Hospital Medicine

Community-acquired pneumonia (CAP) is the second most common cause of hospitalization in the United States, with over 1.5 million unique hospitalizations annually.¹ CAP is also the most common infectious cause of death in US adults.² The 2019 CAP guideline from the American Thoracic Society (ATS) and the Infectious Diseases Society of America (IDSA) provides recommendations on the diagnosis and management of CAP. The guideline provides 16 recommendations, which we have consolidated to highlight practice changing updates in diagnostic testing, risk stratification, and treatment.

Diagnostic Testing

Recommendation 1. In patients with CAP, routine blood cultures, sputum cultures, and urinary antigen tests are not routinely recommended unless severe CAP (Table), history of methicillin-resistant Staphylococcus aureus (MRSA) and/or Pseudomonas infection, or prior hospitalization for which intravenous antibiotics were administered. (Strong recommendation; very low quality of evidence)

The guideline emphasizes that the diagnostic yield of blood/sputum cultures and urinary antigen testing is low. Additionally, high-quality data showing improved clinical outcomes with routine testing of blood cultures and urinary antigens are lacking. Instead, the guideline suggests obtaining blood cultures, urinary antigens, and sputum gram stain and culture only for patients with severe CAP and those being treated for or having prior infection with MRSA or P aeruginosa. They recommend narrowing therapy as appropriate if cultures are negative for either of these two organisms or previous hospitalization with intravenous antibiotics.

Recommendation 2. In patients with CAP, Pneumonia Severity Index (PSI) or CURB-65 (tool based on confusion, urea level, respiratory rate, blood pressure, and age 65 years or older) scores should not be used to determine general medical ward vs intensive care unit care (ICU). (Strong recommendation; low quality of evidence)

The PSI and CURB-65 scores are not validated to determine location of hospital care. Multiple prognostic models have been studied to predict the need for ICU-level care including SMART-COP (systolic blood pressure, multilobar chest radiography involvement, albumin level, respiratory rate, tachycardia, confusion, oxygenation, and arterial pH), SCAPA (Study of Community Acquired Pneumonia Aetiology), and the ATS/IDSA criteria (Table). The positive and negative likelihood ratios for needing ICU admission in CAP with either one major or three or more minor ATS/IDSA criteria are 3.28 and 0.21, respectively.

Treatment

Recommendation 3a. In patients with nonsevere CAP and no risk factors for MRSA or Pseudomonas infection, empiric treatment with a ß-lactam plus macrolide or monotherapy with fluoroquinolones is recommended. (Strong recommendation; high quality of evidence)

Recommendation 3b. In patients with severe CAP and no risk factors for MRSA or Pseudomonas infection, empiric treatment with a ß-lactam plus either a macrolide or fluoroquinolone is recommended. (Strong recommendation; low to moderate quality of evidence)

Microbiologic risk assessment is critical. Risk factors for MRSA or P aeruginosa pneumonia include isolation of these agents in culture and recent hospitalization with receipt of parenteral antibiotics. ß-Lactam monotherapy is not recommended because previous randomized clinical trials (RCTs) demonstrated inferiority of ß-lactam monotherapy to combination therapy for resolution of CAP. The recommended combination therapy for patients with severe CAP without risk factors for MRSA or P aeruginosa infection is a ß-lactam plus either a macrolide or a respiratory fluoroquinolone.

Recommendation 4. In patients with suspected aspiration pneumonia, additional anaerobic coverage is not routinely recommended. (Conditional recommendation; very low quality of evidence)

Aspiration often causes a self-limited pneumonitis that will resolve in 24 to 48 hours with supportive care. Use of additional anaerobic coverage in these patients increases risk for complications (eg, Clostridioides difficile infection) without improving outcomes.

Recommendation 5. In patients with nonsevere CAP, corticosteroids are not routinely recommended. (Conditional recommendation; moderate quality of evidence)

There is no direct evidence that steroids reduce mortality or organ failure in nonsevere CAP. Additionally, the use of steroids in CAP can come with considerable risks (eg, secondary infection, hyperglycemia).

Recommendation 6. In hospitalized patients with CAP, empiric coverage for MRSA or P aeruginosa should be limited to patients meeting specific criteria. (Strong recommendation; moderate quality of evidence)

The guideline highlights the current overuse of extended spectrum antibiotics in patients meeting the previous definition of healthcare-associated pneumonia (HCAP). HCAP was defined by the presence of new chest x-ray infiltrates in patients with various exposures to healthcare settings (eg, chronic dialysis, infusion centers, emergency rooms). Antimicrobial therapy covering MRSA or P aeruginosa should be reserved for patients at risk for MRSA or P aeruginosa infection unless microbiologic testing is negative. Empiric antibiotic selection should incorporate local resistance patterns guided by hospital antibiograms.

Recommendation 7. In adults with CAP, antibiotics should be continued for no less than 5 days with documented clinical stability. (Strong recommendation; moderate quality of evidence)

Hospitalists often determine the length of antibiotic therapy for CAP. Recent studies show extended antibiotic treatment for pneumonia increases risk for adverse events without improving outcomes. Studies also demonstrate patients who receive 5 days of antibiotics total after achieving clinical stability by day 3 do no worse than patients receiving 8 or more days of antibiotics.

This guideline was created by a panel of pulmonologists, infectious disease specialists, general internists, and methodologists using the GRADE (Grading of Recommendations Assessment, Development and Evaluations) approach to draft recommendations. Conflicts of interest were disclosed by all panel members according to the ATS and IDSA policies, and ultimately, two panel members recused themselves owing to conflicts of interest. The inclusion of a large number of RCTs, observational studies, and meta-analyses provides for good generalizability of the guideline published by this group.

Equal support was given in the guideline to all ß-lactams listed, including ampicillin/sulbactam, cefotaxime, ceftriaxone, and ceftaroline, regardless of MRSA risk factors. As the authors explicitly state in the guideline, one of the major reasons for abandoning the HCAP classification was to correct the overuse of anti-MRSA and antipseudomonal therapy.³ It is surprising, then, that the authors would include ceftaroline, a broad-spectrum cephalosporin that covers MRSA, as first-line therapy for patients without risk factors for MRSA.

The guideline also supported the use of a respiratory fluoroquinolone or a ß-lactam with macrolide equally. Although most RCTs have found equal efficacy between these two regimens,⁴ there is growing concern about the safety of fluoroquinolones.⁵ While the authors do encourage clinicians to consider these side effects in the main body of the text, a stronger statement could have been made more prominently to warn clinicians of safety concerns with fluoroquinolones.

Finally, while monotherapy with a ß-lactam was supported for the treatment of nonsevere outpatient CAP, it was not included in the recommendations for the treatment of hospitalized patients. There is conflicting data on this topic. One RCT failed to show noninferiority of monotherapy, but this was most pronounced among patients with severe pneumonia (PSI category IV) or cases with proven atypical infections.⁶ Another RCT found monotherapy to be noninferior to combination therapy for hospitalized patients not admitted to the ICU.⁷ There is also evidence suggesting that many patients hospitalized with pneumonia have viral rather than bacterial infections,⁸ which brings into question the need for antibiotics in this subset entirely. When these findings are considered from a stewardship perspective and patient safety profile, monotherapy with a ß-lactam for hospitalized patients without severe pneumonia could have been considered.

Future research should track the effects of this guideline’s recommendation to narrow empiric therapy on patients empirically treated for MRSA or P aeruginosa infection once sputum and blood cultures are negative, particularly with respect to reduction of time on broad spectrum antimicrobials and clinical outcomes. Similarly, better definitions of which patients require empiric MRSA and P aeruginosa antimicrobial coverage are needed. Ideally, further research will facilitate rapid, cost-­effective, and individualized therapy, particularly with growing concerns for antimicrobial resistance and safety.

The authors have no relevant financial conflicts of interest to disclose.

Community-acquired pneumonia (CAP) is the second most common cause of hospitalization in the United States, with over 1.5 million unique hospitalizations annually.¹ CAP is also the most common infectious cause of death in US adults.² The 2019 CAP guideline from the American Thoracic Society (ATS) and the Infectious Diseases Society of America (IDSA) provides recommendations on the diagnosis and management of CAP. The guideline provides 16 recommendations, which we have consolidated to highlight practice changing updates in diagnostic testing, risk stratification, and treatment.

Diagnostic Testing

Recommendation 1. In patients with CAP, routine blood cultures, sputum cultures, and urinary antigen tests are not routinely recommended unless severe CAP (Table), history of methicillin-resistant Staphylococcus aureus (MRSA) and/or Pseudomonas infection, or prior hospitalization for which intravenous antibiotics were administered. (Strong recommendation; very low quality of evidence)

The guideline emphasizes that the diagnostic yield of blood/sputum cultures and urinary antigen testing is low. Additionally, high-quality data showing improved clinical outcomes with routine testing of blood cultures and urinary antigens are lacking. Instead, the guideline suggests obtaining blood cultures, urinary antigens, and sputum gram stain and culture only for patients with severe CAP and those being treated for or having prior infection with MRSA or P aeruginosa. They recommend narrowing therapy as appropriate if cultures are negative for either of these two organisms or previous hospitalization with intravenous antibiotics.

Recommendation 2. In patients with CAP, Pneumonia Severity Index (PSI) or CURB-65 (tool based on confusion, urea level, respiratory rate, blood pressure, and age 65 years or older) scores should not be used to determine general medical ward vs intensive care unit care (ICU). (Strong recommendation; low quality of evidence)

The PSI and CURB-65 scores are not validated to determine location of hospital care. Multiple prognostic models have been studied to predict the need for ICU-level care including SMART-COP (systolic blood pressure, multilobar chest radiography involvement, albumin level, respiratory rate, tachycardia, confusion, oxygenation, and arterial pH), SCAPA (Study of Community Acquired Pneumonia Aetiology), and the ATS/IDSA criteria (Table). The positive and negative likelihood ratios for needing ICU admission in CAP with either one major or three or more minor ATS/IDSA criteria are 3.28 and 0.21, respectively.

Treatment

Recommendation 3a. In patients with nonsevere CAP and no risk factors for MRSA or Pseudomonas infection, empiric treatment with a ß-lactam plus macrolide or monotherapy with fluoroquinolones is recommended. (Strong recommendation; high quality of evidence)

Recommendation 3b. In patients with severe CAP and no risk factors for MRSA or Pseudomonas infection, empiric treatment with a ß-lactam plus either a macrolide or fluoroquinolone is recommended. (Strong recommendation; low to moderate quality of evidence)

Microbiologic risk assessment is critical. Risk factors for MRSA or P aeruginosa pneumonia include isolation of these agents in culture and recent hospitalization with receipt of parenteral antibiotics. ß-Lactam monotherapy is not recommended because previous randomized clinical trials (RCTs) demonstrated inferiority of ß-lactam monotherapy to combination therapy for resolution of CAP. The recommended combination therapy for patients with severe CAP without risk factors for MRSA or P aeruginosa infection is a ß-lactam plus either a macrolide or a respiratory fluoroquinolone.

Recommendation 4. In patients with suspected aspiration pneumonia, additional anaerobic coverage is not routinely recommended. (Conditional recommendation; very low quality of evidence)

Aspiration often causes a self-limited pneumonitis that will resolve in 24 to 48 hours with supportive care. Use of additional anaerobic coverage in these patients increases risk for complications (eg, Clostridioides difficile infection) without improving outcomes.

Recommendation 5. In patients with nonsevere CAP, corticosteroids are not routinely recommended. (Conditional recommendation; moderate quality of evidence)

There is no direct evidence that steroids reduce mortality or organ failure in nonsevere CAP. Additionally, the use of steroids in CAP can come with considerable risks (eg, secondary infection, hyperglycemia).

Recommendation 6. In hospitalized patients with CAP, empiric coverage for MRSA or P aeruginosa should be limited to patients meeting specific criteria. (Strong recommendation; moderate quality of evidence)

The guideline highlights the current overuse of extended spectrum antibiotics in patients meeting the previous definition of healthcare-associated pneumonia (HCAP). HCAP was defined by the presence of new chest x-ray infiltrates in patients with various exposures to healthcare settings (eg, chronic dialysis, infusion centers, emergency rooms). Antimicrobial therapy covering MRSA or P aeruginosa should be reserved for patients at risk for MRSA or P aeruginosa infection unless microbiologic testing is negative. Empiric antibiotic selection should incorporate local resistance patterns guided by hospital antibiograms.

Recommendation 7. In adults with CAP, antibiotics should be continued for no less than 5 days with documented clinical stability. (Strong recommendation; moderate quality of evidence)

Hospitalists often determine the length of antibiotic therapy for CAP. Recent studies show extended antibiotic treatment for pneumonia increases risk for adverse events without improving outcomes. Studies also demonstrate patients who receive 5 days of antibiotics total after achieving clinical stability by day 3 do no worse than patients receiving 8 or more days of antibiotics.

This guideline was created by a panel of pulmonologists, infectious disease specialists, general internists, and methodologists using the GRADE (Grading of Recommendations Assessment, Development and Evaluations) approach to draft recommendations. Conflicts of interest were disclosed by all panel members according to the ATS and IDSA policies, and ultimately, two panel members recused themselves owing to conflicts of interest. The inclusion of a large number of RCTs, observational studies, and meta-analyses provides for good generalizability of the guideline published by this group.

Equal support was given in the guideline to all ß-lactams listed, including ampicillin/sulbactam, cefotaxime, ceftriaxone, and ceftaroline, regardless of MRSA risk factors. As the authors explicitly state in the guideline, one of the major reasons for abandoning the HCAP classification was to correct the overuse of anti-MRSA and antipseudomonal therapy.³ It is surprising, then, that the authors would include ceftaroline, a broad-spectrum cephalosporin that covers MRSA, as first-line therapy for patients without risk factors for MRSA.

The guideline also supported the use of a respiratory fluoroquinolone or a ß-lactam with macrolide equally. Although most RCTs have found equal efficacy between these two regimens,⁴ there is growing concern about the safety of fluoroquinolones.⁵ While the authors do encourage clinicians to consider these side effects in the main body of the text, a stronger statement could have been made more prominently to warn clinicians of safety concerns with fluoroquinolones.

Finally, while monotherapy with a ß-lactam was supported for the treatment of nonsevere outpatient CAP, it was not included in the recommendations for the treatment of hospitalized patients. There is conflicting data on this topic. One RCT failed to show noninferiority of monotherapy, but this was most pronounced among patients with severe pneumonia (PSI category IV) or cases with proven atypical infections.⁶ Another RCT found monotherapy to be noninferior to combination therapy for hospitalized patients not admitted to the ICU.⁷ There is also evidence suggesting that many patients hospitalized with pneumonia have viral rather than bacterial infections,⁸ which brings into question the need for antibiotics in this subset entirely. When these findings are considered from a stewardship perspective and patient safety profile, monotherapy with a ß-lactam for hospitalized patients without severe pneumonia could have been considered.

Future research should track the effects of this guideline’s recommendation to narrow empiric therapy on patients empirically treated for MRSA or P aeruginosa infection once sputum and blood cultures are negative, particularly with respect to reduction of time on broad spectrum antimicrobials and clinical outcomes. Similarly, better definitions of which patients require empiric MRSA and P aeruginosa antimicrobial coverage are needed. Ideally, further research will facilitate rapid, cost-­effective, and individualized therapy, particularly with growing concerns for antimicrobial resistance and safety.

The authors have no relevant financial conflicts of interest to disclose.

Community-acquired pneumonia (CAP) is the second most common cause of hospitalization in the United States, with over 1.5 million unique hospitalizations annually.¹ CAP is also the most common infectious cause of death in US adults.² The 2019 CAP guideline from the American Thoracic Society (ATS) and the Infectious Diseases Society of America (IDSA) provides recommendations on the diagnosis and management of CAP. The guideline provides 16 recommendations, which we have consolidated to highlight practice changing updates in diagnostic testing, risk stratification, and treatment.

Diagnostic Testing

Recommendation 1. In patients with CAP, routine blood cultures, sputum cultures, and urinary antigen tests are not routinely recommended unless severe CAP (Table), history of methicillin-resistant Staphylococcus aureus (MRSA) and/or Pseudomonas infection, or prior hospitalization for which intravenous antibiotics were administered. (Strong recommendation; very low quality of evidence)

The guideline emphasizes that the diagnostic yield of blood/sputum cultures and urinary antigen testing is low. Additionally, high-quality data showing improved clinical outcomes with routine testing of blood cultures and urinary antigens are lacking. Instead, the guideline suggests obtaining blood cultures, urinary antigens, and sputum gram stain and culture only for patients with severe CAP and those being treated for or having prior infection with MRSA or P aeruginosa. They recommend narrowing therapy as appropriate if cultures are negative for either of these two organisms or previous hospitalization with intravenous antibiotics.

Recommendation 2. In patients with CAP, Pneumonia Severity Index (PSI) or CURB-65 (tool based on confusion, urea level, respiratory rate, blood pressure, and age 65 years or older) scores should not be used to determine general medical ward vs intensive care unit care (ICU). (Strong recommendation; low quality of evidence)

The PSI and CURB-65 scores are not validated to determine location of hospital care. Multiple prognostic models have been studied to predict the need for ICU-level care including SMART-COP (systolic blood pressure, multilobar chest radiography involvement, albumin level, respiratory rate, tachycardia, confusion, oxygenation, and arterial pH), SCAPA (Study of Community Acquired Pneumonia Aetiology), and the ATS/IDSA criteria (Table). The positive and negative likelihood ratios for needing ICU admission in CAP with either one major or three or more minor ATS/IDSA criteria are 3.28 and 0.21, respectively.

Treatment

Recommendation 3a. In patients with nonsevere CAP and no risk factors for MRSA or Pseudomonas infection, empiric treatment with a ß-lactam plus macrolide or monotherapy with fluoroquinolones is recommended. (Strong recommendation; high quality of evidence)

Recommendation 3b. In patients with severe CAP and no risk factors for MRSA or Pseudomonas infection, empiric treatment with a ß-lactam plus either a macrolide or fluoroquinolone is recommended. (Strong recommendation; low to moderate quality of evidence)

Microbiologic risk assessment is critical. Risk factors for MRSA or P aeruginosa pneumonia include isolation of these agents in culture and recent hospitalization with receipt of parenteral antibiotics. ß-Lactam monotherapy is not recommended because previous randomized clinical trials (RCTs) demonstrated inferiority of ß-lactam monotherapy to combination therapy for resolution of CAP. The recommended combination therapy for patients with severe CAP without risk factors for MRSA or P aeruginosa infection is a ß-lactam plus either a macrolide or a respiratory fluoroquinolone.

Recommendation 4. In patients with suspected aspiration pneumonia, additional anaerobic coverage is not routinely recommended. (Conditional recommendation; very low quality of evidence)

Aspiration often causes a self-limited pneumonitis that will resolve in 24 to 48 hours with supportive care. Use of additional anaerobic coverage in these patients increases risk for complications (eg, Clostridioides difficile infection) without improving outcomes.

Recommendation 5. In patients with nonsevere CAP, corticosteroids are not routinely recommended. (Conditional recommendation; moderate quality of evidence)

There is no direct evidence that steroids reduce mortality or organ failure in nonsevere CAP. Additionally, the use of steroids in CAP can come with considerable risks (eg, secondary infection, hyperglycemia).

Recommendation 6. In hospitalized patients with CAP, empiric coverage for MRSA or P aeruginosa should be limited to patients meeting specific criteria. (Strong recommendation; moderate quality of evidence)

The guideline highlights the current overuse of extended spectrum antibiotics in patients meeting the previous definition of healthcare-associated pneumonia (HCAP). HCAP was defined by the presence of new chest x-ray infiltrates in patients with various exposures to healthcare settings (eg, chronic dialysis, infusion centers, emergency rooms). Antimicrobial therapy covering MRSA or P aeruginosa should be reserved for patients at risk for MRSA or P aeruginosa infection unless microbiologic testing is negative. Empiric antibiotic selection should incorporate local resistance patterns guided by hospital antibiograms.

Recommendation 7. In adults with CAP, antibiotics should be continued for no less than 5 days with documented clinical stability. (Strong recommendation; moderate quality of evidence)

Hospitalists often determine the length of antibiotic therapy for CAP. Recent studies show extended antibiotic treatment for pneumonia increases risk for adverse events without improving outcomes. Studies also demonstrate patients who receive 5 days of antibiotics total after achieving clinical stability by day 3 do no worse than patients receiving 8 or more days of antibiotics.

This guideline was created by a panel of pulmonologists, infectious disease specialists, general internists, and methodologists using the GRADE (Grading of Recommendations Assessment, Development and Evaluations) approach to draft recommendations. Conflicts of interest were disclosed by all panel members according to the ATS and IDSA policies, and ultimately, two panel members recused themselves owing to conflicts of interest. The inclusion of a large number of RCTs, observational studies, and meta-analyses provides for good generalizability of the guideline published by this group.

Equal support was given in the guideline to all ß-lactams listed, including ampicillin/sulbactam, cefotaxime, ceftriaxone, and ceftaroline, regardless of MRSA risk factors. As the authors explicitly state in the guideline, one of the major reasons for abandoning the HCAP classification was to correct the overuse of anti-MRSA and antipseudomonal therapy.³ It is surprising, then, that the authors would include ceftaroline, a broad-spectrum cephalosporin that covers MRSA, as first-line therapy for patients without risk factors for MRSA.

The guideline also supported the use of a respiratory fluoroquinolone or a ß-lactam with macrolide equally. Although most RCTs have found equal efficacy between these two regimens,⁴ there is growing concern about the safety of fluoroquinolones.⁵ While the authors do encourage clinicians to consider these side effects in the main body of the text, a stronger statement could have been made more prominently to warn clinicians of safety concerns with fluoroquinolones.

Finally, while monotherapy with a ß-lactam was supported for the treatment of nonsevere outpatient CAP, it was not included in the recommendations for the treatment of hospitalized patients. There is conflicting data on this topic. One RCT failed to show noninferiority of monotherapy, but this was most pronounced among patients with severe pneumonia (PSI category IV) or cases with proven atypical infections.⁶ Another RCT found monotherapy to be noninferior to combination therapy for hospitalized patients not admitted to the ICU.⁷ There is also evidence suggesting that many patients hospitalized with pneumonia have viral rather than bacterial infections,⁸ which brings into question the need for antibiotics in this subset entirely. When these findings are considered from a stewardship perspective and patient safety profile, monotherapy with a ß-lactam for hospitalized patients without severe pneumonia could have been considered.

Future research should track the effects of this guideline’s recommendation to narrow empiric therapy on patients empirically treated for MRSA or P aeruginosa infection once sputum and blood cultures are negative, particularly with respect to reduction of time on broad spectrum antimicrobials and clinical outcomes. Similarly, better definitions of which patients require empiric MRSA and P aeruginosa antimicrobial coverage are needed. Ideally, further research will facilitate rapid, cost-­effective, and individualized therapy, particularly with growing concerns for antimicrobial resistance and safety.

The authors have no relevant financial conflicts of interest to disclose.

According to the most recent Annual Report of the American Association of Poison Control Centers’ National Poison Data System, there were 51,081 patient exposures to acetaminophen or acetaminophen-­containing products treated in a healthcare facility in 2018.¹ The toxicity of acetaminophen is mediated by its metabolism into the electrophile N-acetyl-p-benzoquinone imine (NAPQI).² If the amount of acetaminophen ingested is higher than recommended dosages, glutathione stores may become depleted as the liver tries to detoxify NAPQI and may no longer able to keep up with the demand. As the patient progresses 24 to 36 hours after ingestion, the onset of hepatic injury becomes apparent with elevations in aspartate aminotransferase (AST). AST elevations, should they occur, are almost always present by 36 hours.² Maximum hepatic dysfunction, as well as elevations in prothrombin time, bilirubin, and serum creatinine, occurs 72 to 96 hours after a toxic ingestion. As a result of marked age-associated differences in the process of conjugating acetaminophen metabolites, infants and young children may be less susceptible to acetaminophen-associated hepatotoxicity, compared with their adult counterparts.^3,4

This clinical progress note addresses acetaminophen toxicity treatment, focusing on N-acetylcysteine (NAC) dosing protocols and the 2017 American College of Medical Toxicology (ACMT) position statement, Duration of Intravenous Acetylcysteine Therapy Following Acetaminophen Overdose.⁵ We conducted a literature search via the PubMed database. The authors began by using the following Medical Subject Headings terms “acetaminophen overdose [title],” which yielded 299 articles, “acetaminophen hepatotoxicity [title],” which yielded 283 articles, and “acetaminophen N-acetylcysteine [title],” which yielded 335 titles. Variations of these terms were used to ensure exhaustive search results. The search results were reviewed for applicability to acetaminophen poisoning and especially goal-directed N-acetylcysteine treatment strategies.

Clarifying a patient’s history of ingestion is a crucial first step to the diagnosis and risk evaluation of acetaminophen-poisoned patients. The initial laboratory workup should include, at minimum, measurement (at 4 or more hours after ingestion) of serum acetaminophen level, AST and alanine aminotransferase (ALT), bilirubin, alkaline phosphatase, creatinine, prothrombin time and international normalized ratio (PT/INR), lactate, and ethanol levels. Gastric decontamination is not appropriate in most cases because of the rapid absorption of acetaminophen.² Activated charcoal is most effective when given within the first 2 hours following ingestion. For acetaminophen-­poisoned patients with severe hepatic toxicity, treating clinicians should consider consultation with a medical toxicologist, an individual with expertise, or the local Poison Control Center.⁵

The Rumack-Matthew nomogram provides guidance on the initiation of antidotal treatment for patients with acetaminophen overdose. This nomogram was developed to exhibit high sensitivity in patients at risk for hepatic damage from acetaminophen overdose.⁶ The mainstay for treatment of acetaminophen poisonings is NAC. Early administration of NAC can replenish depleted hepatic glutathione stores, which restores the liver’s ability to detoxify the toxic metabolite NAPQI. When administered within 6 to 8 hours following ingestion, NAC is nearly universally effective at preventing the subsequent hepatic damage from NAPQI.⁷ Consideration of individualized factors should lead the treating clinician to use the Rumack-Matthew nomogram within the appropriate contexts. Factors such as an unreliable timeline of ingestion by patients, concomitantly ingested medications, and patient-specific factors that might alter the absorption of acetaminophen or impact serum acetaminophen concentrations should be considered. While this nomogram is helpful for decision-making pertaining to the initiation of NAC, further management and continuation of treatment must be patient specific and determined by the treating clinician.

In the United States, intravenous and oral formulations of NAC are available for treatment of acetaminophen toxicity. These formulations are equally efficacious in treating acetaminophen toxicity, with the exception of established hepatic failure, for which only the intravenous route has been studied.⁸ Oral administration results in vomiting in approximately 20% of patients, while intravenous administration may uncommonly result in severe anaphylactoid reactions. Three scenarios exist in which the intravenous route is preferred over the oral route: acetaminophen toxicity in pregnant women, acetaminophen-­induced hepatic failure, and intractable vomiting preventing treatment with oral therapy.² The intravenous formulation is Food and Drug Administration–approved for a 21-hour treatment duration, while the oral formulation is traditionally administered for a 72-hour duration.^9,10

Treatment of acetaminophen poisoning must be tailored on an individual basis and should not follow predetermined durations of treatment. Not only might the FDA-approved 21-hour course of intravenous NAC be inadequate, but the standard 72-hour oral regimen duration may be excessive for many patients. Both the 21-hour and 72-hour protocols were developed based on the half-life and expected elimination of acetaminophen from the body. The traditional 72-hour oral protocol used in the United States was approved in 1985 and was based on the observation that, in patients who developed acetaminophen-induced fatal hepatotoxicity, the acetaminophen half-life was 12 hours, requiring five half-lives to eliminate the drug, for a total of 60 hours.¹¹ The FDA then added an additional 12 hours to provide an additional margin of safety. Researchers in the United Kingdom developed an intravenous treatment protocol based on a shorter 4-hour acetaminophen half-life, calculated from observations of a larger pool of acetaminophen-poisoned patients and not just those with fatal hepatotoxicity.¹² This 21-hour protocol—slightly longer than the five 4-hour half-lives needed—was subsequently approved in the United States as the intravenous formulation’s duration in 2004.

Case reports of hepatic failure have described treatment failures of intravenous NAC after the completion of the standard 21 hour treatment regimen.¹³ In a case report by Smith et al, a patient with significant single ingestion is described who ultimately developed hepatotoxicity despite receiving the FDA-­approved 21-hour infusion.¹³ This patient ingested approximately 96 immediate-release, 500-mg acetaminophen tablets over approximately 1 hour. The initial acetaminophen concentration was 264 mcg/mL at 2.25 hours after ingestion. At 21 hours after ingestion, the patient’s acetaminophen levels remained elevated at 116 mcg/mL, but therapy with NAC was discontinued per protocol because AST and ALT levels were normal. A second peak in serum acetaminophen levels occurred after discontinuation of NAC, with levels reaching 228 mcg/mL approximately 48 hours after ingestion. Liver function worsened over the next few days with elevations in AST of greater than 4,000 U/L, PT of 51.4 seconds, and ammonia of 165 mcg/dL. Intravenous NAC was restarted and hepatic function improved gradually. The patient was ultimately discharged from the hospital. This case and others highlight the possible pitfalls of applying a protocolized, fixed treatment duration of NAC to all patients.

Based on individual ingestion timing, quantities, and pharmacokinetics, patients should have pertinent laboratory markers monitored at least every 24 hours from presentation. The clinician may then consider the need to continue or discontinue NAC treatment every 24 hours from the initiation of NAC. Although notable limitations exist for both intravenous and oral treatment regimens, the literature uniformly recommends treating until serum acetaminophen levels are undetectable, regardless of NAC administration route. Based on the metabolism of acetaminophen, as long as serum concentrations of acetaminophen persist, further metabolism to the toxic NAPQI metabolite is possible. The ACMT addressed these concerns with the 2017 position paper, which strongly recommends all of the following criteria to be present prior to discontinuation of intravenous NAC: undetectable acetaminophen concentration, improving hepatic aminotransferases, and improving prognostic markers, including creatinine, lactate, pH, PT/INR, and phosphate.⁵ In cases in which laboratory and clinical parameters are abnormal, continued treatment beyond the established standard NAC treatment durations is warranted. However, once NAC treatment is extended beyond established protocol lengths, the subsequent dosing regimens that should be employed in these patients are not well studied. The ACMT Position Statement states the administration of an additional NAC bolus or extending the duration of the 6.25 mg/kg per hour maintenance infusion may be appropriate in this subset of patients requiring continued treatment.

Based on the positive outcomes in most patients receiving the 21-hour intravenous NAC protocol, the necessity of an additional 2 days of NAC treatment with the 72-hour oral regimen is uncertain. The ACMT Position Statement states “evidence supports using shorter oral NAC courses, provided that liver enzymes and synthetic function are normal or improving, and plasma acetaminophen concentration is undetectable.”⁵ Based on this and available literature, patients with a lower risk for developing hepatotoxicity may have oral NAC therapy discontinued after 24 to 48 hours of treatment if specific lab and patient parameters are within defined criteria.

The ACMT does not differentiate between adult and pediatric recommendations. However, the data used to develop the ACMT goal-directed treatment recommendations included both adult and pediatric patients.^14,15 The same goal-­directed treatment principles may be applied to pediatric patients, though studies have not specifically addressed pediatric goal-directed treatment of acetaminophen poisoning.

Acetaminophen-poisoned patients require evaluation not just prior to initiation of treatment, but also when deciding to discontinue NAC therapy. New literature supports reconsideration of standardized 21- and 72-hour NAC treatment protocols, as well as the notion that treatment duration should be guided by laboratory parameters rather than by route of NAC administration. Goal-directed or patient-tailored treatment is also supported by the ACMT.⁵ Treatment with NAC should continue in patients displaying hepatic toxicity or persistent elevations in acetaminophen levels. In these cases, the optimal treatment regimen in these cases remains unknown and further research into this area is merited.

According to the most recent Annual Report of the American Association of Poison Control Centers’ National Poison Data System, there were 51,081 patient exposures to acetaminophen or acetaminophen-­containing products treated in a healthcare facility in 2018.¹ The toxicity of acetaminophen is mediated by its metabolism into the electrophile N-acetyl-p-benzoquinone imine (NAPQI).² If the amount of acetaminophen ingested is higher than recommended dosages, glutathione stores may become depleted as the liver tries to detoxify NAPQI and may no longer able to keep up with the demand. As the patient progresses 24 to 36 hours after ingestion, the onset of hepatic injury becomes apparent with elevations in aspartate aminotransferase (AST). AST elevations, should they occur, are almost always present by 36 hours.² Maximum hepatic dysfunction, as well as elevations in prothrombin time, bilirubin, and serum creatinine, occurs 72 to 96 hours after a toxic ingestion. As a result of marked age-associated differences in the process of conjugating acetaminophen metabolites, infants and young children may be less susceptible to acetaminophen-associated hepatotoxicity, compared with their adult counterparts.^3,4

This clinical progress note addresses acetaminophen toxicity treatment, focusing on N-acetylcysteine (NAC) dosing protocols and the 2017 American College of Medical Toxicology (ACMT) position statement, Duration of Intravenous Acetylcysteine Therapy Following Acetaminophen Overdose.⁵ We conducted a literature search via the PubMed database. The authors began by using the following Medical Subject Headings terms “acetaminophen overdose [title],” which yielded 299 articles, “acetaminophen hepatotoxicity [title],” which yielded 283 articles, and “acetaminophen N-acetylcysteine [title],” which yielded 335 titles. Variations of these terms were used to ensure exhaustive search results. The search results were reviewed for applicability to acetaminophen poisoning and especially goal-directed N-acetylcysteine treatment strategies.

Clarifying a patient’s history of ingestion is a crucial first step to the diagnosis and risk evaluation of acetaminophen-poisoned patients. The initial laboratory workup should include, at minimum, measurement (at 4 or more hours after ingestion) of serum acetaminophen level, AST and alanine aminotransferase (ALT), bilirubin, alkaline phosphatase, creatinine, prothrombin time and international normalized ratio (PT/INR), lactate, and ethanol levels. Gastric decontamination is not appropriate in most cases because of the rapid absorption of acetaminophen.² Activated charcoal is most effective when given within the first 2 hours following ingestion. For acetaminophen-­poisoned patients with severe hepatic toxicity, treating clinicians should consider consultation with a medical toxicologist, an individual with expertise, or the local Poison Control Center.⁵

The Rumack-Matthew nomogram provides guidance on the initiation of antidotal treatment for patients with acetaminophen overdose. This nomogram was developed to exhibit high sensitivity in patients at risk for hepatic damage from acetaminophen overdose.⁶ The mainstay for treatment of acetaminophen poisonings is NAC. Early administration of NAC can replenish depleted hepatic glutathione stores, which restores the liver’s ability to detoxify the toxic metabolite NAPQI. When administered within 6 to 8 hours following ingestion, NAC is nearly universally effective at preventing the subsequent hepatic damage from NAPQI.⁷ Consideration of individualized factors should lead the treating clinician to use the Rumack-Matthew nomogram within the appropriate contexts. Factors such as an unreliable timeline of ingestion by patients, concomitantly ingested medications, and patient-specific factors that might alter the absorption of acetaminophen or impact serum acetaminophen concentrations should be considered. While this nomogram is helpful for decision-making pertaining to the initiation of NAC, further management and continuation of treatment must be patient specific and determined by the treating clinician.

In the United States, intravenous and oral formulations of NAC are available for treatment of acetaminophen toxicity. These formulations are equally efficacious in treating acetaminophen toxicity, with the exception of established hepatic failure, for which only the intravenous route has been studied.⁸ Oral administration results in vomiting in approximately 20% of patients, while intravenous administration may uncommonly result in severe anaphylactoid reactions. Three scenarios exist in which the intravenous route is preferred over the oral route: acetaminophen toxicity in pregnant women, acetaminophen-­induced hepatic failure, and intractable vomiting preventing treatment with oral therapy.² The intravenous formulation is Food and Drug Administration–approved for a 21-hour treatment duration, while the oral formulation is traditionally administered for a 72-hour duration.^9,10

Treatment of acetaminophen poisoning must be tailored on an individual basis and should not follow predetermined durations of treatment. Not only might the FDA-approved 21-hour course of intravenous NAC be inadequate, but the standard 72-hour oral regimen duration may be excessive for many patients. Both the 21-hour and 72-hour protocols were developed based on the half-life and expected elimination of acetaminophen from the body. The traditional 72-hour oral protocol used in the United States was approved in 1985 and was based on the observation that, in patients who developed acetaminophen-induced fatal hepatotoxicity, the acetaminophen half-life was 12 hours, requiring five half-lives to eliminate the drug, for a total of 60 hours.¹¹ The FDA then added an additional 12 hours to provide an additional margin of safety. Researchers in the United Kingdom developed an intravenous treatment protocol based on a shorter 4-hour acetaminophen half-life, calculated from observations of a larger pool of acetaminophen-poisoned patients and not just those with fatal hepatotoxicity.¹² This 21-hour protocol—slightly longer than the five 4-hour half-lives needed—was subsequently approved in the United States as the intravenous formulation’s duration in 2004.

Case reports of hepatic failure have described treatment failures of intravenous NAC after the completion of the standard 21 hour treatment regimen.¹³ In a case report by Smith et al, a patient with significant single ingestion is described who ultimately developed hepatotoxicity despite receiving the FDA-­approved 21-hour infusion.¹³ This patient ingested approximately 96 immediate-release, 500-mg acetaminophen tablets over approximately 1 hour. The initial acetaminophen concentration was 264 mcg/mL at 2.25 hours after ingestion. At 21 hours after ingestion, the patient’s acetaminophen levels remained elevated at 116 mcg/mL, but therapy with NAC was discontinued per protocol because AST and ALT levels were normal. A second peak in serum acetaminophen levels occurred after discontinuation of NAC, with levels reaching 228 mcg/mL approximately 48 hours after ingestion. Liver function worsened over the next few days with elevations in AST of greater than 4,000 U/L, PT of 51.4 seconds, and ammonia of 165 mcg/dL. Intravenous NAC was restarted and hepatic function improved gradually. The patient was ultimately discharged from the hospital. This case and others highlight the possible pitfalls of applying a protocolized, fixed treatment duration of NAC to all patients.

Based on individual ingestion timing, quantities, and pharmacokinetics, patients should have pertinent laboratory markers monitored at least every 24 hours from presentation. The clinician may then consider the need to continue or discontinue NAC treatment every 24 hours from the initiation of NAC. Although notable limitations exist for both intravenous and oral treatment regimens, the literature uniformly recommends treating until serum acetaminophen levels are undetectable, regardless of NAC administration route. Based on the metabolism of acetaminophen, as long as serum concentrations of acetaminophen persist, further metabolism to the toxic NAPQI metabolite is possible. The ACMT addressed these concerns with the 2017 position paper, which strongly recommends all of the following criteria to be present prior to discontinuation of intravenous NAC: undetectable acetaminophen concentration, improving hepatic aminotransferases, and improving prognostic markers, including creatinine, lactate, pH, PT/INR, and phosphate.⁵ In cases in which laboratory and clinical parameters are abnormal, continued treatment beyond the established standard NAC treatment durations is warranted. However, once NAC treatment is extended beyond established protocol lengths, the subsequent dosing regimens that should be employed in these patients are not well studied. The ACMT Position Statement states the administration of an additional NAC bolus or extending the duration of the 6.25 mg/kg per hour maintenance infusion may be appropriate in this subset of patients requiring continued treatment.

Based on the positive outcomes in most patients receiving the 21-hour intravenous NAC protocol, the necessity of an additional 2 days of NAC treatment with the 72-hour oral regimen is uncertain. The ACMT Position Statement states “evidence supports using shorter oral NAC courses, provided that liver enzymes and synthetic function are normal or improving, and plasma acetaminophen concentration is undetectable.”⁵ Based on this and available literature, patients with a lower risk for developing hepatotoxicity may have oral NAC therapy discontinued after 24 to 48 hours of treatment if specific lab and patient parameters are within defined criteria.

The ACMT does not differentiate between adult and pediatric recommendations. However, the data used to develop the ACMT goal-directed treatment recommendations included both adult and pediatric patients.^14,15 The same goal-­directed treatment principles may be applied to pediatric patients, though studies have not specifically addressed pediatric goal-directed treatment of acetaminophen poisoning.

Acetaminophen-poisoned patients require evaluation not just prior to initiation of treatment, but also when deciding to discontinue NAC therapy. New literature supports reconsideration of standardized 21- and 72-hour NAC treatment protocols, as well as the notion that treatment duration should be guided by laboratory parameters rather than by route of NAC administration. Goal-directed or patient-tailored treatment is also supported by the ACMT.⁵ Treatment with NAC should continue in patients displaying hepatic toxicity or persistent elevations in acetaminophen levels. In these cases, the optimal treatment regimen in these cases remains unknown and further research into this area is merited.

According to the most recent Annual Report of the American Association of Poison Control Centers’ National Poison Data System, there were 51,081 patient exposures to acetaminophen or acetaminophen-­containing products treated in a healthcare facility in 2018.¹ The toxicity of acetaminophen is mediated by its metabolism into the electrophile N-acetyl-p-benzoquinone imine (NAPQI).² If the amount of acetaminophen ingested is higher than recommended dosages, glutathione stores may become depleted as the liver tries to detoxify NAPQI and may no longer able to keep up with the demand. As the patient progresses 24 to 36 hours after ingestion, the onset of hepatic injury becomes apparent with elevations in aspartate aminotransferase (AST). AST elevations, should they occur, are almost always present by 36 hours.² Maximum hepatic dysfunction, as well as elevations in prothrombin time, bilirubin, and serum creatinine, occurs 72 to 96 hours after a toxic ingestion. As a result of marked age-associated differences in the process of conjugating acetaminophen metabolites, infants and young children may be less susceptible to acetaminophen-associated hepatotoxicity, compared with their adult counterparts.^3,4

This clinical progress note addresses acetaminophen toxicity treatment, focusing on N-acetylcysteine (NAC) dosing protocols and the 2017 American College of Medical Toxicology (ACMT) position statement, Duration of Intravenous Acetylcysteine Therapy Following Acetaminophen Overdose.⁵ We conducted a literature search via the PubMed database. The authors began by using the following Medical Subject Headings terms “acetaminophen overdose [title],” which yielded 299 articles, “acetaminophen hepatotoxicity [title],” which yielded 283 articles, and “acetaminophen N-acetylcysteine [title],” which yielded 335 titles. Variations of these terms were used to ensure exhaustive search results. The search results were reviewed for applicability to acetaminophen poisoning and especially goal-directed N-acetylcysteine treatment strategies.

Clarifying a patient’s history of ingestion is a crucial first step to the diagnosis and risk evaluation of acetaminophen-poisoned patients. The initial laboratory workup should include, at minimum, measurement (at 4 or more hours after ingestion) of serum acetaminophen level, AST and alanine aminotransferase (ALT), bilirubin, alkaline phosphatase, creatinine, prothrombin time and international normalized ratio (PT/INR), lactate, and ethanol levels. Gastric decontamination is not appropriate in most cases because of the rapid absorption of acetaminophen.² Activated charcoal is most effective when given within the first 2 hours following ingestion. For acetaminophen-­poisoned patients with severe hepatic toxicity, treating clinicians should consider consultation with a medical toxicologist, an individual with expertise, or the local Poison Control Center.⁵

The Rumack-Matthew nomogram provides guidance on the initiation of antidotal treatment for patients with acetaminophen overdose. This nomogram was developed to exhibit high sensitivity in patients at risk for hepatic damage from acetaminophen overdose.⁶ The mainstay for treatment of acetaminophen poisonings is NAC. Early administration of NAC can replenish depleted hepatic glutathione stores, which restores the liver’s ability to detoxify the toxic metabolite NAPQI. When administered within 6 to 8 hours following ingestion, NAC is nearly universally effective at preventing the subsequent hepatic damage from NAPQI.⁷ Consideration of individualized factors should lead the treating clinician to use the Rumack-Matthew nomogram within the appropriate contexts. Factors such as an unreliable timeline of ingestion by patients, concomitantly ingested medications, and patient-specific factors that might alter the absorption of acetaminophen or impact serum acetaminophen concentrations should be considered. While this nomogram is helpful for decision-making pertaining to the initiation of NAC, further management and continuation of treatment must be patient specific and determined by the treating clinician.

In the United States, intravenous and oral formulations of NAC are available for treatment of acetaminophen toxicity. These formulations are equally efficacious in treating acetaminophen toxicity, with the exception of established hepatic failure, for which only the intravenous route has been studied.⁸ Oral administration results in vomiting in approximately 20% of patients, while intravenous administration may uncommonly result in severe anaphylactoid reactions. Three scenarios exist in which the intravenous route is preferred over the oral route: acetaminophen toxicity in pregnant women, acetaminophen-­induced hepatic failure, and intractable vomiting preventing treatment with oral therapy.² The intravenous formulation is Food and Drug Administration–approved for a 21-hour treatment duration, while the oral formulation is traditionally administered for a 72-hour duration.^9,10

Treatment of acetaminophen poisoning must be tailored on an individual basis and should not follow predetermined durations of treatment. Not only might the FDA-approved 21-hour course of intravenous NAC be inadequate, but the standard 72-hour oral regimen duration may be excessive for many patients. Both the 21-hour and 72-hour protocols were developed based on the half-life and expected elimination of acetaminophen from the body. The traditional 72-hour oral protocol used in the United States was approved in 1985 and was based on the observation that, in patients who developed acetaminophen-induced fatal hepatotoxicity, the acetaminophen half-life was 12 hours, requiring five half-lives to eliminate the drug, for a total of 60 hours.¹¹ The FDA then added an additional 12 hours to provide an additional margin of safety. Researchers in the United Kingdom developed an intravenous treatment protocol based on a shorter 4-hour acetaminophen half-life, calculated from observations of a larger pool of acetaminophen-poisoned patients and not just those with fatal hepatotoxicity.¹² This 21-hour protocol—slightly longer than the five 4-hour half-lives needed—was subsequently approved in the United States as the intravenous formulation’s duration in 2004.

Case reports of hepatic failure have described treatment failures of intravenous NAC after the completion of the standard 21 hour treatment regimen.¹³ In a case report by Smith et al, a patient with significant single ingestion is described who ultimately developed hepatotoxicity despite receiving the FDA-­approved 21-hour infusion.¹³ This patient ingested approximately 96 immediate-release, 500-mg acetaminophen tablets over approximately 1 hour. The initial acetaminophen concentration was 264 mcg/mL at 2.25 hours after ingestion. At 21 hours after ingestion, the patient’s acetaminophen levels remained elevated at 116 mcg/mL, but therapy with NAC was discontinued per protocol because AST and ALT levels were normal. A second peak in serum acetaminophen levels occurred after discontinuation of NAC, with levels reaching 228 mcg/mL approximately 48 hours after ingestion. Liver function worsened over the next few days with elevations in AST of greater than 4,000 U/L, PT of 51.4 seconds, and ammonia of 165 mcg/dL. Intravenous NAC was restarted and hepatic function improved gradually. The patient was ultimately discharged from the hospital. This case and others highlight the possible pitfalls of applying a protocolized, fixed treatment duration of NAC to all patients.

Based on individual ingestion timing, quantities, and pharmacokinetics, patients should have pertinent laboratory markers monitored at least every 24 hours from presentation. The clinician may then consider the need to continue or discontinue NAC treatment every 24 hours from the initiation of NAC. Although notable limitations exist for both intravenous and oral treatment regimens, the literature uniformly recommends treating until serum acetaminophen levels are undetectable, regardless of NAC administration route. Based on the metabolism of acetaminophen, as long as serum concentrations of acetaminophen persist, further metabolism to the toxic NAPQI metabolite is possible. The ACMT addressed these concerns with the 2017 position paper, which strongly recommends all of the following criteria to be present prior to discontinuation of intravenous NAC: undetectable acetaminophen concentration, improving hepatic aminotransferases, and improving prognostic markers, including creatinine, lactate, pH, PT/INR, and phosphate.⁵ In cases in which laboratory and clinical parameters are abnormal, continued treatment beyond the established standard NAC treatment durations is warranted. However, once NAC treatment is extended beyond established protocol lengths, the subsequent dosing regimens that should be employed in these patients are not well studied. The ACMT Position Statement states the administration of an additional NAC bolus or extending the duration of the 6.25 mg/kg per hour maintenance infusion may be appropriate in this subset of patients requiring continued treatment.

Based on the positive outcomes in most patients receiving the 21-hour intravenous NAC protocol, the necessity of an additional 2 days of NAC treatment with the 72-hour oral regimen is uncertain. The ACMT Position Statement states “evidence supports using shorter oral NAC courses, provided that liver enzymes and synthetic function are normal or improving, and plasma acetaminophen concentration is undetectable.”⁵ Based on this and available literature, patients with a lower risk for developing hepatotoxicity may have oral NAC therapy discontinued after 24 to 48 hours of treatment if specific lab and patient parameters are within defined criteria.

The ACMT does not differentiate between adult and pediatric recommendations. However, the data used to develop the ACMT goal-directed treatment recommendations included both adult and pediatric patients.^14,15 The same goal-­directed treatment principles may be applied to pediatric patients, though studies have not specifically addressed pediatric goal-directed treatment of acetaminophen poisoning.

Acetaminophen-poisoned patients require evaluation not just prior to initiation of treatment, but also when deciding to discontinue NAC therapy. New literature supports reconsideration of standardized 21- and 72-hour NAC treatment protocols, as well as the notion that treatment duration should be guided by laboratory parameters rather than by route of NAC administration. Goal-directed or patient-tailored treatment is also supported by the ACMT.⁵ Treatment with NAC should continue in patients displaying hepatic toxicity or persistent elevations in acetaminophen levels. In these cases, the optimal treatment regimen in these cases remains unknown and further research into this area is merited.

Uncomplicated bacteremia, while not precisely defined in the literature, generally implies bacteremia in the absence of a persistent or difficult-to-eradicate infectious source. Bacteremia secondary to focal infections such as skin and soft-tissue infection, pneumonia, pyelonephritis, or urinary tract infection (UTI) accounts for up to 25% of bloodstream infections (BSIs) and usually resolves with prompt and appropriate antimicrobial therapy.^1,2 Current practice guidelines do not adequately address key aspects of the optimal management of gram-negative (GN)–BSI commonly encountered in hospital care.^3-7 Notably, antimicrobial duration, criteria to transition from intravenous (IV) to oral step-down therapy, choice of oral antimicrobials, and reassessment of follow-up blood cultures have not been addressed. In the absence of consensus guidelines, clinicians rely on “conventional wisdom” and clinical experience, which may not be supported by scientific rigor. A growing body of research now challenges some long-standing practices once thought to be standard of care.

In this narrative review, we aim to examine and synthesize emerging information to provide an evidence-based framework in the management of hospitalized patients with GN-BSI. We highlight the unintended consequences and potential harms of excessive antimicrobial exposure and focus on areas in the fundamental approach to duration of therapy, the role of oral antimicrobials, and usefulness of follow-up blood cultures. A comprehensive search of the published literature was performed in PubMed with an emphasis on articles published during 2015-2019 with use of search terms including gram-negative bacteremia, duration, antibiotics, adverse effects, intravascular catheter, and follow-up blood cultures.

Antimicrobial overuse is common and may be driven by concerns for undertreatment. Clinicians may believe that prolonged antimicrobial therapy maximizes cure rates, with treatment duration often defined arbitrarily by a fixed number of “Constantine-units” (dating back to the ancient Roman emperor’s decree of 7 days in a week).^8-10 Recent publications refute this notion and point out that the harms of overprescribing outweigh the perceived benefits of longer treatment duration.

Antimicrobials are lifesaving but not benign; adverse effects are common and costly to our patients and healthcare system. Among 1,488 hospitalized adults who received at least 24 hours of systemic antimicrobials, 20% had an antimicrobial-­associated adverse event, mostly gastrointestinal, renal, or hematologic in nature.¹¹ Prolonged duration of antimicrobials is further associated with adverse effects such as antimicrobial-associated diarrhea, increased rates of Clostridioides difficile infection (CDI), emergence of antimicrobial resistance, and longer hospital length of stay (LOS).^11-15 Vaughn and colleagues conducted the largest observational study to date, evaluating antimicrobial prescriptions for the treatment of nearly 6,500 adults with community-acquired pneumonia in a 43-hospital consortium in Michigan.¹⁴ More than two-thirds of patients received antimicrobial courses (median 8 days) that exceeded guideline-­recommended duration. Patients who received longer antimicrobial courses did not have reduced mortality, readmission, or emergency department visits. More importantly, each excess day of treatment was associated with a relative 5% increase in the odds of antimicrobial-associated adverse effects reported by patients. This is further supported by national and state hospital data that antimicrobial-associated adverse events are an independent predictor of longer LOS.¹²

CDI is commonly linked to destructive changes to the indigenous microbiota of the intestinal flora caused by antimicrobial administration. Stevens and colleagues identified 7,792 hospitalized patients who received at least 2 consecutive days of antimicrobial therapy¹³; comparing 241 cases of CDI with the control group, they observed a dose-dependent risk of CDI associated with increasing cumulative dose, number of antimicrobials, and days of antimicrobial exposure. Compared with patients who received fewer than 4 days of antimicrobials, the adjusted hazard ratios (aHR) for those who received 4-7 days or 8-18 days of therapy were 1.4 (95% CI, 0.8-2.4) and 3.0 (95% CI, 1.9-5.0), respectively. This correlates to a threefold increase in CDI risk for patients who received more than 7 days of antimicrobials. More specifically, the empiric use of antipseudomonal ß-lactams (APBL) for more than 48 hours was also found to be an independent risk factor for CDI among 808 patients with Enterobacteriaceae BSI.¹⁶ The risk of CDI within 90 days of BSI was higher among those who received >48 hours of APBL than it was among those who received ≤48 hours (HR, 3.6; 95% CI, 1.5-9.9).

While C difficile may be the most well-known pathogen implicated in antimicrobial usage, the incidence of multidrug-­resistant (MDR) organisms, either as infectious or colonizing pathogens, is also tied to antimicrobial exposure. Among patients receiving systemic antimicrobials, 6% developed an MDR infection within 90 days.¹¹ Over a 5-year period, Teshome and colleagues evaluated 7,118 critically ill patients and demonstrated that prolonged exposures to APBLs increased the risk of new antimicrobial resistance within 60 days.¹⁵ This resistance pattern was not an institutional or environmental finding but a patient-level finding. For each additional day of cefepime or piperacillin/tazobactam received, the risk of new antimicrobial resistance was increased by 8%. The authors concluded that defining a piperacillin/tazobactam course as 10 vs 7 days would result in a 24% higher relative risk of resistance per patient related to those 3 additional days of antimicrobial exposure.

Catheter complications including thrombophlebitis, infiltration, and infection are serious and frequent problems associated with IV medication administration.¹⁷ Even with short-term use, peripherally inserted central catheters (PICCs) carry a substantial risk of venous thrombosis (superficial and deep veins). The incidence of deep vein thrombosis (DVT) for PICCs is estimated between 5% and 15% for hospitalized patients and 2% and 5% for ambulatory patients.¹⁸ A recent randomized controlled trial (RCT) of oral vs IV antimicrobials for bone and joint infections reported that, compared with patients randomized to oral antimicrobials, those randomized to IV antimicrobials were more likely to have catheter complications (9.4% vs 1.0%; P < .001) and to discontinue therapy earlier (18.9% vs 12.8%; P = .006).¹⁹ Median hospital stay was also significantly longer in the IV group (14 days vs 11 days; P < .001).

Optimization of antimicrobial duration has long been recognized as one of the key strategies in reducing unnecessary antimicrobial exposure, yet high-quality evidence on comparative effectiveness of duration in the setting of bacteremia has been limited until recently.²⁰ The presence of bacteremia is often used as a justification for prolonged courses of antimicrobial regardless of infection source or clinical response. The Infectious Diseases Society of America guidelines suggest 7 to 14 days of treatment for intravascular catheter-associated gram-negative bacteremia, but the optimal duration for non–catheter-related gram-negative bacteremia is not addressed.²¹ This lack of clear guidance and the historical scarcity of robust data make it difficult to inform best practices, which leads to wide variability in clinical practice and 14 days being the most prescribed duration.^22,23

Pooled clinical trials’ data from subsets of patients with bacteremia and those from observational studies have been the best available evidence for the treatment duration of GN-BSI until recently (Table 1).^24-32 Two meta-analyses evaluating RCTs of adult and pediatric patients with pyelonephritis, UTI, peritonitis, and pneumonia found no differences in clinical failure, microbiologic cure, or survival between short and long courses of therapy in the subset of patient with associated bacteremia.^24,25 Six heterogeneous RCTs of short vs long courses of therapy for complicated UTI or pyelonephritis reported no differences in clinical cure rates in the subset of patients with associated GN-BSI.² The observational studies outlined in Table 1 are also consistent with RCT results supporting noninferiority in clinical cure and mortality outcomes between short and long courses of therapy.^26-32 These findings may also be extrapolated to immunocompromised hosts given a considerable representation of 10% to 47% of the study population with immunosuppressive conditions.

Nelson and colleagues conducted the only retrospective study to date reporting conflicting results of higher risk of treatment failure (defined as composite endpoint of mortality or recurrent infection within 90 days of index BSI) in patients receiving a short course of therapy.²⁷ However, the difference was driven by 90-day mortality (8.2% vs 3.3%; P = .04) not recurrent infection (6.7% vs 6.5%; P = 0.93). Giannella and colleagues also evaluated 90-day mortality as a primary endpoint in a much larger cohort of over 850 patients in Italy and found no difference in mortality rates between short and long courses of antimicrobials.³⁰

Yahav and colleagues conducted the first well-designed open-label RCT comparing short and long courses of antimicrobials in uncomplicated GN-BSI.³³ This noninferiority study randomized more than 600 hospitalized patients with adequate source control who were afebrile and hemodynamically stable for ≥48 hours to receive either 7 days or 14 days of therapy. The source of infections was predominantly urinary (68%), and the causative pathogens were 90% Enterobacteriacae, including 20% MDR strains. The primary outcome was a composite of 90-day all-cause mortality or clinical failure defined as either relapse of bacteremia, local or distant complications, readmission, or extended hospital stay >14 days. The authors reported no statistically significant differences in the primary outcome between short (45.8%) and long (48.3%) courses of treatment. In the prespecified post hoc analysis designed to evaluate infection-­related outcomes at an earlier time frame, there were no observed differences in complications, relapses, or mortality between study groups at 14 and 28 days. Further subgroup analysis demonstrated similar results among patients with MDR pathogens, primarily extended-spectrum ß-lactamases (ESBL). Interestingly, there was a more rapid return to baseline activity and functional capacity among patients randomized to a short course of therapy. The authors acknowledged that the patients’ perception of illness while taking antimicrobials may have influenced self-reported well-being and functional performance. In exploratory analysis, prolonged hospitalization and readmission were excluded from the primary study endpoint to mirror outcomes assessed by Nelson and colleagues. There were no statistically significant differences in death, relapses, or complications between groups randomized to short (18.6%) or long (15.1%) courses of therapy, with a risk difference of 3.5% (95% CI, –2.5% to 9.5%) in this study population.

Patients with Pseudomonas aeruginosa BSI often have more chronic medical comorbidities, immunocompromised conditions, higher severity of illness, and more indwelling catheters than do patients with Enterobacteriaceae BSI.³² It is uncertain whether shorter duration of therapy is generalizable to this population, given that Pseudomonas accounted for a relatively low number (8%) of infections in the published RCT.³³ Fabre and colleagues included high-risk patients with >65% of the cohort with severe immunocompromised conditions consisting of stem cell transplantation, recent chemotherapy, or neutropenia, and they reported no difference in 30-day mortality or recurrent infections among patients with pseudomonal BSI regardless of duration of therapy.³²

It is a well-accepted standard of practice that BSI are treated with upfront IV antimicrobials that can rapidly achieve therapeutic serum concentration. Whether IV administration is warranted for the entire duration of therapy, though, remains controversial. Even in an era of highly bioavailable oral antimicrobials, clinicians often assume that IV antimicrobials are more potent and efficacious than oral antimicrobials.^8,9 This belief has contributed to the dogma that IV therapy is necessary irrespective of the associated risks and costs. Oral antimicrobials are often overlooked as alternatives despite established benefits in avoiding complications associated with IV catheters, decreasing hospital LOS, and improving quality of life.³⁴ There are promising clinical data in support of the efficacy and safety of transitioning from sequential-IV to highly bioavailable oral agents for the treatment of uncomplicated bacteremia caused by both gram-positive and gram-negative pathogens.^2,35 Highly bioavailable oral antimicrobials are also increasingly integrated as sequential therapy for deep-seated infections in bone and joint infections, such as vertebral osteomyelitis.^19,36 These findings have been confirmed in a recent RCT demonstrating noninferiority of oral antimicrobial combinations after satisfactory clinical responses to at least 10 days of IV therapy, compared with continued IV regimens, in left-­sided infective endocarditis.³⁶ While not a prespecified endpoint, hospital LOS was shorter among patients randomized to oral antimicrobials.

Although there are no large-scale RCTs sufficiently powered to address the role of oral antimicrobials in the treatment of uncomplicated GN-BSI, some insights can be gleaned from the existing literature (Table 2). In the RCT establishing noninferiority of short vs long courses of antimicrobials for uncomplicated GN-BSI, the majority of patients randomized to 7 days vs 14 days of therapy, 64% and 81%, respectively, were de-escalated to oral antimicrobials, with fluoroquinolones (FQs) being the predominant (>70%) oral regimen, followed by trimethoprim/sulfamethoxazole (T/S) and oral ß-lactams.³³

Despite the Food and Drug Administration warnings of the potentially permanent adverse effects involving tendons, muscles, joints, nerves, and most recently, aortic aneurysms and ruptures,³⁷ FQs remain a unique class of drugs with favorable pharmacodynamic and pharmacokinetic properties that achieve approximately equivalent serum and tissue concentration when administered either intravenously or orally. This advantage was recognized early on as a potential IV-sparing therapeutic option. A prospective RCT that evaluated oral vs IV ciprofloxacin as initial empiric therapy among 141 patients with pyelonephritis or complicated UTI (38% with secondary BSI) reported no significant differences in microbiological failure or clinical response between the two treatment groups.³⁸ Two small RCTs have also demonstrated the safety and effectiveness of sequential-IV antimicrobial to oral FQs in the setting of GN-BSI secondary to urinary source and cholangitis.^39,40 Oral ß-lactams, however, achieve substantially lower serum concentration than do their IV counterparts and, accordingly, may be less reliably effective.²

Five retrospective cohort studies have more directly investigated the role of oral antimicrobials in the setting of GN-BSI secondary to common focal infections (Table 2 and Table 3).^41-45 Two observational studies reported no difference of treatment failure among patients who received IV-only therapy vs those who were switched to oral therapy in bacteremia secondary to UTIs.^41,42 Catheter-associated complications were higher in the IV cohort (6.1% vs 0.4%; P = .03).⁴² In the largest multicenter cohort study to date, which included 1,478 patients with Enterobacteriaceae bacteremia, there was no difference in 30-day mortality or recurrent bacteremia between patients converted to oral step-down therapy and patients who received the full course of IV antimicrobials.⁴³ Furthermore, the median hospital LOS was shorter (5 days vs 7 days; P < .001) among patients who were transitioned to oral therapy, a finding that is consistent with other studies.^39-42 In their analysis, the oral antimicrobials were categorized as low-bioavailability (ß-lactams) or high-bioavailability (FQ and T/S), and there was no difference in outcomes when results were stratified by bioavailability. Mercuro and colleagues reported similar clinical success among patients who received oral ß-lactams and those who received FQs as step-down therapy.⁴⁴ Notably, patients were more likely to tolerate ß-lactams without experiencing adverse effects than were those who received FQs (91.7% vs 82.1%; P = .049). In contrast, Kutob and colleagues compared step-down oral antimicrobials categorized as low bioavailability (ß-lactams), moderate bioavailability (ciprofloxacin and T/S), and high bioavailability (levofloxacin). They reported that treatment failures were significantly higher among patients who received low-bioavailability (14%) and moderate-­bioavailability (12%) antimicrobials, compared with those who received the high-bioavailability agent (2%; P = .02).⁴⁵ Interestingly, the bioavailability of ciprofloxacin reaches 85% and T/S approaches 90%, and they are often categorized as highly bioavailable agents in other studies.^43,46 If they were reclassified as highly bioavailable agents, the study conclusions might differ. Nevertheless, the reported success with oral step-down therapy exceeded 85% in all five studies.^41-45

It is important to acknowledge the possibility of unmeasured confounders in these retrospective, observational studies despite statistical adjustments and that they are likely underpowered to determine the clinical significance of oral bioavailability of antimicrobials. In a meta-analysis of published studies and abstracts that included 2,289 patients with Enterobacteriaceae bacteremia, all-cause mortality was similar between patients de-escalated to an oral FQ, T/S, or ß-lactam.⁴⁶ Overall recurrence of infection (bacteremia or primary site) occurred more frequently in patients transitioned to oral ß-lactams than FQs, but relapse of bacteremia was not statistically different between comparator groups. Bioavailability of the oral agents may not be the sole determinant of higher recurrence; adherence may be poor because of the more frequent dosing required for oral ß-lactams to achieve targeted pharmacokinetics. Additionally, suboptimal dosing of oral ß-lactams noted in the studies may have also contributed to the increased recurrences.

After source control has been achieved and bacterial inoculum burden is sufficiently reduced with appropriate upfront IV therapy, the bioavailability of oral antimicrobials may become less important. However, existing observational data indicate clinical experience is most established with highly bioavailable oral agents, particularly FQs, though the risks vs benefits require careful consideration. For now, the preferred oral agent remains uncertain and selections should be individualized based on susceptibility, patient factors, and other clinical considerations. More importantly, if there are no contraindications or concerns of malabsorption, oral step-down therapy should be initiated as soon as source control and good clinical responses have been achieved.

Routine follow-up blood cultures (FUBCs) are strongly recommended in Staphylococcus aureus bacteremia because of the propensity for endovascular and metastatic infection, which dictates clinical decision-making regarding duration of therapy. In contrast, GN-BSI secondary to focal infections is usually transient, and the need for confirmation of blood culture clearance is less clear. The low yield of FUBC in many clinical settings suggests that it may not be helpful.^47-50 Despite the questionable impact on the clinical management of GN-BSI, FUBCs are routinely ordered in the hospital.^1,47 Although there is no high-quality evidence addressing the utility of FUBC, several observational studies suggest clinicians should reconsider routinely ordering FUBC.

Canzoneri and colleagues retrospectively evaluated 383 episodes of bacteremia with at least one FUBC drawn after the initial blood culture.⁴⁸ On average, 2.32 FUBC were performed per patient for GN-BSI episode, and only 8 patients (5.7%) had persistent bacteremia. Specifically, only 3% had documented positive FUBC among patients with urinary tract source of infection. It was estimated that 17 FUBCs are needed to yield one positive result for GN-BSI. This finding is consistent with results from another study that examined 1,801 episodes of bacteremia, 901 of which were gram-negative organisms, predominantly (67%) Escherichia coli and Klebsiella spp.⁴⁹ Among GN-BSI episodes, FUBCs were performed in 247 cases, with 27 (10.9%) cases demonstrating persistent bacteremia. A nested case-­control analysis between patients with cleared or persistent bacteremia found a lower yield in FUBC with gram-negative organisms and a genitourinary source of infection. Moreover, persistent bacteremia did not influence a change in antimicrobial regimen. Kang and colleagues investigated 1,068 episodes of Klebsiella pneumoniae bacteremia, with FUBCs performed in 862 (80.7%) cases despite only a 7.2% incidence of persistent bacteremia.⁵⁰ The independent risk factors associated with persistent bacteremia were intra-abdominal infection, solid organ transplantation, high Charlson comorbidity index score, and unfavorable treatment responses, which suggests the need for FUBC may be individualized rather than routine.

In the setting of GN-BSI in which the probability of persistent bacteremia is relatively low, especially in genitourinary sources of infection, FUBCs are not warranted. It is uncomfortable for patients and exposes them to harms of false-positive results, leading to antimicrobial administration with possible adverse effects, which can be further compounded by unnecessary testing, potentially missed alternative diagnosis, and increasing hospital LOS.^1,51 Given the low yield of FUBC in GN-BSI, and the lack of association of persistent bacteremia with change in antimicrobial therapy or clinical outcomes, we recommend avoiding FUBC as a test of cure. Documentation of gram-negative blood culture clearance should be reserved for situations in which there is concern for deeper or otherwise uncontrolled source of infection.

The optimal management of gram-negative bacteremia in hospitalized patients is evolving. There is a growing body of evidence supporting shorter duration for a total of 7 days with oral step-down therapy as safe and effective for patients with uncomplicated Enterobacteriaceae bacteremia who have achieved adequate source control and demonstrated clinical stability and improvement. Although comparative data regarding the optimal duration of therapy in the setting of MDR strains such as ESBL Enterobacteriaceae and pseudomonal BSI are limited, available data appear promising in favor of shorter treatment duration with oral step-down therapy. Routine follow-up blood culture is not cost-effective and may result in unnecessary healthcare resource utilization and inappropriate use of antimicrobials. Table 4 provides a framework for the clinical management of GN-BSI in the hospital. Taken together, these steps will facilitate antimicrobial stewardship, limit unnecessary antimicrobial exposure, and improve quality of patient care.

The authors have no conflicts of interest to disclose.

Uncomplicated bacteremia, while not precisely defined in the literature, generally implies bacteremia in the absence of a persistent or difficult-to-eradicate infectious source. Bacteremia secondary to focal infections such as skin and soft-tissue infection, pneumonia, pyelonephritis, or urinary tract infection (UTI) accounts for up to 25% of bloodstream infections (BSIs) and usually resolves with prompt and appropriate antimicrobial therapy.^1,2 Current practice guidelines do not adequately address key aspects of the optimal management of gram-negative (GN)–BSI commonly encountered in hospital care.^3-7 Notably, antimicrobial duration, criteria to transition from intravenous (IV) to oral step-down therapy, choice of oral antimicrobials, and reassessment of follow-up blood cultures have not been addressed. In the absence of consensus guidelines, clinicians rely on “conventional wisdom” and clinical experience, which may not be supported by scientific rigor. A growing body of research now challenges some long-standing practices once thought to be standard of care.

In this narrative review, we aim to examine and synthesize emerging information to provide an evidence-based framework in the management of hospitalized patients with GN-BSI. We highlight the unintended consequences and potential harms of excessive antimicrobial exposure and focus on areas in the fundamental approach to duration of therapy, the role of oral antimicrobials, and usefulness of follow-up blood cultures. A comprehensive search of the published literature was performed in PubMed with an emphasis on articles published during 2015-2019 with use of search terms including gram-negative bacteremia, duration, antibiotics, adverse effects, intravascular catheter, and follow-up blood cultures.

Antimicrobial overuse is common and may be driven by concerns for undertreatment. Clinicians may believe that prolonged antimicrobial therapy maximizes cure rates, with treatment duration often defined arbitrarily by a fixed number of “Constantine-units” (dating back to the ancient Roman emperor’s decree of 7 days in a week).^8-10 Recent publications refute this notion and point out that the harms of overprescribing outweigh the perceived benefits of longer treatment duration.

Antimicrobials are lifesaving but not benign; adverse effects are common and costly to our patients and healthcare system. Among 1,488 hospitalized adults who received at least 24 hours of systemic antimicrobials, 20% had an antimicrobial-­associated adverse event, mostly gastrointestinal, renal, or hematologic in nature.¹¹ Prolonged duration of antimicrobials is further associated with adverse effects such as antimicrobial-associated diarrhea, increased rates of Clostridioides difficile infection (CDI), emergence of antimicrobial resistance, and longer hospital length of stay (LOS).^11-15 Vaughn and colleagues conducted the largest observational study to date, evaluating antimicrobial prescriptions for the treatment of nearly 6,500 adults with community-acquired pneumonia in a 43-hospital consortium in Michigan.¹⁴ More than two-thirds of patients received antimicrobial courses (median 8 days) that exceeded guideline-­recommended duration. Patients who received longer antimicrobial courses did not have reduced mortality, readmission, or emergency department visits. More importantly, each excess day of treatment was associated with a relative 5% increase in the odds of antimicrobial-associated adverse effects reported by patients. This is further supported by national and state hospital data that antimicrobial-associated adverse events are an independent predictor of longer LOS.¹²

CDI is commonly linked to destructive changes to the indigenous microbiota of the intestinal flora caused by antimicrobial administration. Stevens and colleagues identified 7,792 hospitalized patients who received at least 2 consecutive days of antimicrobial therapy¹³; comparing 241 cases of CDI with the control group, they observed a dose-dependent risk of CDI associated with increasing cumulative dose, number of antimicrobials, and days of antimicrobial exposure. Compared with patients who received fewer than 4 days of antimicrobials, the adjusted hazard ratios (aHR) for those who received 4-7 days or 8-18 days of therapy were 1.4 (95% CI, 0.8-2.4) and 3.0 (95% CI, 1.9-5.0), respectively. This correlates to a threefold increase in CDI risk for patients who received more than 7 days of antimicrobials. More specifically, the empiric use of antipseudomonal ß-lactams (APBL) for more than 48 hours was also found to be an independent risk factor for CDI among 808 patients with Enterobacteriaceae BSI.¹⁶ The risk of CDI within 90 days of BSI was higher among those who received >48 hours of APBL than it was among those who received ≤48 hours (HR, 3.6; 95% CI, 1.5-9.9).

While C difficile may be the most well-known pathogen implicated in antimicrobial usage, the incidence of multidrug-­resistant (MDR) organisms, either as infectious or colonizing pathogens, is also tied to antimicrobial exposure. Among patients receiving systemic antimicrobials, 6% developed an MDR infection within 90 days.¹¹ Over a 5-year period, Teshome and colleagues evaluated 7,118 critically ill patients and demonstrated that prolonged exposures to APBLs increased the risk of new antimicrobial resistance within 60 days.¹⁵ This resistance pattern was not an institutional or environmental finding but a patient-level finding. For each additional day of cefepime or piperacillin/tazobactam received, the risk of new antimicrobial resistance was increased by 8%. The authors concluded that defining a piperacillin/tazobactam course as 10 vs 7 days would result in a 24% higher relative risk of resistance per patient related to those 3 additional days of antimicrobial exposure.

Catheter complications including thrombophlebitis, infiltration, and infection are serious and frequent problems associated with IV medication administration.¹⁷ Even with short-term use, peripherally inserted central catheters (PICCs) carry a substantial risk of venous thrombosis (superficial and deep veins). The incidence of deep vein thrombosis (DVT) for PICCs is estimated between 5% and 15% for hospitalized patients and 2% and 5% for ambulatory patients.¹⁸ A recent randomized controlled trial (RCT) of oral vs IV antimicrobials for bone and joint infections reported that, compared with patients randomized to oral antimicrobials, those randomized to IV antimicrobials were more likely to have catheter complications (9.4% vs 1.0%; P < .001) and to discontinue therapy earlier (18.9% vs 12.8%; P = .006).¹⁹ Median hospital stay was also significantly longer in the IV group (14 days vs 11 days; P < .001).

Optimization of antimicrobial duration has long been recognized as one of the key strategies in reducing unnecessary antimicrobial exposure, yet high-quality evidence on comparative effectiveness of duration in the setting of bacteremia has been limited until recently.²⁰ The presence of bacteremia is often used as a justification for prolonged courses of antimicrobial regardless of infection source or clinical response. The Infectious Diseases Society of America guidelines suggest 7 to 14 days of treatment for intravascular catheter-associated gram-negative bacteremia, but the optimal duration for non–catheter-related gram-negative bacteremia is not addressed.²¹ This lack of clear guidance and the historical scarcity of robust data make it difficult to inform best practices, which leads to wide variability in clinical practice and 14 days being the most prescribed duration.^22,23

Pooled clinical trials’ data from subsets of patients with bacteremia and those from observational studies have been the best available evidence for the treatment duration of GN-BSI until recently (Table 1).^24-32 Two meta-analyses evaluating RCTs of adult and pediatric patients with pyelonephritis, UTI, peritonitis, and pneumonia found no differences in clinical failure, microbiologic cure, or survival between short and long courses of therapy in the subset of patient with associated bacteremia.^24,25 Six heterogeneous RCTs of short vs long courses of therapy for complicated UTI or pyelonephritis reported no differences in clinical cure rates in the subset of patients with associated GN-BSI.² The observational studies outlined in Table 1 are also consistent with RCT results supporting noninferiority in clinical cure and mortality outcomes between short and long courses of therapy.^26-32 These findings may also be extrapolated to immunocompromised hosts given a considerable representation of 10% to 47% of the study population with immunosuppressive conditions.

Nelson and colleagues conducted the only retrospective study to date reporting conflicting results of higher risk of treatment failure (defined as composite endpoint of mortality or recurrent infection within 90 days of index BSI) in patients receiving a short course of therapy.²⁷ However, the difference was driven by 90-day mortality (8.2% vs 3.3%; P = .04) not recurrent infection (6.7% vs 6.5%; P = 0.93). Giannella and colleagues also evaluated 90-day mortality as a primary endpoint in a much larger cohort of over 850 patients in Italy and found no difference in mortality rates between short and long courses of antimicrobials.³⁰

Yahav and colleagues conducted the first well-designed open-label RCT comparing short and long courses of antimicrobials in uncomplicated GN-BSI.³³ This noninferiority study randomized more than 600 hospitalized patients with adequate source control who were afebrile and hemodynamically stable for ≥48 hours to receive either 7 days or 14 days of therapy. The source of infections was predominantly urinary (68%), and the causative pathogens were 90% Enterobacteriacae, including 20% MDR strains. The primary outcome was a composite of 90-day all-cause mortality or clinical failure defined as either relapse of bacteremia, local or distant complications, readmission, or extended hospital stay >14 days. The authors reported no statistically significant differences in the primary outcome between short (45.8%) and long (48.3%) courses of treatment. In the prespecified post hoc analysis designed to evaluate infection-­related outcomes at an earlier time frame, there were no observed differences in complications, relapses, or mortality between study groups at 14 and 28 days. Further subgroup analysis demonstrated similar results among patients with MDR pathogens, primarily extended-spectrum ß-lactamases (ESBL). Interestingly, there was a more rapid return to baseline activity and functional capacity among patients randomized to a short course of therapy. The authors acknowledged that the patients’ perception of illness while taking antimicrobials may have influenced self-reported well-being and functional performance. In exploratory analysis, prolonged hospitalization and readmission were excluded from the primary study endpoint to mirror outcomes assessed by Nelson and colleagues. There were no statistically significant differences in death, relapses, or complications between groups randomized to short (18.6%) or long (15.1%) courses of therapy, with a risk difference of 3.5% (95% CI, –2.5% to 9.5%) in this study population.

Patients with Pseudomonas aeruginosa BSI often have more chronic medical comorbidities, immunocompromised conditions, higher severity of illness, and more indwelling catheters than do patients with Enterobacteriaceae BSI.³² It is uncertain whether shorter duration of therapy is generalizable to this population, given that Pseudomonas accounted for a relatively low number (8%) of infections in the published RCT.³³ Fabre and colleagues included high-risk patients with >65% of the cohort with severe immunocompromised conditions consisting of stem cell transplantation, recent chemotherapy, or neutropenia, and they reported no difference in 30-day mortality or recurrent infections among patients with pseudomonal BSI regardless of duration of therapy.³²

It is a well-accepted standard of practice that BSI are treated with upfront IV antimicrobials that can rapidly achieve therapeutic serum concentration. Whether IV administration is warranted for the entire duration of therapy, though, remains controversial. Even in an era of highly bioavailable oral antimicrobials, clinicians often assume that IV antimicrobials are more potent and efficacious than oral antimicrobials.^8,9 This belief has contributed to the dogma that IV therapy is necessary irrespective of the associated risks and costs. Oral antimicrobials are often overlooked as alternatives despite established benefits in avoiding complications associated with IV catheters, decreasing hospital LOS, and improving quality of life.³⁴ There are promising clinical data in support of the efficacy and safety of transitioning from sequential-IV to highly bioavailable oral agents for the treatment of uncomplicated bacteremia caused by both gram-positive and gram-negative pathogens.^2,35 Highly bioavailable oral antimicrobials are also increasingly integrated as sequential therapy for deep-seated infections in bone and joint infections, such as vertebral osteomyelitis.^19,36 These findings have been confirmed in a recent RCT demonstrating noninferiority of oral antimicrobial combinations after satisfactory clinical responses to at least 10 days of IV therapy, compared with continued IV regimens, in left-­sided infective endocarditis.³⁶ While not a prespecified endpoint, hospital LOS was shorter among patients randomized to oral antimicrobials.

Although there are no large-scale RCTs sufficiently powered to address the role of oral antimicrobials in the treatment of uncomplicated GN-BSI, some insights can be gleaned from the existing literature (Table 2). In the RCT establishing noninferiority of short vs long courses of antimicrobials for uncomplicated GN-BSI, the majority of patients randomized to 7 days vs 14 days of therapy, 64% and 81%, respectively, were de-escalated to oral antimicrobials, with fluoroquinolones (FQs) being the predominant (>70%) oral regimen, followed by trimethoprim/sulfamethoxazole (T/S) and oral ß-lactams.³³

Despite the Food and Drug Administration warnings of the potentially permanent adverse effects involving tendons, muscles, joints, nerves, and most recently, aortic aneurysms and ruptures,³⁷ FQs remain a unique class of drugs with favorable pharmacodynamic and pharmacokinetic properties that achieve approximately equivalent serum and tissue concentration when administered either intravenously or orally. This advantage was recognized early on as a potential IV-sparing therapeutic option. A prospective RCT that evaluated oral vs IV ciprofloxacin as initial empiric therapy among 141 patients with pyelonephritis or complicated UTI (38% with secondary BSI) reported no significant differences in microbiological failure or clinical response between the two treatment groups.³⁸ Two small RCTs have also demonstrated the safety and effectiveness of sequential-IV antimicrobial to oral FQs in the setting of GN-BSI secondary to urinary source and cholangitis.^39,40 Oral ß-lactams, however, achieve substantially lower serum concentration than do their IV counterparts and, accordingly, may be less reliably effective.²

Five retrospective cohort studies have more directly investigated the role of oral antimicrobials in the setting of GN-BSI secondary to common focal infections (Table 2 and Table 3).^41-45 Two observational studies reported no difference of treatment failure among patients who received IV-only therapy vs those who were switched to oral therapy in bacteremia secondary to UTIs.^41,42 Catheter-associated complications were higher in the IV cohort (6.1% vs 0.4%; P = .03).⁴² In the largest multicenter cohort study to date, which included 1,478 patients with Enterobacteriaceae bacteremia, there was no difference in 30-day mortality or recurrent bacteremia between patients converted to oral step-down therapy and patients who received the full course of IV antimicrobials.⁴³ Furthermore, the median hospital LOS was shorter (5 days vs 7 days; P < .001) among patients who were transitioned to oral therapy, a finding that is consistent with other studies.^39-42 In their analysis, the oral antimicrobials were categorized as low-bioavailability (ß-lactams) or high-bioavailability (FQ and T/S), and there was no difference in outcomes when results were stratified by bioavailability. Mercuro and colleagues reported similar clinical success among patients who received oral ß-lactams and those who received FQs as step-down therapy.⁴⁴ Notably, patients were more likely to tolerate ß-lactams without experiencing adverse effects than were those who received FQs (91.7% vs 82.1%; P = .049). In contrast, Kutob and colleagues compared step-down oral antimicrobials categorized as low bioavailability (ß-lactams), moderate bioavailability (ciprofloxacin and T/S), and high bioavailability (levofloxacin). They reported that treatment failures were significantly higher among patients who received low-bioavailability (14%) and moderate-­bioavailability (12%) antimicrobials, compared with those who received the high-bioavailability agent (2%; P = .02).⁴⁵ Interestingly, the bioavailability of ciprofloxacin reaches 85% and T/S approaches 90%, and they are often categorized as highly bioavailable agents in other studies.^43,46 If they were reclassified as highly bioavailable agents, the study conclusions might differ. Nevertheless, the reported success with oral step-down therapy exceeded 85% in all five studies.^41-45

It is important to acknowledge the possibility of unmeasured confounders in these retrospective, observational studies despite statistical adjustments and that they are likely underpowered to determine the clinical significance of oral bioavailability of antimicrobials. In a meta-analysis of published studies and abstracts that included 2,289 patients with Enterobacteriaceae bacteremia, all-cause mortality was similar between patients de-escalated to an oral FQ, T/S, or ß-lactam.⁴⁶ Overall recurrence of infection (bacteremia or primary site) occurred more frequently in patients transitioned to oral ß-lactams than FQs, but relapse of bacteremia was not statistically different between comparator groups. Bioavailability of the oral agents may not be the sole determinant of higher recurrence; adherence may be poor because of the more frequent dosing required for oral ß-lactams to achieve targeted pharmacokinetics. Additionally, suboptimal dosing of oral ß-lactams noted in the studies may have also contributed to the increased recurrences.

After source control has been achieved and bacterial inoculum burden is sufficiently reduced with appropriate upfront IV therapy, the bioavailability of oral antimicrobials may become less important. However, existing observational data indicate clinical experience is most established with highly bioavailable oral agents, particularly FQs, though the risks vs benefits require careful consideration. For now, the preferred oral agent remains uncertain and selections should be individualized based on susceptibility, patient factors, and other clinical considerations. More importantly, if there are no contraindications or concerns of malabsorption, oral step-down therapy should be initiated as soon as source control and good clinical responses have been achieved.

Routine follow-up blood cultures (FUBCs) are strongly recommended in Staphylococcus aureus bacteremia because of the propensity for endovascular and metastatic infection, which dictates clinical decision-making regarding duration of therapy. In contrast, GN-BSI secondary to focal infections is usually transient, and the need for confirmation of blood culture clearance is less clear. The low yield of FUBC in many clinical settings suggests that it may not be helpful.^47-50 Despite the questionable impact on the clinical management of GN-BSI, FUBCs are routinely ordered in the hospital.^1,47 Although there is no high-quality evidence addressing the utility of FUBC, several observational studies suggest clinicians should reconsider routinely ordering FUBC.

Canzoneri and colleagues retrospectively evaluated 383 episodes of bacteremia with at least one FUBC drawn after the initial blood culture.⁴⁸ On average, 2.32 FUBC were performed per patient for GN-BSI episode, and only 8 patients (5.7%) had persistent bacteremia. Specifically, only 3% had documented positive FUBC among patients with urinary tract source of infection. It was estimated that 17 FUBCs are needed to yield one positive result for GN-BSI. This finding is consistent with results from another study that examined 1,801 episodes of bacteremia, 901 of which were gram-negative organisms, predominantly (67%) Escherichia coli and Klebsiella spp.⁴⁹ Among GN-BSI episodes, FUBCs were performed in 247 cases, with 27 (10.9%) cases demonstrating persistent bacteremia. A nested case-­control analysis between patients with cleared or persistent bacteremia found a lower yield in FUBC with gram-negative organisms and a genitourinary source of infection. Moreover, persistent bacteremia did not influence a change in antimicrobial regimen. Kang and colleagues investigated 1,068 episodes of Klebsiella pneumoniae bacteremia, with FUBCs performed in 862 (80.7%) cases despite only a 7.2% incidence of persistent bacteremia.⁵⁰ The independent risk factors associated with persistent bacteremia were intra-abdominal infection, solid organ transplantation, high Charlson comorbidity index score, and unfavorable treatment responses, which suggests the need for FUBC may be individualized rather than routine.

In the setting of GN-BSI in which the probability of persistent bacteremia is relatively low, especially in genitourinary sources of infection, FUBCs are not warranted. It is uncomfortable for patients and exposes them to harms of false-positive results, leading to antimicrobial administration with possible adverse effects, which can be further compounded by unnecessary testing, potentially missed alternative diagnosis, and increasing hospital LOS.^1,51 Given the low yield of FUBC in GN-BSI, and the lack of association of persistent bacteremia with change in antimicrobial therapy or clinical outcomes, we recommend avoiding FUBC as a test of cure. Documentation of gram-negative blood culture clearance should be reserved for situations in which there is concern for deeper or otherwise uncontrolled source of infection.

The optimal management of gram-negative bacteremia in hospitalized patients is evolving. There is a growing body of evidence supporting shorter duration for a total of 7 days with oral step-down therapy as safe and effective for patients with uncomplicated Enterobacteriaceae bacteremia who have achieved adequate source control and demonstrated clinical stability and improvement. Although comparative data regarding the optimal duration of therapy in the setting of MDR strains such as ESBL Enterobacteriaceae and pseudomonal BSI are limited, available data appear promising in favor of shorter treatment duration with oral step-down therapy. Routine follow-up blood culture is not cost-effective and may result in unnecessary healthcare resource utilization and inappropriate use of antimicrobials. Table 4 provides a framework for the clinical management of GN-BSI in the hospital. Taken together, these steps will facilitate antimicrobial stewardship, limit unnecessary antimicrobial exposure, and improve quality of patient care.

The authors have no conflicts of interest to disclose.

Uncomplicated bacteremia, while not precisely defined in the literature, generally implies bacteremia in the absence of a persistent or difficult-to-eradicate infectious source. Bacteremia secondary to focal infections such as skin and soft-tissue infection, pneumonia, pyelonephritis, or urinary tract infection (UTI) accounts for up to 25% of bloodstream infections (BSIs) and usually resolves with prompt and appropriate antimicrobial therapy.^1,2 Current practice guidelines do not adequately address key aspects of the optimal management of gram-negative (GN)–BSI commonly encountered in hospital care.^3-7 Notably, antimicrobial duration, criteria to transition from intravenous (IV) to oral step-down therapy, choice of oral antimicrobials, and reassessment of follow-up blood cultures have not been addressed. In the absence of consensus guidelines, clinicians rely on “conventional wisdom” and clinical experience, which may not be supported by scientific rigor. A growing body of research now challenges some long-standing practices once thought to be standard of care.

In this narrative review, we aim to examine and synthesize emerging information to provide an evidence-based framework in the management of hospitalized patients with GN-BSI. We highlight the unintended consequences and potential harms of excessive antimicrobial exposure and focus on areas in the fundamental approach to duration of therapy, the role of oral antimicrobials, and usefulness of follow-up blood cultures. A comprehensive search of the published literature was performed in PubMed with an emphasis on articles published during 2015-2019 with use of search terms including gram-negative bacteremia, duration, antibiotics, adverse effects, intravascular catheter, and follow-up blood cultures.

Antimicrobial overuse is common and may be driven by concerns for undertreatment. Clinicians may believe that prolonged antimicrobial therapy maximizes cure rates, with treatment duration often defined arbitrarily by a fixed number of “Constantine-units” (dating back to the ancient Roman emperor’s decree of 7 days in a week).^8-10 Recent publications refute this notion and point out that the harms of overprescribing outweigh the perceived benefits of longer treatment duration.

Antimicrobials are lifesaving but not benign; adverse effects are common and costly to our patients and healthcare system. Among 1,488 hospitalized adults who received at least 24 hours of systemic antimicrobials, 20% had an antimicrobial-­associated adverse event, mostly gastrointestinal, renal, or hematologic in nature.¹¹ Prolonged duration of antimicrobials is further associated with adverse effects such as antimicrobial-associated diarrhea, increased rates of Clostridioides difficile infection (CDI), emergence of antimicrobial resistance, and longer hospital length of stay (LOS).^11-15 Vaughn and colleagues conducted the largest observational study to date, evaluating antimicrobial prescriptions for the treatment of nearly 6,500 adults with community-acquired pneumonia in a 43-hospital consortium in Michigan.¹⁴ More than two-thirds of patients received antimicrobial courses (median 8 days) that exceeded guideline-­recommended duration. Patients who received longer antimicrobial courses did not have reduced mortality, readmission, or emergency department visits. More importantly, each excess day of treatment was associated with a relative 5% increase in the odds of antimicrobial-associated adverse effects reported by patients. This is further supported by national and state hospital data that antimicrobial-associated adverse events are an independent predictor of longer LOS.¹²

CDI is commonly linked to destructive changes to the indigenous microbiota of the intestinal flora caused by antimicrobial administration. Stevens and colleagues identified 7,792 hospitalized patients who received at least 2 consecutive days of antimicrobial therapy¹³; comparing 241 cases of CDI with the control group, they observed a dose-dependent risk of CDI associated with increasing cumulative dose, number of antimicrobials, and days of antimicrobial exposure. Compared with patients who received fewer than 4 days of antimicrobials, the adjusted hazard ratios (aHR) for those who received 4-7 days or 8-18 days of therapy were 1.4 (95% CI, 0.8-2.4) and 3.0 (95% CI, 1.9-5.0), respectively. This correlates to a threefold increase in CDI risk for patients who received more than 7 days of antimicrobials. More specifically, the empiric use of antipseudomonal ß-lactams (APBL) for more than 48 hours was also found to be an independent risk factor for CDI among 808 patients with Enterobacteriaceae BSI.¹⁶ The risk of CDI within 90 days of BSI was higher among those who received >48 hours of APBL than it was among those who received ≤48 hours (HR, 3.6; 95% CI, 1.5-9.9).

While C difficile may be the most well-known pathogen implicated in antimicrobial usage, the incidence of multidrug-­resistant (MDR) organisms, either as infectious or colonizing pathogens, is also tied to antimicrobial exposure. Among patients receiving systemic antimicrobials, 6% developed an MDR infection within 90 days.¹¹ Over a 5-year period, Teshome and colleagues evaluated 7,118 critically ill patients and demonstrated that prolonged exposures to APBLs increased the risk of new antimicrobial resistance within 60 days.¹⁵ This resistance pattern was not an institutional or environmental finding but a patient-level finding. For each additional day of cefepime or piperacillin/tazobactam received, the risk of new antimicrobial resistance was increased by 8%. The authors concluded that defining a piperacillin/tazobactam course as 10 vs 7 days would result in a 24% higher relative risk of resistance per patient related to those 3 additional days of antimicrobial exposure.

Catheter complications including thrombophlebitis, infiltration, and infection are serious and frequent problems associated with IV medication administration.¹⁷ Even with short-term use, peripherally inserted central catheters (PICCs) carry a substantial risk of venous thrombosis (superficial and deep veins). The incidence of deep vein thrombosis (DVT) for PICCs is estimated between 5% and 15% for hospitalized patients and 2% and 5% for ambulatory patients.¹⁸ A recent randomized controlled trial (RCT) of oral vs IV antimicrobials for bone and joint infections reported that, compared with patients randomized to oral antimicrobials, those randomized to IV antimicrobials were more likely to have catheter complications (9.4% vs 1.0%; P < .001) and to discontinue therapy earlier (18.9% vs 12.8%; P = .006).¹⁹ Median hospital stay was also significantly longer in the IV group (14 days vs 11 days; P < .001).

Optimization of antimicrobial duration has long been recognized as one of the key strategies in reducing unnecessary antimicrobial exposure, yet high-quality evidence on comparative effectiveness of duration in the setting of bacteremia has been limited until recently.²⁰ The presence of bacteremia is often used as a justification for prolonged courses of antimicrobial regardless of infection source or clinical response. The Infectious Diseases Society of America guidelines suggest 7 to 14 days of treatment for intravascular catheter-associated gram-negative bacteremia, but the optimal duration for non–catheter-related gram-negative bacteremia is not addressed.²¹ This lack of clear guidance and the historical scarcity of robust data make it difficult to inform best practices, which leads to wide variability in clinical practice and 14 days being the most prescribed duration.^22,23

Pooled clinical trials’ data from subsets of patients with bacteremia and those from observational studies have been the best available evidence for the treatment duration of GN-BSI until recently (Table 1).^24-32 Two meta-analyses evaluating RCTs of adult and pediatric patients with pyelonephritis, UTI, peritonitis, and pneumonia found no differences in clinical failure, microbiologic cure, or survival between short and long courses of therapy in the subset of patient with associated bacteremia.^24,25 Six heterogeneous RCTs of short vs long courses of therapy for complicated UTI or pyelonephritis reported no differences in clinical cure rates in the subset of patients with associated GN-BSI.² The observational studies outlined in Table 1 are also consistent with RCT results supporting noninferiority in clinical cure and mortality outcomes between short and long courses of therapy.^26-32 These findings may also be extrapolated to immunocompromised hosts given a considerable representation of 10% to 47% of the study population with immunosuppressive conditions.

Nelson and colleagues conducted the only retrospective study to date reporting conflicting results of higher risk of treatment failure (defined as composite endpoint of mortality or recurrent infection within 90 days of index BSI) in patients receiving a short course of therapy.²⁷ However, the difference was driven by 90-day mortality (8.2% vs 3.3%; P = .04) not recurrent infection (6.7% vs 6.5%; P = 0.93). Giannella and colleagues also evaluated 90-day mortality as a primary endpoint in a much larger cohort of over 850 patients in Italy and found no difference in mortality rates between short and long courses of antimicrobials.³⁰

Yahav and colleagues conducted the first well-designed open-label RCT comparing short and long courses of antimicrobials in uncomplicated GN-BSI.³³ This noninferiority study randomized more than 600 hospitalized patients with adequate source control who were afebrile and hemodynamically stable for ≥48 hours to receive either 7 days or 14 days of therapy. The source of infections was predominantly urinary (68%), and the causative pathogens were 90% Enterobacteriacae, including 20% MDR strains. The primary outcome was a composite of 90-day all-cause mortality or clinical failure defined as either relapse of bacteremia, local or distant complications, readmission, or extended hospital stay >14 days. The authors reported no statistically significant differences in the primary outcome between short (45.8%) and long (48.3%) courses of treatment. In the prespecified post hoc analysis designed to evaluate infection-­related outcomes at an earlier time frame, there were no observed differences in complications, relapses, or mortality between study groups at 14 and 28 days. Further subgroup analysis demonstrated similar results among patients with MDR pathogens, primarily extended-spectrum ß-lactamases (ESBL). Interestingly, there was a more rapid return to baseline activity and functional capacity among patients randomized to a short course of therapy. The authors acknowledged that the patients’ perception of illness while taking antimicrobials may have influenced self-reported well-being and functional performance. In exploratory analysis, prolonged hospitalization and readmission were excluded from the primary study endpoint to mirror outcomes assessed by Nelson and colleagues. There were no statistically significant differences in death, relapses, or complications between groups randomized to short (18.6%) or long (15.1%) courses of therapy, with a risk difference of 3.5% (95% CI, –2.5% to 9.5%) in this study population.

Patients with Pseudomonas aeruginosa BSI often have more chronic medical comorbidities, immunocompromised conditions, higher severity of illness, and more indwelling catheters than do patients with Enterobacteriaceae BSI.³² It is uncertain whether shorter duration of therapy is generalizable to this population, given that Pseudomonas accounted for a relatively low number (8%) of infections in the published RCT.³³ Fabre and colleagues included high-risk patients with >65% of the cohort with severe immunocompromised conditions consisting of stem cell transplantation, recent chemotherapy, or neutropenia, and they reported no difference in 30-day mortality or recurrent infections among patients with pseudomonal BSI regardless of duration of therapy.³²

It is a well-accepted standard of practice that BSI are treated with upfront IV antimicrobials that can rapidly achieve therapeutic serum concentration. Whether IV administration is warranted for the entire duration of therapy, though, remains controversial. Even in an era of highly bioavailable oral antimicrobials, clinicians often assume that IV antimicrobials are more potent and efficacious than oral antimicrobials.^8,9 This belief has contributed to the dogma that IV therapy is necessary irrespective of the associated risks and costs. Oral antimicrobials are often overlooked as alternatives despite established benefits in avoiding complications associated with IV catheters, decreasing hospital LOS, and improving quality of life.³⁴ There are promising clinical data in support of the efficacy and safety of transitioning from sequential-IV to highly bioavailable oral agents for the treatment of uncomplicated bacteremia caused by both gram-positive and gram-negative pathogens.^2,35 Highly bioavailable oral antimicrobials are also increasingly integrated as sequential therapy for deep-seated infections in bone and joint infections, such as vertebral osteomyelitis.^19,36 These findings have been confirmed in a recent RCT demonstrating noninferiority of oral antimicrobial combinations after satisfactory clinical responses to at least 10 days of IV therapy, compared with continued IV regimens, in left-­sided infective endocarditis.³⁶ While not a prespecified endpoint, hospital LOS was shorter among patients randomized to oral antimicrobials.

Although there are no large-scale RCTs sufficiently powered to address the role of oral antimicrobials in the treatment of uncomplicated GN-BSI, some insights can be gleaned from the existing literature (Table 2). In the RCT establishing noninferiority of short vs long courses of antimicrobials for uncomplicated GN-BSI, the majority of patients randomized to 7 days vs 14 days of therapy, 64% and 81%, respectively, were de-escalated to oral antimicrobials, with fluoroquinolones (FQs) being the predominant (>70%) oral regimen, followed by trimethoprim/sulfamethoxazole (T/S) and oral ß-lactams.³³

Despite the Food and Drug Administration warnings of the potentially permanent adverse effects involving tendons, muscles, joints, nerves, and most recently, aortic aneurysms and ruptures,³⁷ FQs remain a unique class of drugs with favorable pharmacodynamic and pharmacokinetic properties that achieve approximately equivalent serum and tissue concentration when administered either intravenously or orally. This advantage was recognized early on as a potential IV-sparing therapeutic option. A prospective RCT that evaluated oral vs IV ciprofloxacin as initial empiric therapy among 141 patients with pyelonephritis or complicated UTI (38% with secondary BSI) reported no significant differences in microbiological failure or clinical response between the two treatment groups.³⁸ Two small RCTs have also demonstrated the safety and effectiveness of sequential-IV antimicrobial to oral FQs in the setting of GN-BSI secondary to urinary source and cholangitis.^39,40 Oral ß-lactams, however, achieve substantially lower serum concentration than do their IV counterparts and, accordingly, may be less reliably effective.²

Five retrospective cohort studies have more directly investigated the role of oral antimicrobials in the setting of GN-BSI secondary to common focal infections (Table 2 and Table 3).^41-45 Two observational studies reported no difference of treatment failure among patients who received IV-only therapy vs those who were switched to oral therapy in bacteremia secondary to UTIs.^41,42 Catheter-associated complications were higher in the IV cohort (6.1% vs 0.4%; P = .03).⁴² In the largest multicenter cohort study to date, which included 1,478 patients with Enterobacteriaceae bacteremia, there was no difference in 30-day mortality or recurrent bacteremia between patients converted to oral step-down therapy and patients who received the full course of IV antimicrobials.⁴³ Furthermore, the median hospital LOS was shorter (5 days vs 7 days; P < .001) among patients who were transitioned to oral therapy, a finding that is consistent with other studies.^39-42 In their analysis, the oral antimicrobials were categorized as low-bioavailability (ß-lactams) or high-bioavailability (FQ and T/S), and there was no difference in outcomes when results were stratified by bioavailability. Mercuro and colleagues reported similar clinical success among patients who received oral ß-lactams and those who received FQs as step-down therapy.⁴⁴ Notably, patients were more likely to tolerate ß-lactams without experiencing adverse effects than were those who received FQs (91.7% vs 82.1%; P = .049). In contrast, Kutob and colleagues compared step-down oral antimicrobials categorized as low bioavailability (ß-lactams), moderate bioavailability (ciprofloxacin and T/S), and high bioavailability (levofloxacin). They reported that treatment failures were significantly higher among patients who received low-bioavailability (14%) and moderate-­bioavailability (12%) antimicrobials, compared with those who received the high-bioavailability agent (2%; P = .02).⁴⁵ Interestingly, the bioavailability of ciprofloxacin reaches 85% and T/S approaches 90%, and they are often categorized as highly bioavailable agents in other studies.^43,46 If they were reclassified as highly bioavailable agents, the study conclusions might differ. Nevertheless, the reported success with oral step-down therapy exceeded 85% in all five studies.^41-45

It is important to acknowledge the possibility of unmeasured confounders in these retrospective, observational studies despite statistical adjustments and that they are likely underpowered to determine the clinical significance of oral bioavailability of antimicrobials. In a meta-analysis of published studies and abstracts that included 2,289 patients with Enterobacteriaceae bacteremia, all-cause mortality was similar between patients de-escalated to an oral FQ, T/S, or ß-lactam.⁴⁶ Overall recurrence of infection (bacteremia or primary site) occurred more frequently in patients transitioned to oral ß-lactams than FQs, but relapse of bacteremia was not statistically different between comparator groups. Bioavailability of the oral agents may not be the sole determinant of higher recurrence; adherence may be poor because of the more frequent dosing required for oral ß-lactams to achieve targeted pharmacokinetics. Additionally, suboptimal dosing of oral ß-lactams noted in the studies may have also contributed to the increased recurrences.

After source control has been achieved and bacterial inoculum burden is sufficiently reduced with appropriate upfront IV therapy, the bioavailability of oral antimicrobials may become less important. However, existing observational data indicate clinical experience is most established with highly bioavailable oral agents, particularly FQs, though the risks vs benefits require careful consideration. For now, the preferred oral agent remains uncertain and selections should be individualized based on susceptibility, patient factors, and other clinical considerations. More importantly, if there are no contraindications or concerns of malabsorption, oral step-down therapy should be initiated as soon as source control and good clinical responses have been achieved.

Routine follow-up blood cultures (FUBCs) are strongly recommended in Staphylococcus aureus bacteremia because of the propensity for endovascular and metastatic infection, which dictates clinical decision-making regarding duration of therapy. In contrast, GN-BSI secondary to focal infections is usually transient, and the need for confirmation of blood culture clearance is less clear. The low yield of FUBC in many clinical settings suggests that it may not be helpful.^47-50 Despite the questionable impact on the clinical management of GN-BSI, FUBCs are routinely ordered in the hospital.^1,47 Although there is no high-quality evidence addressing the utility of FUBC, several observational studies suggest clinicians should reconsider routinely ordering FUBC.

Canzoneri and colleagues retrospectively evaluated 383 episodes of bacteremia with at least one FUBC drawn after the initial blood culture.⁴⁸ On average, 2.32 FUBC were performed per patient for GN-BSI episode, and only 8 patients (5.7%) had persistent bacteremia. Specifically, only 3% had documented positive FUBC among patients with urinary tract source of infection. It was estimated that 17 FUBCs are needed to yield one positive result for GN-BSI. This finding is consistent with results from another study that examined 1,801 episodes of bacteremia, 901 of which were gram-negative organisms, predominantly (67%) Escherichia coli and Klebsiella spp.⁴⁹ Among GN-BSI episodes, FUBCs were performed in 247 cases, with 27 (10.9%) cases demonstrating persistent bacteremia. A nested case-­control analysis between patients with cleared or persistent bacteremia found a lower yield in FUBC with gram-negative organisms and a genitourinary source of infection. Moreover, persistent bacteremia did not influence a change in antimicrobial regimen. Kang and colleagues investigated 1,068 episodes of Klebsiella pneumoniae bacteremia, with FUBCs performed in 862 (80.7%) cases despite only a 7.2% incidence of persistent bacteremia.⁵⁰ The independent risk factors associated with persistent bacteremia were intra-abdominal infection, solid organ transplantation, high Charlson comorbidity index score, and unfavorable treatment responses, which suggests the need for FUBC may be individualized rather than routine.

In the setting of GN-BSI in which the probability of persistent bacteremia is relatively low, especially in genitourinary sources of infection, FUBCs are not warranted. It is uncomfortable for patients and exposes them to harms of false-positive results, leading to antimicrobial administration with possible adverse effects, which can be further compounded by unnecessary testing, potentially missed alternative diagnosis, and increasing hospital LOS.^1,51 Given the low yield of FUBC in GN-BSI, and the lack of association of persistent bacteremia with change in antimicrobial therapy or clinical outcomes, we recommend avoiding FUBC as a test of cure. Documentation of gram-negative blood culture clearance should be reserved for situations in which there is concern for deeper or otherwise uncontrolled source of infection.

The optimal management of gram-negative bacteremia in hospitalized patients is evolving. There is a growing body of evidence supporting shorter duration for a total of 7 days with oral step-down therapy as safe and effective for patients with uncomplicated Enterobacteriaceae bacteremia who have achieved adequate source control and demonstrated clinical stability and improvement. Although comparative data regarding the optimal duration of therapy in the setting of MDR strains such as ESBL Enterobacteriaceae and pseudomonal BSI are limited, available data appear promising in favor of shorter treatment duration with oral step-down therapy. Routine follow-up blood culture is not cost-effective and may result in unnecessary healthcare resource utilization and inappropriate use of antimicrobials. Table 4 provides a framework for the clinical management of GN-BSI in the hospital. Taken together, these steps will facilitate antimicrobial stewardship, limit unnecessary antimicrobial exposure, and improve quality of patient care.

The authors have no conflicts of interest to disclose.

Posttraumatic stress disorder (PTSD) is a syndrome that occurs after exposure to a significant traumatic event and is characterized by persistent, debilitating symptoms that fall into four “diagnostic clusters” as outlined in the Diagnostic and Statistical Manual of Mental Disorders-Version V (DSM-V). Patients may experience intrusive thoughts, avoidance of distressing stimuli, persistent negative mood, and hypervigilance, all of which last longer than 1 month.¹

A national survey of United States households conducted during 2001-2003 estimated the 12-month prevalence of PTSD among adults to be 3.5%.² Lifetime prevalence has been found to be between 6.8%³ and 7.8%.⁴ PTSD is more common in veterans. The prevalence of PTSD in veterans differs depending on the conflict in which the veteran participated. Vietnam veterans have an estimated lifetime prevalence of approximately 30%,^5,6 Gulf War veterans approximately 15%,⁷ and veterans of more recent conflicts in Afghanistan and Iraq of approximately 21%.⁸ With the MISSION Act moving more veteran care into the private sector, non-VA inpatient providers will need to become better versed in PTSD.⁹

Patients with PTSD have more contact with the healthcare system, even for non–mental health problems,^8,10-13 and a significantly higher burden of medical comorbities,¹⁴ such as diabetes mellitus, liver disease, gastritis and gastric ulcers, HIV, arthritis,¹⁵ and coronary heart disease.¹⁶ Veterans with PTSD are hospitalized three times more often than are those with no mental health diagnoses,⁸ and patients with psychiatric comorbidities have higher lengths of stay.¹⁷ More than 1.4 million hospitalizations occurring during 2002-2011 had either a primary or secondary associated diagnosis of PTSD, with total inflation-­adjusted charges of 34.9 billion dollars.¹⁸ In the inpatient sample from this study, greater than half were admitted for a primary diagnosis of mental diseases and disorders (Major Diagnostic Category [MDC] 19). Following mental illness, the most common primary diagnoses for men were MDC 5 (Circulatory System, 12.1%), MDC 20 (Alcohol/Drug Use or Induced Mental Disorder, 9.2%), and MDC 4 (Respiratory System, 7.4%), while the most common categories for women were MDC 20 (Alcohol/Drug Use or Induced Mental Disorder, 5.8%), MDC 21 (Injuries, Poison, and Toxic Effect of Drugs, 4.9%), and MDC 6 (Digestive System, 4.5%).¹⁸

In both the inpatient and outpatient settings, a fundamental challenge to comprehensive PTSD management is correctly diagnosing this condition.¹⁹ Confounding the difficulties in diagnosis are numerous comorbidities. In addition to the physical comorbidities described above, more than 70% of patients with PTSD have another psychological comorbidity such as affective disorders, anxiety disorders, or substance use disorder/dependency.²⁰

Given that PTSD may be an underrecognized burden on the healthcare system, we sought to better understand how PTSD could affect hospitalized patients admitted for medical problems by conducting this narrative review. Additionally, three of the authors collaborated with the VA Employee Education Service to conduct a needs assessment of VA hospitalists in 2013. Respondents identified managing and educating patients and families about PTSD as a major educational need (unpublished data available upon request from the corresponding author). Therefore, our aims were to present a synthesis of existing literature, familiarize readers with the tenets of trauma-­informed care as a framework to guide care for these patients, and generate ideas for changes that inpatient providers could implement now. We began by consulting a research librarian at the Clement J. Zablocki VA Medical Center in Milwaukee, Wisconsin, who searched the following databases: PsycInfo, CINAHL, MEDLINE, and PILOTS (a PTSD/trauma specific database). Search terms included hospital, hospitalized, and hospitalization, as well as traumatic stress, posttraumatic stress, and PTSD. Pertinent guidelines and the reference lists from included papers were examined. We focused on papers that described patients admitted for medical problems other than PTSD because those patients who are admitted for PTSD-­related problems should be primarily ma­naged by psychiatry (not hospitalists) with the primary focus of their hospitalization being their PTSD. We also excluded papers about patients developing PTSD secondary to hospitalization, which already has a well-developed literature.^21-23

The literature is sparse describing frequency or type of problems encountered by hospitalized medical patients with PTSD. A recent small survey study reported that 40% of patients anticipated triggers for their PTSD symptoms in the hospital; such triggers included loud noises and being shaken awake.²⁴ Two papers describe case vignettes of patients who had exacerbations of their PTSD while in the Intensive Care Unit (ICU), although neither contain frequency or severity data.^25,26 Approximately 8% of patients in VA ICUs have PTSD,²⁷ and a published abstract suggests that they appear to require more sedation than do patients without PTSD.²⁸ Another published case report describes a patient with recurrent PTSD symptoms (nightmares) after moving into a nursing home.²⁹ These papers suggest other providers have recognized and are concerned about hospitalized patients with PTSD. At present, there are no data to quantify how often hospitalized patients have PTSD exacerbations or how troublesome such exacerbations are to these patients.

Given that there is little empiric literature to guide inpatient management of PTSD as a comorbidity in hospitalized medical patients, we extrapolate some information from the outpatient setting. PTSD is often underdiagnosed and underreported by individual patients in the outpatient setting.³⁰ Failure to have an associated diagnosis of PTSD may lead to underrecognition and undertreatment of these patients by inpatient providers in the hospital setting. Additionally, the numerous psychological and physical comorbidities in PTSD can create unique challenges in properly managing any single problem in these patients.²⁰ Armed with this knowledge, providers should be vigilant in the recognition, assessment, and treatment of PTSD.

Trauma-Informed Care: A Conceptual Model

Trauma-informed care is a mindful and sensitive approach to caring for patients who have suffered trauma.³¹ It requires understanding that many people have suffered trauma in their lives and that the trauma continues to impact many aspects of their lives.³² Trauma-informed care has many advocates and has been implemented across myriad health and social services settings.³³ Its principles can be applied in both the inpatient and outpatient hospital settings. While it is an appropriate approach to patients with PTSD, it is not specific to PTSD. People who have suffered sexual trauma, intimate partner violence, child abuse, or other exposures would also be included in the group of people for whom trauma-informed care is a suitable approach. There are four key assumptions to a trauma-­informed approach to care (the 4 R’s): (1) realization that trauma affects an individual’s coping strategies, relationships, and health; (2) recognition of the signs of trauma; (3) having an appropriate, planned response to patients identified as having suffered a trauma; and (4) resisting retraumatization in the care setting.^31,32

General Approach to Treating Medical Patients With PTSD in the Inpatient Setting

Recognition

Consistent with a trauma-informed care approach, inpatient providers should be able to recognize patients who may have PTSD. First, careful review of the past medical history may show some patients already carry this diagnosis. Second, patients with PTSD often have other comorbidities that could offer a clue that PTSD could be present as well; for example, risk for PTSD is increased when mood, anxiety, or substance use disorders are present.²⁰ When PTSD is suspected, screening is a reasonable next step.

The Primary Care-PTSD-5 (PC-PTSD-5) is a validated screening tool used in the outpatient setting.³⁴ It is easily administered and has good predictive validity (positive likelihood ratio [LR+] of 6.33 and LR– of 0.06). It begins with a question of whether the patient has ever experienced a trauma. A positive initial response triggers a series of five yes/no questions. Answering “yes” to three or more questions is a positive screen. A positive screen should result in consultation to psychiatry to conduct more formal evaluation and guide longer-term management.

Collaboration

Individual trauma-focused psychotherapy is the primary treatment of choice for PTSD with strong evidence supporting its practice.³⁵ This treatment is administered by a psychiatrist or psychologist and will be limited in the inpatient medical setting. Current recommendations suggest pharmacotherapy only when individualized trauma-focused psychotherapy is not available, the patient declines it, or as an adjunct when psychotherapy alone is not effective.³⁶ Therefore, inpatient providers may see patients who are prescribed selective serotonin reuptake inhibitors (eg, paroxetine, fluoxetine) or serotonin and norepinephrine reuptake inhibitors (eg, venlafaxine).³⁶ In the past, PTSD-related nightmares were often treated with prazosin.³⁷ However, a recent randomized controlled trial of prazosin in veterans with PTSD failed to show significant improvement in nightmares.³⁸ Hence, current guidelines do not recommend prazosin as a first-line therapy.³⁹ For hospitalized patients with PTSD symptoms refractory to the interventions outlined herein, particularly those patients with possible borderline personality traits (as suggested by severe anger and impulsivity), we strongly recommend partnering with psychiatry. Finally, given the high prevalence of substance use disorders (SUDs) in PTSD patients, awareness and treatment of comorbidities such as opioid and alcohol dependence must be concurrently addressed.

Individualizing Care

It is essential for the healthcare team to identify ways to meet each patient’s immediate needs. Many of the ideas proposed below are not specific to PTSD; many require an interprofessional approach to care.⁴⁰ From a trauma-informed care standpoint, this is akin to having a planned response for patients who have suffered trauma. Assessing the individual’s needs and incorporating therapeutic modalities such as reflective listening, broadening safe opportunities for control, and providing complementary and integrative medicine (IM) therapies may help manage symptoms and establish rapport.⁴¹ Through reflective listening, a collaborative approach can be established to identify background, triggers, and a safe approach for managing PTSD and its comorbid conditions. Ensuring frequent communication and allowing the patient to be at the center of decision-making establishes a safe environment and promotes positive rapport between the patient and healthcare team.³⁶ Providing a sense of control by involving the patients in their healthcare decisions and in the structure of care delivery may benefit the patients’ well-being. Furthermore, incorporating IM encourages rest and relaxation in the chaotic hospital environment. Suggested IM interventions include deep breathing, aromatherapy, guided imagery, muscle relaxation, and music therapy.^42,43

Key Inpatient Issues Affecting PTSD

In the following sections, we outline common clinical situations that may exacerbate PTSD symptoms and propose some evidence-based responses (Table). In general, nonpharmacologic approaches are favored over pharmacologic approaches for patients with PTSD.

Sleep Hygiene

Sleep problems are very common in patients with PTSD, with nightmares occurring in more than 70% of patients and insomnia in 80%.⁴⁴ In PTSD, sleep problems are linked to poor physical health and other health outcomes^45,46 and may exacerbate other PTSD symptoms.⁴

Treating the sleep problems that occur with PTSD is an important aspect of PTSD care. Usually administered in the outpatient setting, the treatment of choice is cognitive-behavioral therapy (CBT).⁴⁸ Sleep-specific CBT focuses, among other things, on strategies that encourage good sleep hygiene,⁴⁹ which includes promoting regular sleep/wake-up times and specific bedtime routines, avoiding stimulation (eg, light, noise, TV) or excessive liquids before bed, refraining from daytime naps, and using relaxation techniques. Many of these recommendations seem at odds with hospital routines, which may contribute to decompensation of hospitalized patients with PTSD.

While starting sleep-specific CBT in the hospital may not be realistic, we suggest the following goals and strategies as a starting place for promoting healthy sleep for hospitalized patients with PTSD. To begin, factors affecting sleep hygiene should be addressed. Inpatient providers could pay more attention to intravenous (IV) fluid orders, perhaps adjusting them to run only during the daytime hours. Medications can be scheduled at times conducive to maintaining home routines. Avoiding the administration of diuretics close to bedtime may decrease the likelihood of frequent nighttime wakening. Grouping patient care activities, such as bathing or wound care, during daytime hours may allow more opportunities for rest at night. Incorporating uninterrupted sleep protocols, such as quiet hours between 10pm and 6am, may enhance sleep quality in the appropriately selected patient.⁵⁰ Although pharmacological interventions to improve sleep in the hospital may be initially beneficial, nonpharmacological interventions as described above should be incorporated for long-term maintenance of enhanced sleep quality.

Second, providers need to ask about established home bedtime routines and facilitate implementation in the hospital. Through collaboration with patients, providers can incorporate an individualized plan of care for sleep early in hospitalization.⁵⁰ Partnering with nurses is also essential to creating a sleep-friendly environment that can improve patient experiences.⁵¹ Breathing exercises, meditating, listening to music and praying are all examples of “bedtime wind down” strategies recommended in sleep-specific CBT.⁴⁹ Many of these could be successfully implemented in the hospital and may benefit other hospitalized patients too.⁵² In patients with PTSD and obstructive sleep apnea, continuous positive airway pressure (CPAP) reduces nightmares, and if inpatients are on CPAP at home, it should be continued in the hospital.⁵³

Pain

If sleep disturbance is the hallmark of PTSD,⁴⁷ chronic pain is its coconspirator.¹⁵ Uncontrolled pain can make it much more difficult to treat patients with PTSD, which in turn may lead to further decompensation from a mental health standpoint.⁵⁴ SUDs such as alcohol or opioid dependencies are highly comorbid with PTSD⁴⁵ and introduce a layer of complexity when managing painin these patients. Providers should be thoughtful when electing to treat acute or chronic pain with opioids and take particular care to establish realistic therapeutic goals if doing so. While patients with PTSD have a greater likelihood of having an SUD, undertreating pain risks exacerbating underlying PTSD symptoms.

Nonpharmacologic therapies, which include communicating, listening, and expressing compassion and understanding, should be utilized by inpatient providers as a first-line treatment in patients with PTSD who suffer from pain. Additionally, relaxation techniques, physical therapy, and physical activity⁵⁵ can be offered. Pharmacologically, nonopioid medications such as acetaminophen or NSAIDs should always be considered first. Should the use of opioids be deemed necessary, inpatient providers should preferentially use oral over intravenous medications and consider establishing a fixed timeframe for short-term opioids, which should be limited to a few days. Providers should communicate clear expectations with their patients to maximize the desired effect of any specific treatment while minimizing the risk of medication side effects with the goal of agreeing on a short yet effective treatment course.

Anxiety and Anger

One of the most challenging situations for the inpatient provider is encountering a patient who is anxious, angry, or hypervigilant. Mismatch between actual and expected communication between the provider and the patient can lead to frustration and anxiety. A trauma-informed care approach would suggest that frequent and thorough communication with patients may prevent or ameliorate the stresses and anxieties of hospitalization that may manifest as anger because of retraumatization. Hospitalizations usually lead to disruption of normal routine (eg, unpredictable meal times or medication administration), interrupted sleep (eg, woken up for blood draws or provider evaluation), and lack of control of schedule (eg, unsure of exact time when a procedure may be occurring), any of which may trigger symptoms of anxiety and anger in patients with PTSD and lead to hypervigilance.

If situations involving patient anxiety do arise, employ compassion and communication. Extra time spent with the patient, while challenging in the hectic hospital environment, is critical, and nonpharmacological treatments should be the priority. Engaging patients by asking about their PTSD triggers²⁴ may help prevent exacerbations. For example, some patients may specify how they prefer to be woken up to prevent startle reactions. PTSD triggers can be reduced via effective communication with the entire healthcare team. Some immediate yet effective strategies are listening, validation, and negotiation. Benzodiazepine or antipsychotic usage should be avoided.³⁶ Inpatient social work and comanagement with psychiatry involvement may be helpful in more severe exacerbations. A small observational study of patients hospitalized for severe PTSD found an association between walking more during hospitalization and fewer PTSD symptoms,⁵⁶ suggesting that staying active could be helpful for inpatients with PTSD who are able to safely ambulate.

PTSD is a common comorbidity among hospitalized patients in the United States. Typical hospital routines may exacerbate symptoms of PTSD such as anxiety and anger. Inpatient providers can play an important role in making hospitalizations go more smoothly for these patients by using principles consistent with trauma-informed care. Specifically, partnering with patients to construct a plan that preserves their sleep routines and accounts for potential triggers for decompensation can improve the hospital experience for patients with PTSD. Some PTSD interventions require additional investment from the healthcare system to deploy, such as staff training in trauma-­informed care and reflective listening techniques. Electronic health record–based protocols and order sets for patients with PTSD can leverage available resources. Further research should evaluate hospital outcomes that result from a more tailored approach to the care of patients with PTSD. More effective, patient-centered PTSD care could lower rates of leaving against medical advice and improve the inpatient experience for patients and providers alike.

Posttraumatic stress disorder (PTSD) is a syndrome that occurs after exposure to a significant traumatic event and is characterized by persistent, debilitating symptoms that fall into four “diagnostic clusters” as outlined in the Diagnostic and Statistical Manual of Mental Disorders-Version V (DSM-V). Patients may experience intrusive thoughts, avoidance of distressing stimuli, persistent negative mood, and hypervigilance, all of which last longer than 1 month.¹

A national survey of United States households conducted during 2001-2003 estimated the 12-month prevalence of PTSD among adults to be 3.5%.² Lifetime prevalence has been found to be between 6.8%³ and 7.8%.⁴ PTSD is more common in veterans. The prevalence of PTSD in veterans differs depending on the conflict in which the veteran participated. Vietnam veterans have an estimated lifetime prevalence of approximately 30%,^5,6 Gulf War veterans approximately 15%,⁷ and veterans of more recent conflicts in Afghanistan and Iraq of approximately 21%.⁸ With the MISSION Act moving more veteran care into the private sector, non-VA inpatient providers will need to become better versed in PTSD.⁹

Patients with PTSD have more contact with the healthcare system, even for non–mental health problems,^8,10-13 and a significantly higher burden of medical comorbities,¹⁴ such as diabetes mellitus, liver disease, gastritis and gastric ulcers, HIV, arthritis,¹⁵ and coronary heart disease.¹⁶ Veterans with PTSD are hospitalized three times more often than are those with no mental health diagnoses,⁸ and patients with psychiatric comorbidities have higher lengths of stay.¹⁷ More than 1.4 million hospitalizations occurring during 2002-2011 had either a primary or secondary associated diagnosis of PTSD, with total inflation-­adjusted charges of 34.9 billion dollars.¹⁸ In the inpatient sample from this study, greater than half were admitted for a primary diagnosis of mental diseases and disorders (Major Diagnostic Category [MDC] 19). Following mental illness, the most common primary diagnoses for men were MDC 5 (Circulatory System, 12.1%), MDC 20 (Alcohol/Drug Use or Induced Mental Disorder, 9.2%), and MDC 4 (Respiratory System, 7.4%), while the most common categories for women were MDC 20 (Alcohol/Drug Use or Induced Mental Disorder, 5.8%), MDC 21 (Injuries, Poison, and Toxic Effect of Drugs, 4.9%), and MDC 6 (Digestive System, 4.5%).¹⁸

In both the inpatient and outpatient settings, a fundamental challenge to comprehensive PTSD management is correctly diagnosing this condition.¹⁹ Confounding the difficulties in diagnosis are numerous comorbidities. In addition to the physical comorbidities described above, more than 70% of patients with PTSD have another psychological comorbidity such as affective disorders, anxiety disorders, or substance use disorder/dependency.²⁰

Given that PTSD may be an underrecognized burden on the healthcare system, we sought to better understand how PTSD could affect hospitalized patients admitted for medical problems by conducting this narrative review. Additionally, three of the authors collaborated with the VA Employee Education Service to conduct a needs assessment of VA hospitalists in 2013. Respondents identified managing and educating patients and families about PTSD as a major educational need (unpublished data available upon request from the corresponding author). Therefore, our aims were to present a synthesis of existing literature, familiarize readers with the tenets of trauma-­informed care as a framework to guide care for these patients, and generate ideas for changes that inpatient providers could implement now. We began by consulting a research librarian at the Clement J. Zablocki VA Medical Center in Milwaukee, Wisconsin, who searched the following databases: PsycInfo, CINAHL, MEDLINE, and PILOTS (a PTSD/trauma specific database). Search terms included hospital, hospitalized, and hospitalization, as well as traumatic stress, posttraumatic stress, and PTSD. Pertinent guidelines and the reference lists from included papers were examined. We focused on papers that described patients admitted for medical problems other than PTSD because those patients who are admitted for PTSD-­related problems should be primarily ma­naged by psychiatry (not hospitalists) with the primary focus of their hospitalization being their PTSD. We also excluded papers about patients developing PTSD secondary to hospitalization, which already has a well-developed literature.^21-23

The literature is sparse describing frequency or type of problems encountered by hospitalized medical patients with PTSD. A recent small survey study reported that 40% of patients anticipated triggers for their PTSD symptoms in the hospital; such triggers included loud noises and being shaken awake.²⁴ Two papers describe case vignettes of patients who had exacerbations of their PTSD while in the Intensive Care Unit (ICU), although neither contain frequency or severity data.^25,26 Approximately 8% of patients in VA ICUs have PTSD,²⁷ and a published abstract suggests that they appear to require more sedation than do patients without PTSD.²⁸ Another published case report describes a patient with recurrent PTSD symptoms (nightmares) after moving into a nursing home.²⁹ These papers suggest other providers have recognized and are concerned about hospitalized patients with PTSD. At present, there are no data to quantify how often hospitalized patients have PTSD exacerbations or how troublesome such exacerbations are to these patients.

Given that there is little empiric literature to guide inpatient management of PTSD as a comorbidity in hospitalized medical patients, we extrapolate some information from the outpatient setting. PTSD is often underdiagnosed and underreported by individual patients in the outpatient setting.³⁰ Failure to have an associated diagnosis of PTSD may lead to underrecognition and undertreatment of these patients by inpatient providers in the hospital setting. Additionally, the numerous psychological and physical comorbidities in PTSD can create unique challenges in properly managing any single problem in these patients.²⁰ Armed with this knowledge, providers should be vigilant in the recognition, assessment, and treatment of PTSD.

Trauma-Informed Care: A Conceptual Model

Trauma-informed care is a mindful and sensitive approach to caring for patients who have suffered trauma.³¹ It requires understanding that many people have suffered trauma in their lives and that the trauma continues to impact many aspects of their lives.³² Trauma-informed care has many advocates and has been implemented across myriad health and social services settings.³³ Its principles can be applied in both the inpatient and outpatient hospital settings. While it is an appropriate approach to patients with PTSD, it is not specific to PTSD. People who have suffered sexual trauma, intimate partner violence, child abuse, or other exposures would also be included in the group of people for whom trauma-informed care is a suitable approach. There are four key assumptions to a trauma-­informed approach to care (the 4 R’s): (1) realization that trauma affects an individual’s coping strategies, relationships, and health; (2) recognition of the signs of trauma; (3) having an appropriate, planned response to patients identified as having suffered a trauma; and (4) resisting retraumatization in the care setting.^31,32

General Approach to Treating Medical Patients With PTSD in the Inpatient Setting

Recognition

Consistent with a trauma-informed care approach, inpatient providers should be able to recognize patients who may have PTSD. First, careful review of the past medical history may show some patients already carry this diagnosis. Second, patients with PTSD often have other comorbidities that could offer a clue that PTSD could be present as well; for example, risk for PTSD is increased when mood, anxiety, or substance use disorders are present.²⁰ When PTSD is suspected, screening is a reasonable next step.

The Primary Care-PTSD-5 (PC-PTSD-5) is a validated screening tool used in the outpatient setting.³⁴ It is easily administered and has good predictive validity (positive likelihood ratio [LR+] of 6.33 and LR– of 0.06). It begins with a question of whether the patient has ever experienced a trauma. A positive initial response triggers a series of five yes/no questions. Answering “yes” to three or more questions is a positive screen. A positive screen should result in consultation to psychiatry to conduct more formal evaluation and guide longer-term management.

Collaboration

Individual trauma-focused psychotherapy is the primary treatment of choice for PTSD with strong evidence supporting its practice.³⁵ This treatment is administered by a psychiatrist or psychologist and will be limited in the inpatient medical setting. Current recommendations suggest pharmacotherapy only when individualized trauma-focused psychotherapy is not available, the patient declines it, or as an adjunct when psychotherapy alone is not effective.³⁶ Therefore, inpatient providers may see patients who are prescribed selective serotonin reuptake inhibitors (eg, paroxetine, fluoxetine) or serotonin and norepinephrine reuptake inhibitors (eg, venlafaxine).³⁶ In the past, PTSD-related nightmares were often treated with prazosin.³⁷ However, a recent randomized controlled trial of prazosin in veterans with PTSD failed to show significant improvement in nightmares.³⁸ Hence, current guidelines do not recommend prazosin as a first-line therapy.³⁹ For hospitalized patients with PTSD symptoms refractory to the interventions outlined herein, particularly those patients with possible borderline personality traits (as suggested by severe anger and impulsivity), we strongly recommend partnering with psychiatry. Finally, given the high prevalence of substance use disorders (SUDs) in PTSD patients, awareness and treatment of comorbidities such as opioid and alcohol dependence must be concurrently addressed.

Individualizing Care

It is essential for the healthcare team to identify ways to meet each patient’s immediate needs. Many of the ideas proposed below are not specific to PTSD; many require an interprofessional approach to care.⁴⁰ From a trauma-informed care standpoint, this is akin to having a planned response for patients who have suffered trauma. Assessing the individual’s needs and incorporating therapeutic modalities such as reflective listening, broadening safe opportunities for control, and providing complementary and integrative medicine (IM) therapies may help manage symptoms and establish rapport.⁴¹ Through reflective listening, a collaborative approach can be established to identify background, triggers, and a safe approach for managing PTSD and its comorbid conditions. Ensuring frequent communication and allowing the patient to be at the center of decision-making establishes a safe environment and promotes positive rapport between the patient and healthcare team.³⁶ Providing a sense of control by involving the patients in their healthcare decisions and in the structure of care delivery may benefit the patients’ well-being. Furthermore, incorporating IM encourages rest and relaxation in the chaotic hospital environment. Suggested IM interventions include deep breathing, aromatherapy, guided imagery, muscle relaxation, and music therapy.^42,43

Key Inpatient Issues Affecting PTSD

In the following sections, we outline common clinical situations that may exacerbate PTSD symptoms and propose some evidence-based responses (Table). In general, nonpharmacologic approaches are favored over pharmacologic approaches for patients with PTSD.

Sleep Hygiene

Sleep problems are very common in patients with PTSD, with nightmares occurring in more than 70% of patients and insomnia in 80%.⁴⁴ In PTSD, sleep problems are linked to poor physical health and other health outcomes^45,46 and may exacerbate other PTSD symptoms.⁴

Treating the sleep problems that occur with PTSD is an important aspect of PTSD care. Usually administered in the outpatient setting, the treatment of choice is cognitive-behavioral therapy (CBT).⁴⁸ Sleep-specific CBT focuses, among other things, on strategies that encourage good sleep hygiene,⁴⁹ which includes promoting regular sleep/wake-up times and specific bedtime routines, avoiding stimulation (eg, light, noise, TV) or excessive liquids before bed, refraining from daytime naps, and using relaxation techniques. Many of these recommendations seem at odds with hospital routines, which may contribute to decompensation of hospitalized patients with PTSD.

While starting sleep-specific CBT in the hospital may not be realistic, we suggest the following goals and strategies as a starting place for promoting healthy sleep for hospitalized patients with PTSD. To begin, factors affecting sleep hygiene should be addressed. Inpatient providers could pay more attention to intravenous (IV) fluid orders, perhaps adjusting them to run only during the daytime hours. Medications can be scheduled at times conducive to maintaining home routines. Avoiding the administration of diuretics close to bedtime may decrease the likelihood of frequent nighttime wakening. Grouping patient care activities, such as bathing or wound care, during daytime hours may allow more opportunities for rest at night. Incorporating uninterrupted sleep protocols, such as quiet hours between 10pm and 6am, may enhance sleep quality in the appropriately selected patient.⁵⁰ Although pharmacological interventions to improve sleep in the hospital may be initially beneficial, nonpharmacological interventions as described above should be incorporated for long-term maintenance of enhanced sleep quality.

Second, providers need to ask about established home bedtime routines and facilitate implementation in the hospital. Through collaboration with patients, providers can incorporate an individualized plan of care for sleep early in hospitalization.⁵⁰ Partnering with nurses is also essential to creating a sleep-friendly environment that can improve patient experiences.⁵¹ Breathing exercises, meditating, listening to music and praying are all examples of “bedtime wind down” strategies recommended in sleep-specific CBT.⁴⁹ Many of these could be successfully implemented in the hospital and may benefit other hospitalized patients too.⁵² In patients with PTSD and obstructive sleep apnea, continuous positive airway pressure (CPAP) reduces nightmares, and if inpatients are on CPAP at home, it should be continued in the hospital.⁵³

Pain

If sleep disturbance is the hallmark of PTSD,⁴⁷ chronic pain is its coconspirator.¹⁵ Uncontrolled pain can make it much more difficult to treat patients with PTSD, which in turn may lead to further decompensation from a mental health standpoint.⁵⁴ SUDs such as alcohol or opioid dependencies are highly comorbid with PTSD⁴⁵ and introduce a layer of complexity when managing painin these patients. Providers should be thoughtful when electing to treat acute or chronic pain with opioids and take particular care to establish realistic therapeutic goals if doing so. While patients with PTSD have a greater likelihood of having an SUD, undertreating pain risks exacerbating underlying PTSD symptoms.

Nonpharmacologic therapies, which include communicating, listening, and expressing compassion and understanding, should be utilized by inpatient providers as a first-line treatment in patients with PTSD who suffer from pain. Additionally, relaxation techniques, physical therapy, and physical activity⁵⁵ can be offered. Pharmacologically, nonopioid medications such as acetaminophen or NSAIDs should always be considered first. Should the use of opioids be deemed necessary, inpatient providers should preferentially use oral over intravenous medications and consider establishing a fixed timeframe for short-term opioids, which should be limited to a few days. Providers should communicate clear expectations with their patients to maximize the desired effect of any specific treatment while minimizing the risk of medication side effects with the goal of agreeing on a short yet effective treatment course.

Anxiety and Anger

One of the most challenging situations for the inpatient provider is encountering a patient who is anxious, angry, or hypervigilant. Mismatch between actual and expected communication between the provider and the patient can lead to frustration and anxiety. A trauma-informed care approach would suggest that frequent and thorough communication with patients may prevent or ameliorate the stresses and anxieties of hospitalization that may manifest as anger because of retraumatization. Hospitalizations usually lead to disruption of normal routine (eg, unpredictable meal times or medication administration), interrupted sleep (eg, woken up for blood draws or provider evaluation), and lack of control of schedule (eg, unsure of exact time when a procedure may be occurring), any of which may trigger symptoms of anxiety and anger in patients with PTSD and lead to hypervigilance.

If situations involving patient anxiety do arise, employ compassion and communication. Extra time spent with the patient, while challenging in the hectic hospital environment, is critical, and nonpharmacological treatments should be the priority. Engaging patients by asking about their PTSD triggers²⁴ may help prevent exacerbations. For example, some patients may specify how they prefer to be woken up to prevent startle reactions. PTSD triggers can be reduced via effective communication with the entire healthcare team. Some immediate yet effective strategies are listening, validation, and negotiation. Benzodiazepine or antipsychotic usage should be avoided.³⁶ Inpatient social work and comanagement with psychiatry involvement may be helpful in more severe exacerbations. A small observational study of patients hospitalized for severe PTSD found an association between walking more during hospitalization and fewer PTSD symptoms,⁵⁶ suggesting that staying active could be helpful for inpatients with PTSD who are able to safely ambulate.

PTSD is a common comorbidity among hospitalized patients in the United States. Typical hospital routines may exacerbate symptoms of PTSD such as anxiety and anger. Inpatient providers can play an important role in making hospitalizations go more smoothly for these patients by using principles consistent with trauma-informed care. Specifically, partnering with patients to construct a plan that preserves their sleep routines and accounts for potential triggers for decompensation can improve the hospital experience for patients with PTSD. Some PTSD interventions require additional investment from the healthcare system to deploy, such as staff training in trauma-­informed care and reflective listening techniques. Electronic health record–based protocols and order sets for patients with PTSD can leverage available resources. Further research should evaluate hospital outcomes that result from a more tailored approach to the care of patients with PTSD. More effective, patient-centered PTSD care could lower rates of leaving against medical advice and improve the inpatient experience for patients and providers alike.

Posttraumatic stress disorder (PTSD) is a syndrome that occurs after exposure to a significant traumatic event and is characterized by persistent, debilitating symptoms that fall into four “diagnostic clusters” as outlined in the Diagnostic and Statistical Manual of Mental Disorders-Version V (DSM-V). Patients may experience intrusive thoughts, avoidance of distressing stimuli, persistent negative mood, and hypervigilance, all of which last longer than 1 month.¹

A national survey of United States households conducted during 2001-2003 estimated the 12-month prevalence of PTSD among adults to be 3.5%.² Lifetime prevalence has been found to be between 6.8%³ and 7.8%.⁴ PTSD is more common in veterans. The prevalence of PTSD in veterans differs depending on the conflict in which the veteran participated. Vietnam veterans have an estimated lifetime prevalence of approximately 30%,^5,6 Gulf War veterans approximately 15%,⁷ and veterans of more recent conflicts in Afghanistan and Iraq of approximately 21%.⁸ With the MISSION Act moving more veteran care into the private sector, non-VA inpatient providers will need to become better versed in PTSD.⁹

Patients with PTSD have more contact with the healthcare system, even for non–mental health problems,^8,10-13 and a significantly higher burden of medical comorbities,¹⁴ such as diabetes mellitus, liver disease, gastritis and gastric ulcers, HIV, arthritis,¹⁵ and coronary heart disease.¹⁶ Veterans with PTSD are hospitalized three times more often than are those with no mental health diagnoses,⁸ and patients with psychiatric comorbidities have higher lengths of stay.¹⁷ More than 1.4 million hospitalizations occurring during 2002-2011 had either a primary or secondary associated diagnosis of PTSD, with total inflation-­adjusted charges of 34.9 billion dollars.¹⁸ In the inpatient sample from this study, greater than half were admitted for a primary diagnosis of mental diseases and disorders (Major Diagnostic Category [MDC] 19). Following mental illness, the most common primary diagnoses for men were MDC 5 (Circulatory System, 12.1%), MDC 20 (Alcohol/Drug Use or Induced Mental Disorder, 9.2%), and MDC 4 (Respiratory System, 7.4%), while the most common categories for women were MDC 20 (Alcohol/Drug Use or Induced Mental Disorder, 5.8%), MDC 21 (Injuries, Poison, and Toxic Effect of Drugs, 4.9%), and MDC 6 (Digestive System, 4.5%).¹⁸

In both the inpatient and outpatient settings, a fundamental challenge to comprehensive PTSD management is correctly diagnosing this condition.¹⁹ Confounding the difficulties in diagnosis are numerous comorbidities. In addition to the physical comorbidities described above, more than 70% of patients with PTSD have another psychological comorbidity such as affective disorders, anxiety disorders, or substance use disorder/dependency.²⁰

Given that PTSD may be an underrecognized burden on the healthcare system, we sought to better understand how PTSD could affect hospitalized patients admitted for medical problems by conducting this narrative review. Additionally, three of the authors collaborated with the VA Employee Education Service to conduct a needs assessment of VA hospitalists in 2013. Respondents identified managing and educating patients and families about PTSD as a major educational need (unpublished data available upon request from the corresponding author). Therefore, our aims were to present a synthesis of existing literature, familiarize readers with the tenets of trauma-­informed care as a framework to guide care for these patients, and generate ideas for changes that inpatient providers could implement now. We began by consulting a research librarian at the Clement J. Zablocki VA Medical Center in Milwaukee, Wisconsin, who searched the following databases: PsycInfo, CINAHL, MEDLINE, and PILOTS (a PTSD/trauma specific database). Search terms included hospital, hospitalized, and hospitalization, as well as traumatic stress, posttraumatic stress, and PTSD. Pertinent guidelines and the reference lists from included papers were examined. We focused on papers that described patients admitted for medical problems other than PTSD because those patients who are admitted for PTSD-­related problems should be primarily ma­naged by psychiatry (not hospitalists) with the primary focus of their hospitalization being their PTSD. We also excluded papers about patients developing PTSD secondary to hospitalization, which already has a well-developed literature.^21-23

The literature is sparse describing frequency or type of problems encountered by hospitalized medical patients with PTSD. A recent small survey study reported that 40% of patients anticipated triggers for their PTSD symptoms in the hospital; such triggers included loud noises and being shaken awake.²⁴ Two papers describe case vignettes of patients who had exacerbations of their PTSD while in the Intensive Care Unit (ICU), although neither contain frequency or severity data.^25,26 Approximately 8% of patients in VA ICUs have PTSD,²⁷ and a published abstract suggests that they appear to require more sedation than do patients without PTSD.²⁸ Another published case report describes a patient with recurrent PTSD symptoms (nightmares) after moving into a nursing home.²⁹ These papers suggest other providers have recognized and are concerned about hospitalized patients with PTSD. At present, there are no data to quantify how often hospitalized patients have PTSD exacerbations or how troublesome such exacerbations are to these patients.

Given that there is little empiric literature to guide inpatient management of PTSD as a comorbidity in hospitalized medical patients, we extrapolate some information from the outpatient setting. PTSD is often underdiagnosed and underreported by individual patients in the outpatient setting.³⁰ Failure to have an associated diagnosis of PTSD may lead to underrecognition and undertreatment of these patients by inpatient providers in the hospital setting. Additionally, the numerous psychological and physical comorbidities in PTSD can create unique challenges in properly managing any single problem in these patients.²⁰ Armed with this knowledge, providers should be vigilant in the recognition, assessment, and treatment of PTSD.

Trauma-Informed Care: A Conceptual Model

Trauma-informed care is a mindful and sensitive approach to caring for patients who have suffered trauma.³¹ It requires understanding that many people have suffered trauma in their lives and that the trauma continues to impact many aspects of their lives.³² Trauma-informed care has many advocates and has been implemented across myriad health and social services settings.³³ Its principles can be applied in both the inpatient and outpatient hospital settings. While it is an appropriate approach to patients with PTSD, it is not specific to PTSD. People who have suffered sexual trauma, intimate partner violence, child abuse, or other exposures would also be included in the group of people for whom trauma-informed care is a suitable approach. There are four key assumptions to a trauma-­informed approach to care (the 4 R’s): (1) realization that trauma affects an individual’s coping strategies, relationships, and health; (2) recognition of the signs of trauma; (3) having an appropriate, planned response to patients identified as having suffered a trauma; and (4) resisting retraumatization in the care setting.^31,32

General Approach to Treating Medical Patients With PTSD in the Inpatient Setting

Recognition

Consistent with a trauma-informed care approach, inpatient providers should be able to recognize patients who may have PTSD. First, careful review of the past medical history may show some patients already carry this diagnosis. Second, patients with PTSD often have other comorbidities that could offer a clue that PTSD could be present as well; for example, risk for PTSD is increased when mood, anxiety, or substance use disorders are present.²⁰ When PTSD is suspected, screening is a reasonable next step.

The Primary Care-PTSD-5 (PC-PTSD-5) is a validated screening tool used in the outpatient setting.³⁴ It is easily administered and has good predictive validity (positive likelihood ratio [LR+] of 6.33 and LR– of 0.06). It begins with a question of whether the patient has ever experienced a trauma. A positive initial response triggers a series of five yes/no questions. Answering “yes” to three or more questions is a positive screen. A positive screen should result in consultation to psychiatry to conduct more formal evaluation and guide longer-term management.

Collaboration

Individual trauma-focused psychotherapy is the primary treatment of choice for PTSD with strong evidence supporting its practice.³⁵ This treatment is administered by a psychiatrist or psychologist and will be limited in the inpatient medical setting. Current recommendations suggest pharmacotherapy only when individualized trauma-focused psychotherapy is not available, the patient declines it, or as an adjunct when psychotherapy alone is not effective.³⁶ Therefore, inpatient providers may see patients who are prescribed selective serotonin reuptake inhibitors (eg, paroxetine, fluoxetine) or serotonin and norepinephrine reuptake inhibitors (eg, venlafaxine).³⁶ In the past, PTSD-related nightmares were often treated with prazosin.³⁷ However, a recent randomized controlled trial of prazosin in veterans with PTSD failed to show significant improvement in nightmares.³⁸ Hence, current guidelines do not recommend prazosin as a first-line therapy.³⁹ For hospitalized patients with PTSD symptoms refractory to the interventions outlined herein, particularly those patients with possible borderline personality traits (as suggested by severe anger and impulsivity), we strongly recommend partnering with psychiatry. Finally, given the high prevalence of substance use disorders (SUDs) in PTSD patients, awareness and treatment of comorbidities such as opioid and alcohol dependence must be concurrently addressed.

Individualizing Care

It is essential for the healthcare team to identify ways to meet each patient’s immediate needs. Many of the ideas proposed below are not specific to PTSD; many require an interprofessional approach to care.⁴⁰ From a trauma-informed care standpoint, this is akin to having a planned response for patients who have suffered trauma. Assessing the individual’s needs and incorporating therapeutic modalities such as reflective listening, broadening safe opportunities for control, and providing complementary and integrative medicine (IM) therapies may help manage symptoms and establish rapport.⁴¹ Through reflective listening, a collaborative approach can be established to identify background, triggers, and a safe approach for managing PTSD and its comorbid conditions. Ensuring frequent communication and allowing the patient to be at the center of decision-making establishes a safe environment and promotes positive rapport between the patient and healthcare team.³⁶ Providing a sense of control by involving the patients in their healthcare decisions and in the structure of care delivery may benefit the patients’ well-being. Furthermore, incorporating IM encourages rest and relaxation in the chaotic hospital environment. Suggested IM interventions include deep breathing, aromatherapy, guided imagery, muscle relaxation, and music therapy.^42,43

Key Inpatient Issues Affecting PTSD

In the following sections, we outline common clinical situations that may exacerbate PTSD symptoms and propose some evidence-based responses (Table). In general, nonpharmacologic approaches are favored over pharmacologic approaches for patients with PTSD.

Sleep Hygiene

Sleep problems are very common in patients with PTSD, with nightmares occurring in more than 70% of patients and insomnia in 80%.⁴⁴ In PTSD, sleep problems are linked to poor physical health and other health outcomes^45,46 and may exacerbate other PTSD symptoms.⁴

Treating the sleep problems that occur with PTSD is an important aspect of PTSD care. Usually administered in the outpatient setting, the treatment of choice is cognitive-behavioral therapy (CBT).⁴⁸ Sleep-specific CBT focuses, among other things, on strategies that encourage good sleep hygiene,⁴⁹ which includes promoting regular sleep/wake-up times and specific bedtime routines, avoiding stimulation (eg, light, noise, TV) or excessive liquids before bed, refraining from daytime naps, and using relaxation techniques. Many of these recommendations seem at odds with hospital routines, which may contribute to decompensation of hospitalized patients with PTSD.

While starting sleep-specific CBT in the hospital may not be realistic, we suggest the following goals and strategies as a starting place for promoting healthy sleep for hospitalized patients with PTSD. To begin, factors affecting sleep hygiene should be addressed. Inpatient providers could pay more attention to intravenous (IV) fluid orders, perhaps adjusting them to run only during the daytime hours. Medications can be scheduled at times conducive to maintaining home routines. Avoiding the administration of diuretics close to bedtime may decrease the likelihood of frequent nighttime wakening. Grouping patient care activities, such as bathing or wound care, during daytime hours may allow more opportunities for rest at night. Incorporating uninterrupted sleep protocols, such as quiet hours between 10pm and 6am, may enhance sleep quality in the appropriately selected patient.⁵⁰ Although pharmacological interventions to improve sleep in the hospital may be initially beneficial, nonpharmacological interventions as described above should be incorporated for long-term maintenance of enhanced sleep quality.

Second, providers need to ask about established home bedtime routines and facilitate implementation in the hospital. Through collaboration with patients, providers can incorporate an individualized plan of care for sleep early in hospitalization.⁵⁰ Partnering with nurses is also essential to creating a sleep-friendly environment that can improve patient experiences.⁵¹ Breathing exercises, meditating, listening to music and praying are all examples of “bedtime wind down” strategies recommended in sleep-specific CBT.⁴⁹ Many of these could be successfully implemented in the hospital and may benefit other hospitalized patients too.⁵² In patients with PTSD and obstructive sleep apnea, continuous positive airway pressure (CPAP) reduces nightmares, and if inpatients are on CPAP at home, it should be continued in the hospital.⁵³

Pain

If sleep disturbance is the hallmark of PTSD,⁴⁷ chronic pain is its coconspirator.¹⁵ Uncontrolled pain can make it much more difficult to treat patients with PTSD, which in turn may lead to further decompensation from a mental health standpoint.⁵⁴ SUDs such as alcohol or opioid dependencies are highly comorbid with PTSD⁴⁵ and introduce a layer of complexity when managing painin these patients. Providers should be thoughtful when electing to treat acute or chronic pain with opioids and take particular care to establish realistic therapeutic goals if doing so. While patients with PTSD have a greater likelihood of having an SUD, undertreating pain risks exacerbating underlying PTSD symptoms.

Nonpharmacologic therapies, which include communicating, listening, and expressing compassion and understanding, should be utilized by inpatient providers as a first-line treatment in patients with PTSD who suffer from pain. Additionally, relaxation techniques, physical therapy, and physical activity⁵⁵ can be offered. Pharmacologically, nonopioid medications such as acetaminophen or NSAIDs should always be considered first. Should the use of opioids be deemed necessary, inpatient providers should preferentially use oral over intravenous medications and consider establishing a fixed timeframe for short-term opioids, which should be limited to a few days. Providers should communicate clear expectations with their patients to maximize the desired effect of any specific treatment while minimizing the risk of medication side effects with the goal of agreeing on a short yet effective treatment course.

Anxiety and Anger

One of the most challenging situations for the inpatient provider is encountering a patient who is anxious, angry, or hypervigilant. Mismatch between actual and expected communication between the provider and the patient can lead to frustration and anxiety. A trauma-informed care approach would suggest that frequent and thorough communication with patients may prevent or ameliorate the stresses and anxieties of hospitalization that may manifest as anger because of retraumatization. Hospitalizations usually lead to disruption of normal routine (eg, unpredictable meal times or medication administration), interrupted sleep (eg, woken up for blood draws or provider evaluation), and lack of control of schedule (eg, unsure of exact time when a procedure may be occurring), any of which may trigger symptoms of anxiety and anger in patients with PTSD and lead to hypervigilance.

If situations involving patient anxiety do arise, employ compassion and communication. Extra time spent with the patient, while challenging in the hectic hospital environment, is critical, and nonpharmacological treatments should be the priority. Engaging patients by asking about their PTSD triggers²⁴ may help prevent exacerbations. For example, some patients may specify how they prefer to be woken up to prevent startle reactions. PTSD triggers can be reduced via effective communication with the entire healthcare team. Some immediate yet effective strategies are listening, validation, and negotiation. Benzodiazepine or antipsychotic usage should be avoided.³⁶ Inpatient social work and comanagement with psychiatry involvement may be helpful in more severe exacerbations. A small observational study of patients hospitalized for severe PTSD found an association between walking more during hospitalization and fewer PTSD symptoms,⁵⁶ suggesting that staying active could be helpful for inpatients with PTSD who are able to safely ambulate.

PTSD is a common comorbidity among hospitalized patients in the United States. Typical hospital routines may exacerbate symptoms of PTSD such as anxiety and anger. Inpatient providers can play an important role in making hospitalizations go more smoothly for these patients by using principles consistent with trauma-informed care. Specifically, partnering with patients to construct a plan that preserves their sleep routines and accounts for potential triggers for decompensation can improve the hospital experience for patients with PTSD. Some PTSD interventions require additional investment from the healthcare system to deploy, such as staff training in trauma-­informed care and reflective listening techniques. Electronic health record–based protocols and order sets for patients with PTSD can leverage available resources. Further research should evaluate hospital outcomes that result from a more tailored approach to the care of patients with PTSD. More effective, patient-centered PTSD care could lower rates of leaving against medical advice and improve the inpatient experience for patients and providers alike.

A previously healthy 4-year-old boy presented to his pediatrician for nasal congestion, left ear pain, and intermittent fevers, which he’d been experiencing for 2 days. His exam was consistent with acute otitis media. Cefdinir was prescribed given a rash allergy to amoxicillin. His fever, congestion, and otalgia improved the next day.

Three days later he developed abdominal pain, fever, and labored breathing; his mother brought him to the emergency department (ED). His temperature was 38.0 °C, heart rate 141 beats per minute, blood pressure 117/71 mm Hg, respiratory rate 22 breaths per minute; he had oxygen saturation of 96% on ambient air. Despite mild accessory muscle use, he appeared comfortable and interactive. His left tympanic membrane was bulging without erythema. His neck was supple and mucous membranes moist. He had neither cervical lymphadenopathy nor conjunctival pallor. The cardiopulmonary exam was normal except for tachycardia. His abdomen was soft and not distended without organomegaly or tenderness.

Upper respiratory tract symptoms are commonly encountered in pediatrics and most often result from self-limited viral processes. Evaluation of a child with upper respiratory tract symptoms aims to identify serious causes like meningitis, as well as assessing the need for antimicrobial therapy. Supportive management is often appropriate in otitis media. His new, more concerning symptoms portend either a progression of the original process causing his upper respiratory tract symptoms or a separate etiology. It is key to determine which signs and symptoms are associated with the primary process and which are compensatory or secondary. If he were to be more ill appearing, for example, it is possible that his respiratory distress may be related to an underlying systemic illness rather than a primary lung process. Respiratory distress, abdominal pain, and fever could be a result of sepsis from an intrabdominal process such as ruptured appendicitis, intussusception, or malrotation with volvulus. Other causes of sepsis, such as meningitis or severe mastoiditis, both rare complications of otitis media, should be considered, although he does not appear severely ill. Acute myelogenous leukemia or other malignancies and illnesses associated with immunodeficiency can present with sepsis and chloromas in the middle ear that can be misconstrued as otitis media.

A chest radiograph demonstrated left lower lobe patchy opacities concerning for pneumonia. Rapid respiratory syncytial virus and influenza antigen test results were negative. Laboratory testing for general bloodwork was not obtained. He was administered a single dose of intramuscular ceftriaxone, prescribed a 5-day course of azithromycin, and discharged home. The child’s breathing gradually improved, but he continued to have subjective fevers. Two days later, he developed dark red urine. His mother brought him back to the outpatient clinic.

At the time of the ED visit, a diagnosis of community-acquired pneumonia was plausible given fever, mildly increased work of breathing, and an opacification on chest radiography. Most community-acquired pneumonia is caused by viruses; common bacterial causes for his age include Streptococcus pneumoniae and Moraxella catarrhalis. The first-line treatment for uncomplicated community-acquired pneumonia in children is amoxicillin, but this was appropriately avoided given his allergy.

The persistent fevers are surprising. The improvement in breathing corresponds to the treatment (and resolution) of community-acquired pneumonia. However, the development of dark urine does not. Red urine—in the absence of ingested pigments (such as those found in beets)—usually results from hematuria, hemoglobinuria, or myoglobinuria. Gross hematuria can originate from the kidneys to the urethral meatus. Abdominal masses, kidney trauma, or underlying kidney disease may all present with gross hematuria (or microscopic hematuria, seen only on urinalysis). The urine should be examined for the presence of heme, protein, and for evidence of infection; microscopy should be performed to examine for cellular casts and dysmorphic red cells. Tests of renal function, a comprehensive metabolic panel, evaluation of hematologic indexes, and assessments of inflammatory markers should be performed.

The child lived with his parents and had no siblings. He experienced no physical trauma, and there was no family history of kidney disease or hematuria. His father had a persistent cough and fever for 1 month, but recovered around the time the patient began to experience his initial symptoms. This was the patient’s third diagnosis of pneumonia. He had not traveled and was up to date with immunizations. He attended day care.

The fact that this is not the first episode of “pneumonia” raises important possibilities. The most likely one is that the child has had multiple viral infections; however, he could have an underlying primary immunodeficiency (PI) that predisposes him to recurrent infections. More severe PIs often present with recurrent sepsis, bacteremia, and failure to thrive, none of which were present in this case. Less severe PIs (such as selective IgA deficiency) could be possible. Another possibility is that these recurrent episodes of pneumonia are a relapsing and remitting noninfectious process, such as an antineutrophil cytoplasmic antibodies–associated vasculitis or anti–glomerular basement membrane disease. The patient’s father’s recent prolonged respiratory symptoms may be suggestive of pertussis or a “walking pneumonia” potentially caused by Mycoplasma or another atypical bacterium.

His temperature was 36.9 °C, heart rate 107 beats per minute, blood pressure was 106/67 mm Hg, and respiratory rate was 24 breaths per minute with oxygen saturation of 100% on ambient air. He was well appearing. His mucous membranes were moist, and oropharynx was clear. He had scleral icterus. The cardiopulmonary exam was normal. He had no significant lymphadenopathy, hepatosplenomegaly, or rashes.

The finding of jaundice is an important diagnostic pivot point, especially when combined with hematuria. The next step is determining if the jaundice is resulting from unconjugated or conjugated hyperbilirubinemia; the former most often stems from hemolysis or impairment in conjugation, while the latter results from intrahepatic or extrahepatic biliary defects. Tests for hepatobiliary injury including evaluations of alanine and aspartate aminotransferases and alkaline phosphatase, as well as for hepatic function such as tests of coagulation, should be performed.

The patient was referred to the ED and admitted for further evaluation. A complete blood count revealed a white blood cell (WBC) count of 10,700/µL (61% polymorphonuclear neutrophils, 30% lymphocytes, 5% monocytes, 3% eosinophils, 1% basophils), hemoglobin count was 10.3 g/dL (reticulocyte 2% with absolute reticulocyte count 58,400/μL), and platelet count was 265,000/µL. Components of the basic metabolic panel were within reference ranges except for a mildly elevated blood urea nitrogen level of 14 mg/dL with normal creatinine level of 0.3 mg/dL. Total protein was 6.7 g/dL (reference range, 6.4-8.3) and albumin 3.9 g/dL (reference range, 3.4-4.8). Alkaline phosphatase level was 188 U/L (reference range, 44-147), aspartate aminotransferase level 76 U/L (reference range, 0-40), and alanine aminotransferase level 12 U/L (reference range, 7-40). Total bilirubin level was 2.4 mg/dL (reference range, less than 1.5) with direct bilirubin level of 0.4 mg/dL. His C-reactive protein level was 1.5 mg/mL (reference range, 0-0.75). Creatinine kinase (CK) level was 2,550 U/L (reference range, 2-198). International Normalized Ratio (INR) was 1.0. Urinalysis was notable for 2+ proteinuria, large hemoglobin pigment, and 6 red blood cells per high power field (reference range, 0-4).

His blood urea nitrogen is elevated, reflecting either prerenal azotemia or increased absorption of nitrogenous products. Unconjugated hyperbilirubinemia may result from impaired hepatic bilirubin uptake (such as in heart failure or portosystemic shunts), impaired bilirubin conjugation (resulting from genetic conditions or drugs), or excess bilirubin production (such as in hemolysis); his anemia and lack of other evidence of hepatic dysfunction point to hemolysis as the etiology. The reticulocyte production index is approximately 1%, which suggests that an increase in erythrocyte generation is present but inadequate. This, however, does not mean that an erythrocyte production abnormality is present since reticulocytosis can be delayed in many cases of acute hemolytic anemia. It is also possible that the same hemolytic process is affecting mature and immature erythrocytes. A peripheral blood smear should be reviewed for evidence of intravascular hemolysis and testing for autoimmune hemolysis should be performed. Notably, his white blood cell and platelet counts are preserved, which makes a bone marrow–involved malignancy or infiltrative process less likely. The alkaline phosphatase elevation may result from either intrahepatic or extrahepatic biliopathy; bone damage is also possible. The elevation of aspartate aminotransferase, CK, and potassium, along with marked urinary heme pigment, may indicate muscle damage; the most common myositis in children is benign acute childhood myositis resulting from viral infection. However, the moderate level of CK elevation seen in this case is nonspecific and can result from many different etiologies. A metabolic myopathy, such as carnitine palmitoyltransferase II deficiency, can be made worse by metabolic stress and result in rhabdomyolysis; the presentations of inborn errors of metabolism are varied and a planned-out, stepwise approach in evaluation is fundamental.

Lactic acid dehydrogenase (LDH) level was 1,457 U/L (reference range, 140-280), and haptoglobin level was less than 6 mg/dL (reference range, 30-200). Peripheral blood smear demonstrated occasional atypical, reactive-appearing lymphocytes with red cell clumping and agglutination, as well as rare target, burr, and fragmented red cells. Test results for urine myoglobin were negative. Results of urine culture were negative. No blood culture was collected.

The elevated LDH, decreased haptoglobin, and findings on the peripheral blood smear confirm hemolysis. The clumping of erythrocytes can be artifactual in the preparation of peripheral smears, but when considered in the context of hemolysis, may be clinically important. Clumping of erythrocytes on the peripheral smear indicates the binding of a protein to antigens on the erythrocyte membrane; when this occurs below body temperature, this is consistent with the presence of a “cold agglutinin,” usually an IgM antibody directed at erythrocyte surface antigens that causes agglutination and destruction, especially in cooler areas of the body. This is a well-known complication of Mycoplasma pneumoniae infections as well as Epstein-Barr virus (EBV) infections; it may also occur with lymphoid malignancies or autoimmune disease.

Direct Coombs IgG test findings were negative, direct Coombs C3 test was positive, and direct Coombs polyspecific test was positive. M pneumoniae IgG antibody level was 1.4 mg/dL (reference ranges: <0.9, negative; 0.91-1.09, equivocal; >1.1, positive); M pneumoniae IgM level was 529 U/mL (reference range: <770, negative). EBV capsid IgM and IgG levels were undetectable. EBV nuclear antigen IgG level was also undetectable. EBV viral load was fewer than 10 copies/mL. Antinuclear antibodies (ANA) level was negative. General IgE and IgM levels were normal, at 11 and 81 mg/dL, respectively. Repeat complete blood count showed WBC of 7,800/µL, hemoglobin of 8.7 g/dL, and platelet count of 341,000/µL. The patient’s hemoglobin remained stable during hospitalization.

This directed testing is helpful in further classifying the patient’s hemolytic anemia. Autoimmune hemolytic anemias are classified into warm antibody–mediated, cold antibody–mediated, and mixed-type forms; drug-induced and alloimmune hemolytic anemias also occur. In addition, both systemic lupus erythematosus and antiphospholipid antibody syndrome can have hemolytic anemia with variable Coombs testing results; neither fit well in this case. The absence of red blood cell–directed IgG antibodies substantially decreases the likelihood of warm antibody–mediated hemolytic anemia. In cold antibody–mediated hemolytic anemia, antibodies bind to the erythrocyte membrane and then adhere to complement C3, which leads to both intravascular and extravascular hemolysis. Important types of cold antibody–mediated hemolytic anemia in children are primary and secondary cold agglutinin disease, along with paroxysmal cold hemoglobinuria. The Donath-Landsteiner test can be helpful in differentiating these conditions. Antibodies to Mycoplasma may be delayed in response to acute infection, and a child who is reinfected may only produce IgG antibodies. Given the patient’s clinical stability and previous health, the most likely diagnosis is Mycoplasma-induced cold antibody–mediated hemolytic anemia. It may be helpful to check convalescent titers to Mycoplasma in 2 to 4 weeks.

Donath-Landsteiner (D-L) antibody test results were positive. Medication-derived hemolytic anemia testing was conducted, but the presence of positive D-L antibody makes the test results inconclusive. This ultimately led to a diagnosis of paroxysmal cold hemoglobinuria (PCH), presumably triggered by a viral syndrome. Convalescent titers to Mycoplasma were not checked given clinical improvement. Because the patient’s hemoglobin was stable during hospitalization, he was not treated with steroids. His parents were counseled on avoiding cold temperatures for several days. Within 1 month, his hemoglobin had recovered without further evidence of hemolysis.

Hemolytic anemia refers to the accelerated destruction of red blood cells and can be further classified as acquired or hereditary.¹ Hereditary conditions causing hemolytic anemia include enzymopathies (eg, glucose-6-phosphate dehydrogenase deficiency), hemoglobinopathies (eg, sickle cell disease), and membrane abnormalities (eg, hereditary spherocytosis). Acquired pathologies include microangiopathic hemolytic anemia (MAHA), anemias directly caused by certain infections such as malaria, and immune-mediated (Coombs-positive) hemolytic anemias.

MAHA can sometimes be life-threatening and is therefore important to identify quickly. In the right clinical context, such processes may be rapidly recognized by the presence of schistocytes on blood smear in addition to an elevated serum LDH level. Schistocytes suggest mechanical destruction of erythrocytes in the vasculature, the hallmark of MAHA. Important MAHAs include thrombocytopenic purpura, hemolytic-uremic syndrome, and disseminated intravascular coagulation. Though this patient did have a mildly elevated LDH, MAHA was less likely because there were no schistocytes on the blood smear.

Autoimmune hemolytic anemias (AIHAs) are another important subset of acquired hemolytic anemias. AIHAs occur when there is antibody-mediated destruction of erythrocytes. The direct Coombs test evaluates for antibody- or complement-­coated erythrocytes. After administration of anti-IgG and anti-­C3 serum, the test evaluates for agglutination of the red cells caused by attached antibodies or complement. Coombs-­positive AIHA can also be categorized by the temperature of agglutination. “Warm” hemolysis often involves IgG autoantibodies (ie, warm agglutinins), while “cold” antibodies, usually IgM autoantibodies, bind at colder temperatures (0-4 °C) and activate complements, including C3. In this patient, the Coombs C3 was positive while the Coombs IgG was negative, which is more suggestive of a cold complement–mediated pathway.

Cold AIHA can be further categorized into primary cold agglutinin disease, secondary cold agglutinin disease, and PCH. Primary cold agglutinin disease is an autoimmune disorder that mostly occurs in adults. Secondary cold AIHA can often be triggered by bacterial infection (commonly M pneumoniae) or viruses including EBV, measles, and mumps.² Medications, including penicillin and cephalosporins, can also be implicated. Secondary cold AIHA is also linked with autoimmune diseases, such as systemic lupus erythematosus and lymphoproliferative disorders. PCH can be identified with the unique presence of a specific autoantibody (ie, D-L autoantibody) that agglutinates at cold temperatures but dissociates on subsequent rewarming.³ Complement remains affixed and activates hemolysis.

The D-L antibody responsible for PCH is an IgG antibody to the P-antigen present on the erythrocyte surface. Since the Coombs test is conducted at normal temperature, it will be positive for the affixed complement but not for IgG. The underlying mechanism for PCH was proposed by Julius Donath, MD, and Karl Landsteiner, MD, in 1904 and is considered to be the first description of autoimmune disease being precipitated by antibodies.⁴ The D-L antibody test itself is uncommonly performed and somewhat difficult to interpret, particularly in adults, and may lead to false-negative results.⁵

PCH is an acquired, cold AIHA more common to children^6,7 and may account for up to 33% of pediatric AIHA cases.⁸ Typical presentation is after an upper respiratory tract illness; however, the trigger is often not identified. Implicated triggers include a number of viruses.⁹ Clinical presentation includes findings of intravascular hemolysis similar to those in our patient. The pathogenic IgG autoantibody is polyclonal and is likely formed because of immune stimulation, which is consistent with the predominance of nonmalignant triggers of this disease process.¹⁰ Hemolysis and associated symptoms are often exacerbated with cold exposure; both typically resolve within 2 weeks. In recurrent cases, which are a minority, immunosuppression may be considered.¹⁰

PCH remains an often-understated cause of hemolytic anemia particularly in children. Lacking obvious pathognomonic clinical symptoms, it may be overlooked for other forms of AIHA or MAHA. However, with a structured approach to evaluation, as with this patient who had hematuria and jaundice, early diagnosis can prevent an unnecessarily extensive workup and can provide reassurance to patient and parents. By understanding the basic categories of hemolytic anemia, the relevant blood testing available, and interpretation of Coombs test results, clinicians can ensure that PCH is a diagnosis that is not left out in the cold.

• Examination for schistocytes on a blood smear can help identify life-threatening causes of hemolytic anemia.
• Characterization of cold AIHA includes defining the underlying etiology as primary cold agglutinin disease, secondary cold agglutinin disease, or PCH.
• PCH is a cold AIHA that is an underrecognized cause of hemolytic anemia in children. The diagnosis of PCH is made by testing for the presence of the D-L antibody.

A previously healthy 4-year-old boy presented to his pediatrician for nasal congestion, left ear pain, and intermittent fevers, which he’d been experiencing for 2 days. His exam was consistent with acute otitis media. Cefdinir was prescribed given a rash allergy to amoxicillin. His fever, congestion, and otalgia improved the next day.

Three days later he developed abdominal pain, fever, and labored breathing; his mother brought him to the emergency department (ED). His temperature was 38.0 °C, heart rate 141 beats per minute, blood pressure 117/71 mm Hg, respiratory rate 22 breaths per minute; he had oxygen saturation of 96% on ambient air. Despite mild accessory muscle use, he appeared comfortable and interactive. His left tympanic membrane was bulging without erythema. His neck was supple and mucous membranes moist. He had neither cervical lymphadenopathy nor conjunctival pallor. The cardiopulmonary exam was normal except for tachycardia. His abdomen was soft and not distended without organomegaly or tenderness.

Upper respiratory tract symptoms are commonly encountered in pediatrics and most often result from self-limited viral processes. Evaluation of a child with upper respiratory tract symptoms aims to identify serious causes like meningitis, as well as assessing the need for antimicrobial therapy. Supportive management is often appropriate in otitis media. His new, more concerning symptoms portend either a progression of the original process causing his upper respiratory tract symptoms or a separate etiology. It is key to determine which signs and symptoms are associated with the primary process and which are compensatory or secondary. If he were to be more ill appearing, for example, it is possible that his respiratory distress may be related to an underlying systemic illness rather than a primary lung process. Respiratory distress, abdominal pain, and fever could be a result of sepsis from an intrabdominal process such as ruptured appendicitis, intussusception, or malrotation with volvulus. Other causes of sepsis, such as meningitis or severe mastoiditis, both rare complications of otitis media, should be considered, although he does not appear severely ill. Acute myelogenous leukemia or other malignancies and illnesses associated with immunodeficiency can present with sepsis and chloromas in the middle ear that can be misconstrued as otitis media.

A chest radiograph demonstrated left lower lobe patchy opacities concerning for pneumonia. Rapid respiratory syncytial virus and influenza antigen test results were negative. Laboratory testing for general bloodwork was not obtained. He was administered a single dose of intramuscular ceftriaxone, prescribed a 5-day course of azithromycin, and discharged home. The child’s breathing gradually improved, but he continued to have subjective fevers. Two days later, he developed dark red urine. His mother brought him back to the outpatient clinic.

At the time of the ED visit, a diagnosis of community-acquired pneumonia was plausible given fever, mildly increased work of breathing, and an opacification on chest radiography. Most community-acquired pneumonia is caused by viruses; common bacterial causes for his age include Streptococcus pneumoniae and Moraxella catarrhalis. The first-line treatment for uncomplicated community-acquired pneumonia in children is amoxicillin, but this was appropriately avoided given his allergy.

The persistent fevers are surprising. The improvement in breathing corresponds to the treatment (and resolution) of community-acquired pneumonia. However, the development of dark urine does not. Red urine—in the absence of ingested pigments (such as those found in beets)—usually results from hematuria, hemoglobinuria, or myoglobinuria. Gross hematuria can originate from the kidneys to the urethral meatus. Abdominal masses, kidney trauma, or underlying kidney disease may all present with gross hematuria (or microscopic hematuria, seen only on urinalysis). The urine should be examined for the presence of heme, protein, and for evidence of infection; microscopy should be performed to examine for cellular casts and dysmorphic red cells. Tests of renal function, a comprehensive metabolic panel, evaluation of hematologic indexes, and assessments of inflammatory markers should be performed.

The child lived with his parents and had no siblings. He experienced no physical trauma, and there was no family history of kidney disease or hematuria. His father had a persistent cough and fever for 1 month, but recovered around the time the patient began to experience his initial symptoms. This was the patient’s third diagnosis of pneumonia. He had not traveled and was up to date with immunizations. He attended day care.

The fact that this is not the first episode of “pneumonia” raises important possibilities. The most likely one is that the child has had multiple viral infections; however, he could have an underlying primary immunodeficiency (PI) that predisposes him to recurrent infections. More severe PIs often present with recurrent sepsis, bacteremia, and failure to thrive, none of which were present in this case. Less severe PIs (such as selective IgA deficiency) could be possible. Another possibility is that these recurrent episodes of pneumonia are a relapsing and remitting noninfectious process, such as an antineutrophil cytoplasmic antibodies–associated vasculitis or anti–glomerular basement membrane disease. The patient’s father’s recent prolonged respiratory symptoms may be suggestive of pertussis or a “walking pneumonia” potentially caused by Mycoplasma or another atypical bacterium.

His temperature was 36.9 °C, heart rate 107 beats per minute, blood pressure was 106/67 mm Hg, and respiratory rate was 24 breaths per minute with oxygen saturation of 100% on ambient air. He was well appearing. His mucous membranes were moist, and oropharynx was clear. He had scleral icterus. The cardiopulmonary exam was normal. He had no significant lymphadenopathy, hepatosplenomegaly, or rashes.

The finding of jaundice is an important diagnostic pivot point, especially when combined with hematuria. The next step is determining if the jaundice is resulting from unconjugated or conjugated hyperbilirubinemia; the former most often stems from hemolysis or impairment in conjugation, while the latter results from intrahepatic or extrahepatic biliary defects. Tests for hepatobiliary injury including evaluations of alanine and aspartate aminotransferases and alkaline phosphatase, as well as for hepatic function such as tests of coagulation, should be performed.

The patient was referred to the ED and admitted for further evaluation. A complete blood count revealed a white blood cell (WBC) count of 10,700/µL (61% polymorphonuclear neutrophils, 30% lymphocytes, 5% monocytes, 3% eosinophils, 1% basophils), hemoglobin count was 10.3 g/dL (reticulocyte 2% with absolute reticulocyte count 58,400/μL), and platelet count was 265,000/µL. Components of the basic metabolic panel were within reference ranges except for a mildly elevated blood urea nitrogen level of 14 mg/dL with normal creatinine level of 0.3 mg/dL. Total protein was 6.7 g/dL (reference range, 6.4-8.3) and albumin 3.9 g/dL (reference range, 3.4-4.8). Alkaline phosphatase level was 188 U/L (reference range, 44-147), aspartate aminotransferase level 76 U/L (reference range, 0-40), and alanine aminotransferase level 12 U/L (reference range, 7-40). Total bilirubin level was 2.4 mg/dL (reference range, less than 1.5) with direct bilirubin level of 0.4 mg/dL. His C-reactive protein level was 1.5 mg/mL (reference range, 0-0.75). Creatinine kinase (CK) level was 2,550 U/L (reference range, 2-198). International Normalized Ratio (INR) was 1.0. Urinalysis was notable for 2+ proteinuria, large hemoglobin pigment, and 6 red blood cells per high power field (reference range, 0-4).

His blood urea nitrogen is elevated, reflecting either prerenal azotemia or increased absorption of nitrogenous products. Unconjugated hyperbilirubinemia may result from impaired hepatic bilirubin uptake (such as in heart failure or portosystemic shunts), impaired bilirubin conjugation (resulting from genetic conditions or drugs), or excess bilirubin production (such as in hemolysis); his anemia and lack of other evidence of hepatic dysfunction point to hemolysis as the etiology. The reticulocyte production index is approximately 1%, which suggests that an increase in erythrocyte generation is present but inadequate. This, however, does not mean that an erythrocyte production abnormality is present since reticulocytosis can be delayed in many cases of acute hemolytic anemia. It is also possible that the same hemolytic process is affecting mature and immature erythrocytes. A peripheral blood smear should be reviewed for evidence of intravascular hemolysis and testing for autoimmune hemolysis should be performed. Notably, his white blood cell and platelet counts are preserved, which makes a bone marrow–involved malignancy or infiltrative process less likely. The alkaline phosphatase elevation may result from either intrahepatic or extrahepatic biliopathy; bone damage is also possible. The elevation of aspartate aminotransferase, CK, and potassium, along with marked urinary heme pigment, may indicate muscle damage; the most common myositis in children is benign acute childhood myositis resulting from viral infection. However, the moderate level of CK elevation seen in this case is nonspecific and can result from many different etiologies. A metabolic myopathy, such as carnitine palmitoyltransferase II deficiency, can be made worse by metabolic stress and result in rhabdomyolysis; the presentations of inborn errors of metabolism are varied and a planned-out, stepwise approach in evaluation is fundamental.

Lactic acid dehydrogenase (LDH) level was 1,457 U/L (reference range, 140-280), and haptoglobin level was less than 6 mg/dL (reference range, 30-200). Peripheral blood smear demonstrated occasional atypical, reactive-appearing lymphocytes with red cell clumping and agglutination, as well as rare target, burr, and fragmented red cells. Test results for urine myoglobin were negative. Results of urine culture were negative. No blood culture was collected.

The elevated LDH, decreased haptoglobin, and findings on the peripheral blood smear confirm hemolysis. The clumping of erythrocytes can be artifactual in the preparation of peripheral smears, but when considered in the context of hemolysis, may be clinically important. Clumping of erythrocytes on the peripheral smear indicates the binding of a protein to antigens on the erythrocyte membrane; when this occurs below body temperature, this is consistent with the presence of a “cold agglutinin,” usually an IgM antibody directed at erythrocyte surface antigens that causes agglutination and destruction, especially in cooler areas of the body. This is a well-known complication of Mycoplasma pneumoniae infections as well as Epstein-Barr virus (EBV) infections; it may also occur with lymphoid malignancies or autoimmune disease.

Direct Coombs IgG test findings were negative, direct Coombs C3 test was positive, and direct Coombs polyspecific test was positive. M pneumoniae IgG antibody level was 1.4 mg/dL (reference ranges: <0.9, negative; 0.91-1.09, equivocal; >1.1, positive); M pneumoniae IgM level was 529 U/mL (reference range: <770, negative). EBV capsid IgM and IgG levels were undetectable. EBV nuclear antigen IgG level was also undetectable. EBV viral load was fewer than 10 copies/mL. Antinuclear antibodies (ANA) level was negative. General IgE and IgM levels were normal, at 11 and 81 mg/dL, respectively. Repeat complete blood count showed WBC of 7,800/µL, hemoglobin of 8.7 g/dL, and platelet count of 341,000/µL. The patient’s hemoglobin remained stable during hospitalization.

This directed testing is helpful in further classifying the patient’s hemolytic anemia. Autoimmune hemolytic anemias are classified into warm antibody–mediated, cold antibody–mediated, and mixed-type forms; drug-induced and alloimmune hemolytic anemias also occur. In addition, both systemic lupus erythematosus and antiphospholipid antibody syndrome can have hemolytic anemia with variable Coombs testing results; neither fit well in this case. The absence of red blood cell–directed IgG antibodies substantially decreases the likelihood of warm antibody–mediated hemolytic anemia. In cold antibody–mediated hemolytic anemia, antibodies bind to the erythrocyte membrane and then adhere to complement C3, which leads to both intravascular and extravascular hemolysis. Important types of cold antibody–mediated hemolytic anemia in children are primary and secondary cold agglutinin disease, along with paroxysmal cold hemoglobinuria. The Donath-Landsteiner test can be helpful in differentiating these conditions. Antibodies to Mycoplasma may be delayed in response to acute infection, and a child who is reinfected may only produce IgG antibodies. Given the patient’s clinical stability and previous health, the most likely diagnosis is Mycoplasma-induced cold antibody–mediated hemolytic anemia. It may be helpful to check convalescent titers to Mycoplasma in 2 to 4 weeks.

Donath-Landsteiner (D-L) antibody test results were positive. Medication-derived hemolytic anemia testing was conducted, but the presence of positive D-L antibody makes the test results inconclusive. This ultimately led to a diagnosis of paroxysmal cold hemoglobinuria (PCH), presumably triggered by a viral syndrome. Convalescent titers to Mycoplasma were not checked given clinical improvement. Because the patient’s hemoglobin was stable during hospitalization, he was not treated with steroids. His parents were counseled on avoiding cold temperatures for several days. Within 1 month, his hemoglobin had recovered without further evidence of hemolysis.

Hemolytic anemia refers to the accelerated destruction of red blood cells and can be further classified as acquired or hereditary.¹ Hereditary conditions causing hemolytic anemia include enzymopathies (eg, glucose-6-phosphate dehydrogenase deficiency), hemoglobinopathies (eg, sickle cell disease), and membrane abnormalities (eg, hereditary spherocytosis). Acquired pathologies include microangiopathic hemolytic anemia (MAHA), anemias directly caused by certain infections such as malaria, and immune-mediated (Coombs-positive) hemolytic anemias.

MAHA can sometimes be life-threatening and is therefore important to identify quickly. In the right clinical context, such processes may be rapidly recognized by the presence of schistocytes on blood smear in addition to an elevated serum LDH level. Schistocytes suggest mechanical destruction of erythrocytes in the vasculature, the hallmark of MAHA. Important MAHAs include thrombocytopenic purpura, hemolytic-uremic syndrome, and disseminated intravascular coagulation. Though this patient did have a mildly elevated LDH, MAHA was less likely because there were no schistocytes on the blood smear.

Autoimmune hemolytic anemias (AIHAs) are another important subset of acquired hemolytic anemias. AIHAs occur when there is antibody-mediated destruction of erythrocytes. The direct Coombs test evaluates for antibody- or complement-­coated erythrocytes. After administration of anti-IgG and anti-­C3 serum, the test evaluates for agglutination of the red cells caused by attached antibodies or complement. Coombs-­positive AIHA can also be categorized by the temperature of agglutination. “Warm” hemolysis often involves IgG autoantibodies (ie, warm agglutinins), while “cold” antibodies, usually IgM autoantibodies, bind at colder temperatures (0-4 °C) and activate complements, including C3. In this patient, the Coombs C3 was positive while the Coombs IgG was negative, which is more suggestive of a cold complement–mediated pathway.

Cold AIHA can be further categorized into primary cold agglutinin disease, secondary cold agglutinin disease, and PCH. Primary cold agglutinin disease is an autoimmune disorder that mostly occurs in adults. Secondary cold AIHA can often be triggered by bacterial infection (commonly M pneumoniae) or viruses including EBV, measles, and mumps.² Medications, including penicillin and cephalosporins, can also be implicated. Secondary cold AIHA is also linked with autoimmune diseases, such as systemic lupus erythematosus and lymphoproliferative disorders. PCH can be identified with the unique presence of a specific autoantibody (ie, D-L autoantibody) that agglutinates at cold temperatures but dissociates on subsequent rewarming.³ Complement remains affixed and activates hemolysis.

The D-L antibody responsible for PCH is an IgG antibody to the P-antigen present on the erythrocyte surface. Since the Coombs test is conducted at normal temperature, it will be positive for the affixed complement but not for IgG. The underlying mechanism for PCH was proposed by Julius Donath, MD, and Karl Landsteiner, MD, in 1904 and is considered to be the first description of autoimmune disease being precipitated by antibodies.⁴ The D-L antibody test itself is uncommonly performed and somewhat difficult to interpret, particularly in adults, and may lead to false-negative results.⁵

PCH is an acquired, cold AIHA more common to children^6,7 and may account for up to 33% of pediatric AIHA cases.⁸ Typical presentation is after an upper respiratory tract illness; however, the trigger is often not identified. Implicated triggers include a number of viruses.⁹ Clinical presentation includes findings of intravascular hemolysis similar to those in our patient. The pathogenic IgG autoantibody is polyclonal and is likely formed because of immune stimulation, which is consistent with the predominance of nonmalignant triggers of this disease process.¹⁰ Hemolysis and associated symptoms are often exacerbated with cold exposure; both typically resolve within 2 weeks. In recurrent cases, which are a minority, immunosuppression may be considered.¹⁰

PCH remains an often-understated cause of hemolytic anemia particularly in children. Lacking obvious pathognomonic clinical symptoms, it may be overlooked for other forms of AIHA or MAHA. However, with a structured approach to evaluation, as with this patient who had hematuria and jaundice, early diagnosis can prevent an unnecessarily extensive workup and can provide reassurance to patient and parents. By understanding the basic categories of hemolytic anemia, the relevant blood testing available, and interpretation of Coombs test results, clinicians can ensure that PCH is a diagnosis that is not left out in the cold.

• Examination for schistocytes on a blood smear can help identify life-threatening causes of hemolytic anemia.
• Characterization of cold AIHA includes defining the underlying etiology as primary cold agglutinin disease, secondary cold agglutinin disease, or PCH.
• PCH is a cold AIHA that is an underrecognized cause of hemolytic anemia in children. The diagnosis of PCH is made by testing for the presence of the D-L antibody.

A previously healthy 4-year-old boy presented to his pediatrician for nasal congestion, left ear pain, and intermittent fevers, which he’d been experiencing for 2 days. His exam was consistent with acute otitis media. Cefdinir was prescribed given a rash allergy to amoxicillin. His fever, congestion, and otalgia improved the next day.

Three days later he developed abdominal pain, fever, and labored breathing; his mother brought him to the emergency department (ED). His temperature was 38.0 °C, heart rate 141 beats per minute, blood pressure 117/71 mm Hg, respiratory rate 22 breaths per minute; he had oxygen saturation of 96% on ambient air. Despite mild accessory muscle use, he appeared comfortable and interactive. His left tympanic membrane was bulging without erythema. His neck was supple and mucous membranes moist. He had neither cervical lymphadenopathy nor conjunctival pallor. The cardiopulmonary exam was normal except for tachycardia. His abdomen was soft and not distended without organomegaly or tenderness.

Upper respiratory tract symptoms are commonly encountered in pediatrics and most often result from self-limited viral processes. Evaluation of a child with upper respiratory tract symptoms aims to identify serious causes like meningitis, as well as assessing the need for antimicrobial therapy. Supportive management is often appropriate in otitis media. His new, more concerning symptoms portend either a progression of the original process causing his upper respiratory tract symptoms or a separate etiology. It is key to determine which signs and symptoms are associated with the primary process and which are compensatory or secondary. If he were to be more ill appearing, for example, it is possible that his respiratory distress may be related to an underlying systemic illness rather than a primary lung process. Respiratory distress, abdominal pain, and fever could be a result of sepsis from an intrabdominal process such as ruptured appendicitis, intussusception, or malrotation with volvulus. Other causes of sepsis, such as meningitis or severe mastoiditis, both rare complications of otitis media, should be considered, although he does not appear severely ill. Acute myelogenous leukemia or other malignancies and illnesses associated with immunodeficiency can present with sepsis and chloromas in the middle ear that can be misconstrued as otitis media.

A chest radiograph demonstrated left lower lobe patchy opacities concerning for pneumonia. Rapid respiratory syncytial virus and influenza antigen test results were negative. Laboratory testing for general bloodwork was not obtained. He was administered a single dose of intramuscular ceftriaxone, prescribed a 5-day course of azithromycin, and discharged home. The child’s breathing gradually improved, but he continued to have subjective fevers. Two days later, he developed dark red urine. His mother brought him back to the outpatient clinic.

At the time of the ED visit, a diagnosis of community-acquired pneumonia was plausible given fever, mildly increased work of breathing, and an opacification on chest radiography. Most community-acquired pneumonia is caused by viruses; common bacterial causes for his age include Streptococcus pneumoniae and Moraxella catarrhalis. The first-line treatment for uncomplicated community-acquired pneumonia in children is amoxicillin, but this was appropriately avoided given his allergy.

The persistent fevers are surprising. The improvement in breathing corresponds to the treatment (and resolution) of community-acquired pneumonia. However, the development of dark urine does not. Red urine—in the absence of ingested pigments (such as those found in beets)—usually results from hematuria, hemoglobinuria, or myoglobinuria. Gross hematuria can originate from the kidneys to the urethral meatus. Abdominal masses, kidney trauma, or underlying kidney disease may all present with gross hematuria (or microscopic hematuria, seen only on urinalysis). The urine should be examined for the presence of heme, protein, and for evidence of infection; microscopy should be performed to examine for cellular casts and dysmorphic red cells. Tests of renal function, a comprehensive metabolic panel, evaluation of hematologic indexes, and assessments of inflammatory markers should be performed.

The child lived with his parents and had no siblings. He experienced no physical trauma, and there was no family history of kidney disease or hematuria. His father had a persistent cough and fever for 1 month, but recovered around the time the patient began to experience his initial symptoms. This was the patient’s third diagnosis of pneumonia. He had not traveled and was up to date with immunizations. He attended day care.

The fact that this is not the first episode of “pneumonia” raises important possibilities. The most likely one is that the child has had multiple viral infections; however, he could have an underlying primary immunodeficiency (PI) that predisposes him to recurrent infections. More severe PIs often present with recurrent sepsis, bacteremia, and failure to thrive, none of which were present in this case. Less severe PIs (such as selective IgA deficiency) could be possible. Another possibility is that these recurrent episodes of pneumonia are a relapsing and remitting noninfectious process, such as an antineutrophil cytoplasmic antibodies–associated vasculitis or anti–glomerular basement membrane disease. The patient’s father’s recent prolonged respiratory symptoms may be suggestive of pertussis or a “walking pneumonia” potentially caused by Mycoplasma or another atypical bacterium.

His temperature was 36.9 °C, heart rate 107 beats per minute, blood pressure was 106/67 mm Hg, and respiratory rate was 24 breaths per minute with oxygen saturation of 100% on ambient air. He was well appearing. His mucous membranes were moist, and oropharynx was clear. He had scleral icterus. The cardiopulmonary exam was normal. He had no significant lymphadenopathy, hepatosplenomegaly, or rashes.

The finding of jaundice is an important diagnostic pivot point, especially when combined with hematuria. The next step is determining if the jaundice is resulting from unconjugated or conjugated hyperbilirubinemia; the former most often stems from hemolysis or impairment in conjugation, while the latter results from intrahepatic or extrahepatic biliary defects. Tests for hepatobiliary injury including evaluations of alanine and aspartate aminotransferases and alkaline phosphatase, as well as for hepatic function such as tests of coagulation, should be performed.

The patient was referred to the ED and admitted for further evaluation. A complete blood count revealed a white blood cell (WBC) count of 10,700/µL (61% polymorphonuclear neutrophils, 30% lymphocytes, 5% monocytes, 3% eosinophils, 1% basophils), hemoglobin count was 10.3 g/dL (reticulocyte 2% with absolute reticulocyte count 58,400/μL), and platelet count was 265,000/µL. Components of the basic metabolic panel were within reference ranges except for a mildly elevated blood urea nitrogen level of 14 mg/dL with normal creatinine level of 0.3 mg/dL. Total protein was 6.7 g/dL (reference range, 6.4-8.3) and albumin 3.9 g/dL (reference range, 3.4-4.8). Alkaline phosphatase level was 188 U/L (reference range, 44-147), aspartate aminotransferase level 76 U/L (reference range, 0-40), and alanine aminotransferase level 12 U/L (reference range, 7-40). Total bilirubin level was 2.4 mg/dL (reference range, less than 1.5) with direct bilirubin level of 0.4 mg/dL. His C-reactive protein level was 1.5 mg/mL (reference range, 0-0.75). Creatinine kinase (CK) level was 2,550 U/L (reference range, 2-198). International Normalized Ratio (INR) was 1.0. Urinalysis was notable for 2+ proteinuria, large hemoglobin pigment, and 6 red blood cells per high power field (reference range, 0-4).

His blood urea nitrogen is elevated, reflecting either prerenal azotemia or increased absorption of nitrogenous products. Unconjugated hyperbilirubinemia may result from impaired hepatic bilirubin uptake (such as in heart failure or portosystemic shunts), impaired bilirubin conjugation (resulting from genetic conditions or drugs), or excess bilirubin production (such as in hemolysis); his anemia and lack of other evidence of hepatic dysfunction point to hemolysis as the etiology. The reticulocyte production index is approximately 1%, which suggests that an increase in erythrocyte generation is present but inadequate. This, however, does not mean that an erythrocyte production abnormality is present since reticulocytosis can be delayed in many cases of acute hemolytic anemia. It is also possible that the same hemolytic process is affecting mature and immature erythrocytes. A peripheral blood smear should be reviewed for evidence of intravascular hemolysis and testing for autoimmune hemolysis should be performed. Notably, his white blood cell and platelet counts are preserved, which makes a bone marrow–involved malignancy or infiltrative process less likely. The alkaline phosphatase elevation may result from either intrahepatic or extrahepatic biliopathy; bone damage is also possible. The elevation of aspartate aminotransferase, CK, and potassium, along with marked urinary heme pigment, may indicate muscle damage; the most common myositis in children is benign acute childhood myositis resulting from viral infection. However, the moderate level of CK elevation seen in this case is nonspecific and can result from many different etiologies. A metabolic myopathy, such as carnitine palmitoyltransferase II deficiency, can be made worse by metabolic stress and result in rhabdomyolysis; the presentations of inborn errors of metabolism are varied and a planned-out, stepwise approach in evaluation is fundamental.

Lactic acid dehydrogenase (LDH) level was 1,457 U/L (reference range, 140-280), and haptoglobin level was less than 6 mg/dL (reference range, 30-200). Peripheral blood smear demonstrated occasional atypical, reactive-appearing lymphocytes with red cell clumping and agglutination, as well as rare target, burr, and fragmented red cells. Test results for urine myoglobin were negative. Results of urine culture were negative. No blood culture was collected.

The elevated LDH, decreased haptoglobin, and findings on the peripheral blood smear confirm hemolysis. The clumping of erythrocytes can be artifactual in the preparation of peripheral smears, but when considered in the context of hemolysis, may be clinically important. Clumping of erythrocytes on the peripheral smear indicates the binding of a protein to antigens on the erythrocyte membrane; when this occurs below body temperature, this is consistent with the presence of a “cold agglutinin,” usually an IgM antibody directed at erythrocyte surface antigens that causes agglutination and destruction, especially in cooler areas of the body. This is a well-known complication of Mycoplasma pneumoniae infections as well as Epstein-Barr virus (EBV) infections; it may also occur with lymphoid malignancies or autoimmune disease.

Direct Coombs IgG test findings were negative, direct Coombs C3 test was positive, and direct Coombs polyspecific test was positive. M pneumoniae IgG antibody level was 1.4 mg/dL (reference ranges: <0.9, negative; 0.91-1.09, equivocal; >1.1, positive); M pneumoniae IgM level was 529 U/mL (reference range: <770, negative). EBV capsid IgM and IgG levels were undetectable. EBV nuclear antigen IgG level was also undetectable. EBV viral load was fewer than 10 copies/mL. Antinuclear antibodies (ANA) level was negative. General IgE and IgM levels were normal, at 11 and 81 mg/dL, respectively. Repeat complete blood count showed WBC of 7,800/µL, hemoglobin of 8.7 g/dL, and platelet count of 341,000/µL. The patient’s hemoglobin remained stable during hospitalization.

This directed testing is helpful in further classifying the patient’s hemolytic anemia. Autoimmune hemolytic anemias are classified into warm antibody–mediated, cold antibody–mediated, and mixed-type forms; drug-induced and alloimmune hemolytic anemias also occur. In addition, both systemic lupus erythematosus and antiphospholipid antibody syndrome can have hemolytic anemia with variable Coombs testing results; neither fit well in this case. The absence of red blood cell–directed IgG antibodies substantially decreases the likelihood of warm antibody–mediated hemolytic anemia. In cold antibody–mediated hemolytic anemia, antibodies bind to the erythrocyte membrane and then adhere to complement C3, which leads to both intravascular and extravascular hemolysis. Important types of cold antibody–mediated hemolytic anemia in children are primary and secondary cold agglutinin disease, along with paroxysmal cold hemoglobinuria. The Donath-Landsteiner test can be helpful in differentiating these conditions. Antibodies to Mycoplasma may be delayed in response to acute infection, and a child who is reinfected may only produce IgG antibodies. Given the patient’s clinical stability and previous health, the most likely diagnosis is Mycoplasma-induced cold antibody–mediated hemolytic anemia. It may be helpful to check convalescent titers to Mycoplasma in 2 to 4 weeks.

Donath-Landsteiner (D-L) antibody test results were positive. Medication-derived hemolytic anemia testing was conducted, but the presence of positive D-L antibody makes the test results inconclusive. This ultimately led to a diagnosis of paroxysmal cold hemoglobinuria (PCH), presumably triggered by a viral syndrome. Convalescent titers to Mycoplasma were not checked given clinical improvement. Because the patient’s hemoglobin was stable during hospitalization, he was not treated with steroids. His parents were counseled on avoiding cold temperatures for several days. Within 1 month, his hemoglobin had recovered without further evidence of hemolysis.

Hemolytic anemia refers to the accelerated destruction of red blood cells and can be further classified as acquired or hereditary.¹ Hereditary conditions causing hemolytic anemia include enzymopathies (eg, glucose-6-phosphate dehydrogenase deficiency), hemoglobinopathies (eg, sickle cell disease), and membrane abnormalities (eg, hereditary spherocytosis). Acquired pathologies include microangiopathic hemolytic anemia (MAHA), anemias directly caused by certain infections such as malaria, and immune-mediated (Coombs-positive) hemolytic anemias.

MAHA can sometimes be life-threatening and is therefore important to identify quickly. In the right clinical context, such processes may be rapidly recognized by the presence of schistocytes on blood smear in addition to an elevated serum LDH level. Schistocytes suggest mechanical destruction of erythrocytes in the vasculature, the hallmark of MAHA. Important MAHAs include thrombocytopenic purpura, hemolytic-uremic syndrome, and disseminated intravascular coagulation. Though this patient did have a mildly elevated LDH, MAHA was less likely because there were no schistocytes on the blood smear.

Autoimmune hemolytic anemias (AIHAs) are another important subset of acquired hemolytic anemias. AIHAs occur when there is antibody-mediated destruction of erythrocytes. The direct Coombs test evaluates for antibody- or complement-­coated erythrocytes. After administration of anti-IgG and anti-­C3 serum, the test evaluates for agglutination of the red cells caused by attached antibodies or complement. Coombs-­positive AIHA can also be categorized by the temperature of agglutination. “Warm” hemolysis often involves IgG autoantibodies (ie, warm agglutinins), while “cold” antibodies, usually IgM autoantibodies, bind at colder temperatures (0-4 °C) and activate complements, including C3. In this patient, the Coombs C3 was positive while the Coombs IgG was negative, which is more suggestive of a cold complement–mediated pathway.

Cold AIHA can be further categorized into primary cold agglutinin disease, secondary cold agglutinin disease, and PCH. Primary cold agglutinin disease is an autoimmune disorder that mostly occurs in adults. Secondary cold AIHA can often be triggered by bacterial infection (commonly M pneumoniae) or viruses including EBV, measles, and mumps.² Medications, including penicillin and cephalosporins, can also be implicated. Secondary cold AIHA is also linked with autoimmune diseases, such as systemic lupus erythematosus and lymphoproliferative disorders. PCH can be identified with the unique presence of a specific autoantibody (ie, D-L autoantibody) that agglutinates at cold temperatures but dissociates on subsequent rewarming.³ Complement remains affixed and activates hemolysis.

The D-L antibody responsible for PCH is an IgG antibody to the P-antigen present on the erythrocyte surface. Since the Coombs test is conducted at normal temperature, it will be positive for the affixed complement but not for IgG. The underlying mechanism for PCH was proposed by Julius Donath, MD, and Karl Landsteiner, MD, in 1904 and is considered to be the first description of autoimmune disease being precipitated by antibodies.⁴ The D-L antibody test itself is uncommonly performed and somewhat difficult to interpret, particularly in adults, and may lead to false-negative results.⁵

PCH is an acquired, cold AIHA more common to children^6,7 and may account for up to 33% of pediatric AIHA cases.⁸ Typical presentation is after an upper respiratory tract illness; however, the trigger is often not identified. Implicated triggers include a number of viruses.⁹ Clinical presentation includes findings of intravascular hemolysis similar to those in our patient. The pathogenic IgG autoantibody is polyclonal and is likely formed because of immune stimulation, which is consistent with the predominance of nonmalignant triggers of this disease process.¹⁰ Hemolysis and associated symptoms are often exacerbated with cold exposure; both typically resolve within 2 weeks. In recurrent cases, which are a minority, immunosuppression may be considered.¹⁰

PCH remains an often-understated cause of hemolytic anemia particularly in children. Lacking obvious pathognomonic clinical symptoms, it may be overlooked for other forms of AIHA or MAHA. However, with a structured approach to evaluation, as with this patient who had hematuria and jaundice, early diagnosis can prevent an unnecessarily extensive workup and can provide reassurance to patient and parents. By understanding the basic categories of hemolytic anemia, the relevant blood testing available, and interpretation of Coombs test results, clinicians can ensure that PCH is a diagnosis that is not left out in the cold.

• Examination for schistocytes on a blood smear can help identify life-threatening causes of hemolytic anemia.
• Characterization of cold AIHA includes defining the underlying etiology as primary cold agglutinin disease, secondary cold agglutinin disease, or PCH.
• PCH is a cold AIHA that is an underrecognized cause of hemolytic anemia in children. The diagnosis of PCH is made by testing for the presence of the D-L antibody.

The story of the coronavirus disease 2019 (COVID-19) pandemic in the United States has been defined, in part, by a persistent shortage of medical supplies that has made it difficult and dangerous for healthcare workers to care for infected patients. States, health systems, and even individual hospitals are currently competing against one another—sometimes at auction—to obtain personal protective equipment (PPE). This “Wild West” scenario has resulted in bizarre stories involving attempts to obtain PPE. One health system recently described a James Bond–like pursuit of essential PPE, complete with a covert trip to an industrial warehouse, trucks filled with masks but labeled as food delivery vehicles, and an intervention by a United States congressman.¹ Many states have experienced analogous, but still atypical, stories: masks flown in from China using the private jet of a professional sports team owner,² widespread use of novel sterilization modalities to allow PPE reuse,³ and one attempt to purchase price-gouged PPE from the host of the show “Shark Tank.”⁴ In some cases, hospitals and healthcare workers have pleaded for PPE on fundraising and social media sites.⁵

These profound deviations from operations of contemporary health system supply chains would have seemed beyond belief just a few months ago. Instead, they now echo the collective experiences of healthcare stakeholders trying to obtain PPE to protect their frontline healthcare workers during the COVID-19 pandemic.

How did we get into this situation? The manufacture of medical supplies like gowns and masks is a highly competitive business with very slim margins, and as a result, medical equipment manufacturers aim to match their supply with the market’s demand, with hospitals and health systems using just-in-time ordering to limit excess inventory.⁶ While this approach adds efficiency and reduces costs, it also renders manufacturers and customers vulnerable to supply disruptions and shortages when need surges. The COVID-19 pandemic represents perhaps the most extreme example of a massive, widespread surge in demand that occurred multifocally and in a highly compressed time frame. Unlike other industries (eg, consumer paper products), however, in which demand exceeding supply causes inconvenience, the lack of PPE has led to critical public health consequences, with lives of both healthcare workers and vulnerable patients lost because of these shortages of medical equipment.

There are many reasons for the PPE crisis. As noted above, manufacturers have prioritized efficiency over the ability to quickly increase production. They adhere to just-in-time ordering rather than planning for a surge in demand with extra production capacity, all to avoid having warehouses filled with unsold products if surges never occur. This strategy, compounded by the fact that most PPE in the United States is imported from areas in Asia that were profoundly affected early on by COVID-19, led to the observed widespread shortages. When PPE became unavailable from usual suppliers, hospitals were unable to locate other sources of existing PPE because of a lack of transparency about where PPE could be found and how it could be obtained. The Food and Drug Administration and other federal regulatory agencies maintained strict regulations around PPE production and, despite the crisis, made few exceptions.⁷ The FDA did grant a few Emergency Use Authorizations (EUAs) for certain improvised, decontaminated, or alternative respirators (eg, the Chinese-made KN95), but it has only very infrequently issued EUAs to allow domestic manufacturers to ramp up production.⁸ These failures were accompanied by an serious increase in PPE use, leading to spikes in price, price gouging, and hoarding,⁹ problems that were further magnified as health systems and hospitals were forced to compete with nonhealthcare businesses for PPE.

The Defense Production Act (DPA) gives the federal government the power to increase production of goods needed during a crisis⁸ to offer purchasing guarantees, coordinate federal agencies, and regulate distribution and pricing. However, the current administration’s failure to mount a coordinated federal response has contributed to the observed market instability, medical supply shortages, and public health crisis we face. We have previously recommended that the federal government use the power of the DPA to reduce manufacturers’ risk of being uncompensated for excess supply, support temporary reductions in regulatory barriers, and create mandatory centralized reporting of PPE supply, including completed PPE and its components.¹⁰ We stand by these recommendations but also acknowledge that hospitals and health systems may be simultaneously considering how to best prepare for future crises and even surges in demand over the next 18 months as the COVID-19 pandemic continues.

1. Encourage mandates at the hospital, health system, and state level regarding minimum inventory levels for essential equipment. Stockpiles are essential for emergency preparedness. In the long term, these sorts of stockpiles are economically infeasible without government help to maintain them. In the near term, however, it is sensible that hospitals and health systems would maintain a minimum of 2 weeks’ worth of PPE to prepare for expected regional spikes in COVID-19 cases. However, a soon-to-published study suggests that over 40% of hospitals had a PPE stockpile of less than 2 weeks.¹¹ Although this survey was conducted at the height of the shortage, it suggests that there is opportunity for improvement.

2. Coordinate efforts among states and health systems to collect and report inventory, regionalize resources, and coordinate their distribution. The best example of this is the seven-state purchasing consortium announced by New York Governor Andrew Cuomo in early May.¹² Unfortunately, since the announcement, there have been few details about whether the states were successful in their effort to reduce prices or to obtain PPE in bulk. Still, hospitals and health systems could join or emulate purchasing collaboratives to allow resources to be better allocated according to need. There are barriers to such collaboratives because the market is currently set up to encourage competition among health systems and hospitals. During the pandemic, however, cooperation has increasingly been favored over competition in science and healthcare delivery. There are also existing hospital purchasing collaboratives (eg, Premier, Inc¹³), which have taken steps to vet suppliers and improve access to PPE, but it is not clear how successful these efforts have been to date.

3. Advocate for strong federal leadership, including support for increased domestic manufacturing; replenishment and maintenance of state and health system stockpiles of PPE, ventilators, and medications; and development of a centrally coordinated PPE allocation and distribution process. While hospitals and health systems may favor remaining as apolitical as possible, the need for a federal response to stabilize the PPE market may be too urgent and necessary to ignore.

As hospitals and health systems prepare for continued surges in COVID-19 cases, they face challenges in providing PPE for frontline clinicians and staff. A federal plan to enhance nimbleness in responding to multifocal, geographic outbreaks and ensure awareness regarding inventory would improve our chances to successfully navigate the next pandemic and optimize the protection of our health workers, patients, and public health. In the absence of such a plan, hospitals should maintain a minimum of 2 weeks’ worth of PPE to prepare for expected regional spikes in COVID-19 cases and should continue to attempt to coordinate efforts among states and health systems to collect and report inventory, regionalize resources, and coordinate their distribution.

The story of the coronavirus disease 2019 (COVID-19) pandemic in the United States has been defined, in part, by a persistent shortage of medical supplies that has made it difficult and dangerous for healthcare workers to care for infected patients. States, health systems, and even individual hospitals are currently competing against one another—sometimes at auction—to obtain personal protective equipment (PPE). This “Wild West” scenario has resulted in bizarre stories involving attempts to obtain PPE. One health system recently described a James Bond–like pursuit of essential PPE, complete with a covert trip to an industrial warehouse, trucks filled with masks but labeled as food delivery vehicles, and an intervention by a United States congressman.¹ Many states have experienced analogous, but still atypical, stories: masks flown in from China using the private jet of a professional sports team owner,² widespread use of novel sterilization modalities to allow PPE reuse,³ and one attempt to purchase price-gouged PPE from the host of the show “Shark Tank.”⁴ In some cases, hospitals and healthcare workers have pleaded for PPE on fundraising and social media sites.⁵

These profound deviations from operations of contemporary health system supply chains would have seemed beyond belief just a few months ago. Instead, they now echo the collective experiences of healthcare stakeholders trying to obtain PPE to protect their frontline healthcare workers during the COVID-19 pandemic.

How did we get into this situation? The manufacture of medical supplies like gowns and masks is a highly competitive business with very slim margins, and as a result, medical equipment manufacturers aim to match their supply with the market’s demand, with hospitals and health systems using just-in-time ordering to limit excess inventory.⁶ While this approach adds efficiency and reduces costs, it also renders manufacturers and customers vulnerable to supply disruptions and shortages when need surges. The COVID-19 pandemic represents perhaps the most extreme example of a massive, widespread surge in demand that occurred multifocally and in a highly compressed time frame. Unlike other industries (eg, consumer paper products), however, in which demand exceeding supply causes inconvenience, the lack of PPE has led to critical public health consequences, with lives of both healthcare workers and vulnerable patients lost because of these shortages of medical equipment.

There are many reasons for the PPE crisis. As noted above, manufacturers have prioritized efficiency over the ability to quickly increase production. They adhere to just-in-time ordering rather than planning for a surge in demand with extra production capacity, all to avoid having warehouses filled with unsold products if surges never occur. This strategy, compounded by the fact that most PPE in the United States is imported from areas in Asia that were profoundly affected early on by COVID-19, led to the observed widespread shortages. When PPE became unavailable from usual suppliers, hospitals were unable to locate other sources of existing PPE because of a lack of transparency about where PPE could be found and how it could be obtained. The Food and Drug Administration and other federal regulatory agencies maintained strict regulations around PPE production and, despite the crisis, made few exceptions.⁷ The FDA did grant a few Emergency Use Authorizations (EUAs) for certain improvised, decontaminated, or alternative respirators (eg, the Chinese-made KN95), but it has only very infrequently issued EUAs to allow domestic manufacturers to ramp up production.⁸ These failures were accompanied by an serious increase in PPE use, leading to spikes in price, price gouging, and hoarding,⁹ problems that were further magnified as health systems and hospitals were forced to compete with nonhealthcare businesses for PPE.

The Defense Production Act (DPA) gives the federal government the power to increase production of goods needed during a crisis⁸ to offer purchasing guarantees, coordinate federal agencies, and regulate distribution and pricing. However, the current administration’s failure to mount a coordinated federal response has contributed to the observed market instability, medical supply shortages, and public health crisis we face. We have previously recommended that the federal government use the power of the DPA to reduce manufacturers’ risk of being uncompensated for excess supply, support temporary reductions in regulatory barriers, and create mandatory centralized reporting of PPE supply, including completed PPE and its components.¹⁰ We stand by these recommendations but also acknowledge that hospitals and health systems may be simultaneously considering how to best prepare for future crises and even surges in demand over the next 18 months as the COVID-19 pandemic continues.

1. Encourage mandates at the hospital, health system, and state level regarding minimum inventory levels for essential equipment. Stockpiles are essential for emergency preparedness. In the long term, these sorts of stockpiles are economically infeasible without government help to maintain them. In the near term, however, it is sensible that hospitals and health systems would maintain a minimum of 2 weeks’ worth of PPE to prepare for expected regional spikes in COVID-19 cases. However, a soon-to-published study suggests that over 40% of hospitals had a PPE stockpile of less than 2 weeks.¹¹ Although this survey was conducted at the height of the shortage, it suggests that there is opportunity for improvement.

2. Coordinate efforts among states and health systems to collect and report inventory, regionalize resources, and coordinate their distribution. The best example of this is the seven-state purchasing consortium announced by New York Governor Andrew Cuomo in early May.¹² Unfortunately, since the announcement, there have been few details about whether the states were successful in their effort to reduce prices or to obtain PPE in bulk. Still, hospitals and health systems could join or emulate purchasing collaboratives to allow resources to be better allocated according to need. There are barriers to such collaboratives because the market is currently set up to encourage competition among health systems and hospitals. During the pandemic, however, cooperation has increasingly been favored over competition in science and healthcare delivery. There are also existing hospital purchasing collaboratives (eg, Premier, Inc¹³), which have taken steps to vet suppliers and improve access to PPE, but it is not clear how successful these efforts have been to date.

3. Advocate for strong federal leadership, including support for increased domestic manufacturing; replenishment and maintenance of state and health system stockpiles of PPE, ventilators, and medications; and development of a centrally coordinated PPE allocation and distribution process. While hospitals and health systems may favor remaining as apolitical as possible, the need for a federal response to stabilize the PPE market may be too urgent and necessary to ignore.

As hospitals and health systems prepare for continued surges in COVID-19 cases, they face challenges in providing PPE for frontline clinicians and staff. A federal plan to enhance nimbleness in responding to multifocal, geographic outbreaks and ensure awareness regarding inventory would improve our chances to successfully navigate the next pandemic and optimize the protection of our health workers, patients, and public health. In the absence of such a plan, hospitals should maintain a minimum of 2 weeks’ worth of PPE to prepare for expected regional spikes in COVID-19 cases and should continue to attempt to coordinate efforts among states and health systems to collect and report inventory, regionalize resources, and coordinate their distribution.

The story of the coronavirus disease 2019 (COVID-19) pandemic in the United States has been defined, in part, by a persistent shortage of medical supplies that has made it difficult and dangerous for healthcare workers to care for infected patients. States, health systems, and even individual hospitals are currently competing against one another—sometimes at auction—to obtain personal protective equipment (PPE). This “Wild West” scenario has resulted in bizarre stories involving attempts to obtain PPE. One health system recently described a James Bond–like pursuit of essential PPE, complete with a covert trip to an industrial warehouse, trucks filled with masks but labeled as food delivery vehicles, and an intervention by a United States congressman.¹ Many states have experienced analogous, but still atypical, stories: masks flown in from China using the private jet of a professional sports team owner,² widespread use of novel sterilization modalities to allow PPE reuse,³ and one attempt to purchase price-gouged PPE from the host of the show “Shark Tank.”⁴ In some cases, hospitals and healthcare workers have pleaded for PPE on fundraising and social media sites.⁵

These profound deviations from operations of contemporary health system supply chains would have seemed beyond belief just a few months ago. Instead, they now echo the collective experiences of healthcare stakeholders trying to obtain PPE to protect their frontline healthcare workers during the COVID-19 pandemic.

How did we get into this situation? The manufacture of medical supplies like gowns and masks is a highly competitive business with very slim margins, and as a result, medical equipment manufacturers aim to match their supply with the market’s demand, with hospitals and health systems using just-in-time ordering to limit excess inventory.⁶ While this approach adds efficiency and reduces costs, it also renders manufacturers and customers vulnerable to supply disruptions and shortages when need surges. The COVID-19 pandemic represents perhaps the most extreme example of a massive, widespread surge in demand that occurred multifocally and in a highly compressed time frame. Unlike other industries (eg, consumer paper products), however, in which demand exceeding supply causes inconvenience, the lack of PPE has led to critical public health consequences, with lives of both healthcare workers and vulnerable patients lost because of these shortages of medical equipment.

There are many reasons for the PPE crisis. As noted above, manufacturers have prioritized efficiency over the ability to quickly increase production. They adhere to just-in-time ordering rather than planning for a surge in demand with extra production capacity, all to avoid having warehouses filled with unsold products if surges never occur. This strategy, compounded by the fact that most PPE in the United States is imported from areas in Asia that were profoundly affected early on by COVID-19, led to the observed widespread shortages. When PPE became unavailable from usual suppliers, hospitals were unable to locate other sources of existing PPE because of a lack of transparency about where PPE could be found and how it could be obtained. The Food and Drug Administration and other federal regulatory agencies maintained strict regulations around PPE production and, despite the crisis, made few exceptions.⁷ The FDA did grant a few Emergency Use Authorizations (EUAs) for certain improvised, decontaminated, or alternative respirators (eg, the Chinese-made KN95), but it has only very infrequently issued EUAs to allow domestic manufacturers to ramp up production.⁸ These failures were accompanied by an serious increase in PPE use, leading to spikes in price, price gouging, and hoarding,⁹ problems that were further magnified as health systems and hospitals were forced to compete with nonhealthcare businesses for PPE.

The Defense Production Act (DPA) gives the federal government the power to increase production of goods needed during a crisis⁸ to offer purchasing guarantees, coordinate federal agencies, and regulate distribution and pricing. However, the current administration’s failure to mount a coordinated federal response has contributed to the observed market instability, medical supply shortages, and public health crisis we face. We have previously recommended that the federal government use the power of the DPA to reduce manufacturers’ risk of being uncompensated for excess supply, support temporary reductions in regulatory barriers, and create mandatory centralized reporting of PPE supply, including completed PPE and its components.¹⁰ We stand by these recommendations but also acknowledge that hospitals and health systems may be simultaneously considering how to best prepare for future crises and even surges in demand over the next 18 months as the COVID-19 pandemic continues.

1. Encourage mandates at the hospital, health system, and state level regarding minimum inventory levels for essential equipment. Stockpiles are essential for emergency preparedness. In the long term, these sorts of stockpiles are economically infeasible without government help to maintain them. In the near term, however, it is sensible that hospitals and health systems would maintain a minimum of 2 weeks’ worth of PPE to prepare for expected regional spikes in COVID-19 cases. However, a soon-to-published study suggests that over 40% of hospitals had a PPE stockpile of less than 2 weeks.¹¹ Although this survey was conducted at the height of the shortage, it suggests that there is opportunity for improvement.

2. Coordinate efforts among states and health systems to collect and report inventory, regionalize resources, and coordinate their distribution. The best example of this is the seven-state purchasing consortium announced by New York Governor Andrew Cuomo in early May.¹² Unfortunately, since the announcement, there have been few details about whether the states were successful in their effort to reduce prices or to obtain PPE in bulk. Still, hospitals and health systems could join or emulate purchasing collaboratives to allow resources to be better allocated according to need. There are barriers to such collaboratives because the market is currently set up to encourage competition among health systems and hospitals. During the pandemic, however, cooperation has increasingly been favored over competition in science and healthcare delivery. There are also existing hospital purchasing collaboratives (eg, Premier, Inc¹³), which have taken steps to vet suppliers and improve access to PPE, but it is not clear how successful these efforts have been to date.

3. Advocate for strong federal leadership, including support for increased domestic manufacturing; replenishment and maintenance of state and health system stockpiles of PPE, ventilators, and medications; and development of a centrally coordinated PPE allocation and distribution process. While hospitals and health systems may favor remaining as apolitical as possible, the need for a federal response to stabilize the PPE market may be too urgent and necessary to ignore.

As hospitals and health systems prepare for continued surges in COVID-19 cases, they face challenges in providing PPE for frontline clinicians and staff. A federal plan to enhance nimbleness in responding to multifocal, geographic outbreaks and ensure awareness regarding inventory would improve our chances to successfully navigate the next pandemic and optimize the protection of our health workers, patients, and public health. In the absence of such a plan, hospitals should maintain a minimum of 2 weeks’ worth of PPE to prepare for expected regional spikes in COVID-19 cases and should continue to attempt to coordinate efforts among states and health systems to collect and report inventory, regionalize resources, and coordinate their distribution.

When my grandfather, who speaks limited English, was admitted to a hospital following a stroke amid the coronavirus disease 2019 (COVID-19) pandemic, my family was understandably worried. Sure enough, within just hours of his admission, we were told our normally very calm and beloved Nana was experiencing significant agitation and delirium. He did not understand nurses’ efforts to calm him down, became even more confused, and was eventually sedated and placed in physical restraints. Even though my family’s presence might have prevented some or all of this terrible series of events, the hospital’s visiting policies during the wave of COVID-19 admissions meant that we were forced to wait in the parking lot as they transpired. The hospital’s policy at the time only allowed visitors for pediatrics, end-of-life care, or labor, not for patients with delirium or altered mental status. We were given the option to make a video call, but my grandfather’s stroke had almost completely taken away his vision. Instead of sitting by his side, comforting him, providing explanations in voices he knew and a language he understood, we were left imagining how difficult it must be to suddenly wake up in an unfamiliar environment, with strangers speaking a different language, limited vision, and your arms and legs tied. Intellectually, I understood the hospital’s goals to minimize transmission, but spiritually and emotionally, it felt very cruel and very wrong.

The next day, we successfully petitioned administration to make an exception for one visitor. We argued that our presence would allow for removal of the sedation and restraints. The clinical team agreed that video calls were insufficient in his situation; he was allowed a visitor. We decided that it should be my mother. As soon my grandfather heard her familiar voice, there was a dramatic improvement. He immediately became calmer and restraints were no longer necessary. The team was grateful for a better physical exam and my grandfather was more cooperative with physical therapy. A few days later, unfortunately, the hospital let us know that they had reevaluated their position on my mother’s visits and that she posed an unnecessary COVID-19 risk to medical staff and other patients. And as soon as she left, my grandfather was again agitated and confused for the remaining 3 days of his hospitalization. Although we are grateful that his delirium resolved once he returned home, delirium also has the potential to lead to long-term cognitive impairment.¹

The COVID-19 pandemic has required hospitals around the world to make difficult decisions about how to balance minimizing disease transmission with continuing to provide compassionate and high-quality patient care. Of these many dilemmas, developing flexible visitor policies is particularly difficult. Currently, the Centers for Disease Control and Prevention and many state health departments encourage limiting visitation in general but recognize the need for exceptions in special circumstances such as in end-of-life settings or altered mental status.^2-4

At the hospital level, there is substantial variation in visitation policies among hospitals. Near our family home in San Jose, California, one hospital currently allows visitation for pediatric patients, pregnant patients, end-of-life patients, surgical patients, and patients in the emergency department, as well as those with mental disabilities or safety needs.⁵ A mere 10 minutes away, another hospital has implemented a very different policy that allows only one visitor for pregnant patients and in end-of-life settings; there are no exceptions for patients with cognitive or physical disabilities.⁶ Other hospitals in the United States have gone even further, not permitting visitors even for those at the end of life.⁷ These patients are forced to spend their last few moments alone.

From an infection control perspective, there are certainly valid reasons to limit visitation. Even with temperature screenings, any movement into and out of a hospital poses a risk of transmitting disease. Infected but asymptomatic persons are known to transmit the disease. Additionally, hospitals still treat non–COVID-19 patients who are most susceptible to severe illness should they develop COVID-19 infection. Early in the COVID-19 pandemic, limitations in testing capacity, personal protective equipment (PPE), and staffing made it challenging to ensure safe visitation. In many cases, it was almost impossible to mitigate the transmission risk that visitors posed. Because many hospitals did not have the capacity to test all symptomatic patients, they could not reliably limit visits to COVID-19–positive patients. Additionally, without enough PPE for healthcare workers, hospitals could not afford for visitors to use additional PPE.

Now that testing is more readily available and some aspects of the PPE shortages have been addressed, we should not forget that visitation has significant benefits for both patients’ psychological well-being and their overall outcomes.⁸ Putting aside the emotional support that the physical presence of loved ones can offer, a large body of research indicates that allowing visitors can also meaningfully improve other important patient outcomes. Specifically, the presence of visitors is associated with less fear,⁹ reduced delirium,¹⁰ and even faster recovery.⁸ In many cases, family members can also help improve hospital safety surveillance and catch medical errors.¹¹

I saw firsthand how these benefits are particularly true for visitors who are also a patient’s primary caretaker. When my mother visited my grandfather, she was not simply a visitor but instead served as an active member of the care team. In addition to providing emotional comfort, my mother oriented him to his surroundings, successfully encouraged oral intake, and even caught some medication errors. Particularly for patients with cognitive impairment, caretakers know the patient better than anyone on the clinical team, and their absence can negatively affect the quality of care.

As a family member who also has familiarity with the healthcare system, I share hospitals’ concerns about wanting to minimize disease transmission. I recognize that, even with PPE and screenings, there is still a chance that visitors unknowingly spread COVID-19 to others in the hospital. On a personal level, however, it feels inhumane to maintain this policy even when it affects particularly vulnerable patients like my grandfather. As some hospitals are already doing,¹² we can take steps to allow visitors for such patients while minimizing the likelihood of COVID-19 disease transmission from visitors. Arriving visitors can be screened and required to wear PPE. While these measures may not eliminate the risk of COVID-19 transmission from visitors, they will likely reduce it significantly when implemented properly and make possible a more humane experience for all.¹³

Fortunately, my grandfather is now recovering comfortably at home, surrounded by his loved ones. To this day, however, he has not forgotten what it was like to be confused and alone in the hospital after his stroke. Even with loved ones around, a stroke is a profoundly distressing experience. To go through such an experience alone is even worse. Because of our petitioning, my grandfather was at least allowed a visitor for part of his stay. Other patients are not even allowed that. As we plan for the pandemic’s next waves, hospitals should reevaluate their visitor policies to ensure that their most vulnerable patients do not have to suffer alone.

The author sincerely thanks Dr Allan Goroll (Massachusetts General Hospital/Harvard Medical School) for his mentorship and critical review of this manuscript.

The author has nothing to disclose.

When my grandfather, who speaks limited English, was admitted to a hospital following a stroke amid the coronavirus disease 2019 (COVID-19) pandemic, my family was understandably worried. Sure enough, within just hours of his admission, we were told our normally very calm and beloved Nana was experiencing significant agitation and delirium. He did not understand nurses’ efforts to calm him down, became even more confused, and was eventually sedated and placed in physical restraints. Even though my family’s presence might have prevented some or all of this terrible series of events, the hospital’s visiting policies during the wave of COVID-19 admissions meant that we were forced to wait in the parking lot as they transpired. The hospital’s policy at the time only allowed visitors for pediatrics, end-of-life care, or labor, not for patients with delirium or altered mental status. We were given the option to make a video call, but my grandfather’s stroke had almost completely taken away his vision. Instead of sitting by his side, comforting him, providing explanations in voices he knew and a language he understood, we were left imagining how difficult it must be to suddenly wake up in an unfamiliar environment, with strangers speaking a different language, limited vision, and your arms and legs tied. Intellectually, I understood the hospital’s goals to minimize transmission, but spiritually and emotionally, it felt very cruel and very wrong.

The next day, we successfully petitioned administration to make an exception for one visitor. We argued that our presence would allow for removal of the sedation and restraints. The clinical team agreed that video calls were insufficient in his situation; he was allowed a visitor. We decided that it should be my mother. As soon my grandfather heard her familiar voice, there was a dramatic improvement. He immediately became calmer and restraints were no longer necessary. The team was grateful for a better physical exam and my grandfather was more cooperative with physical therapy. A few days later, unfortunately, the hospital let us know that they had reevaluated their position on my mother’s visits and that she posed an unnecessary COVID-19 risk to medical staff and other patients. And as soon as she left, my grandfather was again agitated and confused for the remaining 3 days of his hospitalization. Although we are grateful that his delirium resolved once he returned home, delirium also has the potential to lead to long-term cognitive impairment.¹

The COVID-19 pandemic has required hospitals around the world to make difficult decisions about how to balance minimizing disease transmission with continuing to provide compassionate and high-quality patient care. Of these many dilemmas, developing flexible visitor policies is particularly difficult. Currently, the Centers for Disease Control and Prevention and many state health departments encourage limiting visitation in general but recognize the need for exceptions in special circumstances such as in end-of-life settings or altered mental status.^2-4

At the hospital level, there is substantial variation in visitation policies among hospitals. Near our family home in San Jose, California, one hospital currently allows visitation for pediatric patients, pregnant patients, end-of-life patients, surgical patients, and patients in the emergency department, as well as those with mental disabilities or safety needs.⁵ A mere 10 minutes away, another hospital has implemented a very different policy that allows only one visitor for pregnant patients and in end-of-life settings; there are no exceptions for patients with cognitive or physical disabilities.⁶ Other hospitals in the United States have gone even further, not permitting visitors even for those at the end of life.⁷ These patients are forced to spend their last few moments alone.

From an infection control perspective, there are certainly valid reasons to limit visitation. Even with temperature screenings, any movement into and out of a hospital poses a risk of transmitting disease. Infected but asymptomatic persons are known to transmit the disease. Additionally, hospitals still treat non–COVID-19 patients who are most susceptible to severe illness should they develop COVID-19 infection. Early in the COVID-19 pandemic, limitations in testing capacity, personal protective equipment (PPE), and staffing made it challenging to ensure safe visitation. In many cases, it was almost impossible to mitigate the transmission risk that visitors posed. Because many hospitals did not have the capacity to test all symptomatic patients, they could not reliably limit visits to COVID-19–positive patients. Additionally, without enough PPE for healthcare workers, hospitals could not afford for visitors to use additional PPE.

Now that testing is more readily available and some aspects of the PPE shortages have been addressed, we should not forget that visitation has significant benefits for both patients’ psychological well-being and their overall outcomes.⁸ Putting aside the emotional support that the physical presence of loved ones can offer, a large body of research indicates that allowing visitors can also meaningfully improve other important patient outcomes. Specifically, the presence of visitors is associated with less fear,⁹ reduced delirium,¹⁰ and even faster recovery.⁸ In many cases, family members can also help improve hospital safety surveillance and catch medical errors.¹¹

I saw firsthand how these benefits are particularly true for visitors who are also a patient’s primary caretaker. When my mother visited my grandfather, she was not simply a visitor but instead served as an active member of the care team. In addition to providing emotional comfort, my mother oriented him to his surroundings, successfully encouraged oral intake, and even caught some medication errors. Particularly for patients with cognitive impairment, caretakers know the patient better than anyone on the clinical team, and their absence can negatively affect the quality of care.

As a family member who also has familiarity with the healthcare system, I share hospitals’ concerns about wanting to minimize disease transmission. I recognize that, even with PPE and screenings, there is still a chance that visitors unknowingly spread COVID-19 to others in the hospital. On a personal level, however, it feels inhumane to maintain this policy even when it affects particularly vulnerable patients like my grandfather. As some hospitals are already doing,¹² we can take steps to allow visitors for such patients while minimizing the likelihood of COVID-19 disease transmission from visitors. Arriving visitors can be screened and required to wear PPE. While these measures may not eliminate the risk of COVID-19 transmission from visitors, they will likely reduce it significantly when implemented properly and make possible a more humane experience for all.¹³

Fortunately, my grandfather is now recovering comfortably at home, surrounded by his loved ones. To this day, however, he has not forgotten what it was like to be confused and alone in the hospital after his stroke. Even with loved ones around, a stroke is a profoundly distressing experience. To go through such an experience alone is even worse. Because of our petitioning, my grandfather was at least allowed a visitor for part of his stay. Other patients are not even allowed that. As we plan for the pandemic’s next waves, hospitals should reevaluate their visitor policies to ensure that their most vulnerable patients do not have to suffer alone.

The author sincerely thanks Dr Allan Goroll (Massachusetts General Hospital/Harvard Medical School) for his mentorship and critical review of this manuscript.

The author has nothing to disclose.

When my grandfather, who speaks limited English, was admitted to a hospital following a stroke amid the coronavirus disease 2019 (COVID-19) pandemic, my family was understandably worried. Sure enough, within just hours of his admission, we were told our normally very calm and beloved Nana was experiencing significant agitation and delirium. He did not understand nurses’ efforts to calm him down, became even more confused, and was eventually sedated and placed in physical restraints. Even though my family’s presence might have prevented some or all of this terrible series of events, the hospital’s visiting policies during the wave of COVID-19 admissions meant that we were forced to wait in the parking lot as they transpired. The hospital’s policy at the time only allowed visitors for pediatrics, end-of-life care, or labor, not for patients with delirium or altered mental status. We were given the option to make a video call, but my grandfather’s stroke had almost completely taken away his vision. Instead of sitting by his side, comforting him, providing explanations in voices he knew and a language he understood, we were left imagining how difficult it must be to suddenly wake up in an unfamiliar environment, with strangers speaking a different language, limited vision, and your arms and legs tied. Intellectually, I understood the hospital’s goals to minimize transmission, but spiritually and emotionally, it felt very cruel and very wrong.

The next day, we successfully petitioned administration to make an exception for one visitor. We argued that our presence would allow for removal of the sedation and restraints. The clinical team agreed that video calls were insufficient in his situation; he was allowed a visitor. We decided that it should be my mother. As soon my grandfather heard her familiar voice, there was a dramatic improvement. He immediately became calmer and restraints were no longer necessary. The team was grateful for a better physical exam and my grandfather was more cooperative with physical therapy. A few days later, unfortunately, the hospital let us know that they had reevaluated their position on my mother’s visits and that she posed an unnecessary COVID-19 risk to medical staff and other patients. And as soon as she left, my grandfather was again agitated and confused for the remaining 3 days of his hospitalization. Although we are grateful that his delirium resolved once he returned home, delirium also has the potential to lead to long-term cognitive impairment.¹

The COVID-19 pandemic has required hospitals around the world to make difficult decisions about how to balance minimizing disease transmission with continuing to provide compassionate and high-quality patient care. Of these many dilemmas, developing flexible visitor policies is particularly difficult. Currently, the Centers for Disease Control and Prevention and many state health departments encourage limiting visitation in general but recognize the need for exceptions in special circumstances such as in end-of-life settings or altered mental status.^2-4

At the hospital level, there is substantial variation in visitation policies among hospitals. Near our family home in San Jose, California, one hospital currently allows visitation for pediatric patients, pregnant patients, end-of-life patients, surgical patients, and patients in the emergency department, as well as those with mental disabilities or safety needs.⁵ A mere 10 minutes away, another hospital has implemented a very different policy that allows only one visitor for pregnant patients and in end-of-life settings; there are no exceptions for patients with cognitive or physical disabilities.⁶ Other hospitals in the United States have gone even further, not permitting visitors even for those at the end of life.⁷ These patients are forced to spend their last few moments alone.

From an infection control perspective, there are certainly valid reasons to limit visitation. Even with temperature screenings, any movement into and out of a hospital poses a risk of transmitting disease. Infected but asymptomatic persons are known to transmit the disease. Additionally, hospitals still treat non–COVID-19 patients who are most susceptible to severe illness should they develop COVID-19 infection. Early in the COVID-19 pandemic, limitations in testing capacity, personal protective equipment (PPE), and staffing made it challenging to ensure safe visitation. In many cases, it was almost impossible to mitigate the transmission risk that visitors posed. Because many hospitals did not have the capacity to test all symptomatic patients, they could not reliably limit visits to COVID-19–positive patients. Additionally, without enough PPE for healthcare workers, hospitals could not afford for visitors to use additional PPE.

Now that testing is more readily available and some aspects of the PPE shortages have been addressed, we should not forget that visitation has significant benefits for both patients’ psychological well-being and their overall outcomes.⁸ Putting aside the emotional support that the physical presence of loved ones can offer, a large body of research indicates that allowing visitors can also meaningfully improve other important patient outcomes. Specifically, the presence of visitors is associated with less fear,⁹ reduced delirium,¹⁰ and even faster recovery.⁸ In many cases, family members can also help improve hospital safety surveillance and catch medical errors.¹¹

I saw firsthand how these benefits are particularly true for visitors who are also a patient’s primary caretaker. When my mother visited my grandfather, she was not simply a visitor but instead served as an active member of the care team. In addition to providing emotional comfort, my mother oriented him to his surroundings, successfully encouraged oral intake, and even caught some medication errors. Particularly for patients with cognitive impairment, caretakers know the patient better than anyone on the clinical team, and their absence can negatively affect the quality of care.

As a family member who also has familiarity with the healthcare system, I share hospitals’ concerns about wanting to minimize disease transmission. I recognize that, even with PPE and screenings, there is still a chance that visitors unknowingly spread COVID-19 to others in the hospital. On a personal level, however, it feels inhumane to maintain this policy even when it affects particularly vulnerable patients like my grandfather. As some hospitals are already doing,¹² we can take steps to allow visitors for such patients while minimizing the likelihood of COVID-19 disease transmission from visitors. Arriving visitors can be screened and required to wear PPE. While these measures may not eliminate the risk of COVID-19 transmission from visitors, they will likely reduce it significantly when implemented properly and make possible a more humane experience for all.¹³

Fortunately, my grandfather is now recovering comfortably at home, surrounded by his loved ones. To this day, however, he has not forgotten what it was like to be confused and alone in the hospital after his stroke. Even with loved ones around, a stroke is a profoundly distressing experience. To go through such an experience alone is even worse. Because of our petitioning, my grandfather was at least allowed a visitor for part of his stay. Other patients are not even allowed that. As we plan for the pandemic’s next waves, hospitals should reevaluate their visitor policies to ensure that their most vulnerable patients do not have to suffer alone.

The author sincerely thanks Dr Allan Goroll (Massachusetts General Hospital/Harvard Medical School) for his mentorship and critical review of this manuscript.

The author has nothing to disclose.

The coronavirus disease of 2019 (COVID-19) pandemic has created innumerable challenges on scales both global and personal while straining health systems and their personnel. Hospitalists and hospital medicine groups are experiencing unique burdens as they confront the pandemic on the frontlines. Hospital medicine groups are being challenged by the rapid operational changes necessary in preparing for and caring for patients with COVID-19. These challenges include drafting new diagnostic and management algorithms, establishing and enacting policies on personal protective equipment (PPE) and patient and provider testing, modifying staffing protocols including deploying staff to new roles or integrating non-hospitalists into hospital medicine roles, and developing capacity for patient surges¹—all in the setting of uncertainty about how the pandemic may affect individual hospitals or health systems and how long these repercussions may last. In this perspective, we describe key lessons we have learned in leading our hospital medicine group during the COVID-19 pandemic: how to apply emotional intelligence to proactively address the emotional effects of the crisis.

In the early days of the COVID-19 pandemic, the evolving knowledge of the disease process, changing national and local public health guidelines, and instability of the PPE supply chain necessitated rapid change. This pace no longer allowed for our typical time frame of weeks to months for implementation of large-scale operational changes; instead, it demanded adaptation in hours to days. We operated under a strategy of developing new workflows and policies that were logical and reflected the best available information at the time.

For instance, our hospital medicine service cared for some of the earliest-identified COVID-19 patients in the United States in early February 2020. Our initial operational plan for caring for patients with COVID-19 involved grouping these patients on a limited number of direct-care hospitalist teams. The advantages of this approach, which benefitted from low numbers of initial patients, were clear: consolidation of clinical and operational knowledge (including optimal PPE practices) in a few individuals, streamlining communication with infectious diseases specialists and public health departments, and requiring change on only a couple of teams while allowing others to continue their usual workflow. However, we soon learned that providers caring for COVID-19 patients were experiencing an onslaught of negative emotions: fear of contracting the virus themselves or carrying it home to infect loved ones, anxiety of not understanding the clinical disease or having treatments to offer, resentment of having been randomly assigned to the team that would care for these patients, and loneliness of being a sole provider experiencing these emotions. We found ourselves in the position of managing these emotional responses reactively.

To reduce the distress experienced by our hospitalists and to lead more effectively, we realized the need to proactively address the emotional effects that the pandemic was having. Several authors who have written about valuable leadership lessons during this time have noted the importance of acknowledging the emotional tolls of such a crisis and creating venues for hospitalists to share their experiences.^1-4 However, solely adding “wellness” as a checklist item for leaders to address fails to capture the nuances of the complex human emotions that hospitalists may endure at this time and how these emotions influence frontline hospitalists’ responses to operational changes. It is critically important for hospital medicine leaders to employ emotional intelligence, defined as “the ability to monitor one’s own and others’ feelings and emotions, to discriminate among them and to use this information to guide one’s thinking and actions.”^5-7 Integrating emotional intelligence allows hospital medicine leaders to anticipate, identify, articulate, and manage the emotional responses to necessary changes and stresses that occur during a crisis such as the COVID-19 pandemic.

As we applied principles of emotional intelligence to our leadership response to the COVID crisis, we found the following seven techniques effective (Appendix Table):

1. ASK. Leaders should ask individual hospitalists “How are you feeling?” instead of “How are you doing?” or “How can I help?” This question may feel too intimate for some, or leaders may worry that the question feels patronizing; however, in our experience, hospitalists respond positively to this prompt, welcome the opportunity to communicate their feelings, and value being heard. Moreover, when hospitalists feel overwhelmed, they may not be able to determine what help they do or do not need. By understanding the emotions of frontline hospitalists, leaders may be better able to address those emotions directly, find solutions to problems, and anticipate reactions to future policies.⁴

2. SHARE. Leaders should model what they ask of frontline hospitalists and share their own feelings, even if they are experiencing mixed or negative emotions. For instance, a leader who is feeling saddened about the death of a patient can begin a meeting by sharing this sentiment. By allowing themselves to display vulnerability, leaders demonstrate courage and promote a culture of openness, honesty, and mutual trust.

3. INITIATE. Leaders should embrace difficult conversations and be transparent about uncertainty, although they may not have the answers and may need to take local responsibility for consequences of decisions made externally, such as those made by the health system or government. Confronting difficult discussions and being transparent about “unknowns” provides acknowledgement, reassurance, and shared experience that expresses to the hospitalist group that, while the future may be unsettled, they will face it together.

4. ANTICIPATE. Leaders should anticipate the emotional responses to operational changes while designing them and rolling them out. While negative emotions may heavily outweigh positive emotions in times of crisis, we have also found that harnessing positive emotions when designing operational initiatives can assist with successful implementation. For example, by surveying our hospitalists, we found that many felt enthusiastic about caring for patients with COVID-19, curious about new skill sets, and passionate about helping in a time of crisis. By generating a list of these hospitalists up front, we were able to preferentially staff COVID-19 teams with providers who were eager to care for those patients and, thereby, minimize anxiety among those who were more apprehensive.

5. ENCOURAGE. Leaders should provide time and space (including virtually) for hospitalists to discuss their emotions.⁸ We found that creating multiple layers of opportunity for expression allows for engagement with a wider range of hospitalists, some of whom may be reluctant to share feelings openly or to a group, whereas others may enjoy the opportunity to reveal their feelings publicly. These varied venues for emotional expression may range from brief individual check-ins to open “office hours” to dedicated meetings such as “Hospitalist Town Halls.” For instance, spending the first few minutes of a meeting with a smaller group by encouraging each participant to share something personal can build community and mutual understanding, as well as cue leaders in to where participants may be on the emotional landscape.

6. NURTURE. Beyond inviting the expression of emotions, leaders should ensure that hospitalists have access to more formal systems of support, especially for hospitalists who may be experiencing more intense negative emotions. Support may be provided through unit- or team-based debriefing sessions, health-system sponsored support programs, or individual counseling sessions.^4,8

7. APPRECIATE. Leaders should deliberately foster gratitude by sincerely and frequently expressing their appreciation. Because expressing gratitude builds resiliency,⁹ cultivating a culture of gratitude may bolster resilience in the entire hospital medicine group. Opportunities for thankfulness abound as hospitalists volunteer for extra shifts, cover for ill colleagues, participate in new working groups and task forces, and sacrifice their personal safety on the front lines. We often incorporate statements of appreciation into one-on-one conversations with hospitalists, during operational and divisional meetings, and in email. We also built gratitude expressions into the daily work on the Respiratory Isolation Unit at our hospital via daily interdisciplinary huddles for frontline providers to share their experiences and emotions. During huddles, providers are asked to pair negative emotions with suggestions for improvement and to share a moment of gratitude. This helps to engender a spirit of camaraderie, shared mission, and collective optimism.

Hospitalists are experiencing a wide range of emotions related to the COVID-19 pandemic. Hospital medicine leaders must have strategies to understand the emotions providers are experiencing. Being aware of and acknowledging these emotions up front can help leaders plan and implement the operational changes necessary to manage the crisis. Because our health system and city have fortunately been spared the worst of the pandemic so far without large volumes of patients with COVID-19, we recognize that the strategies above may be challenging for leaders in overwhelmed health systems. However, we hope that leaders at all levels can apply the lessons we have learned: to ask hospitalists how they are feeling, share their own feelings, initiate difficult conversations when needed, anticipate the emotional effects of operational changes, encourage expressions of emotion in multiple venues, nurture hospitalists who need more formal support, and appreciate frontline hospitalists. While the emotional needs of hospitalists will undoubtedly change over time as the pandemic evolves, we suspect that these strategies will continue to be important over the coming weeks, months, and longer as we settle into the postpandemic world.

The coronavirus disease of 2019 (COVID-19) pandemic has created innumerable challenges on scales both global and personal while straining health systems and their personnel. Hospitalists and hospital medicine groups are experiencing unique burdens as they confront the pandemic on the frontlines. Hospital medicine groups are being challenged by the rapid operational changes necessary in preparing for and caring for patients with COVID-19. These challenges include drafting new diagnostic and management algorithms, establishing and enacting policies on personal protective equipment (PPE) and patient and provider testing, modifying staffing protocols including deploying staff to new roles or integrating non-hospitalists into hospital medicine roles, and developing capacity for patient surges¹—all in the setting of uncertainty about how the pandemic may affect individual hospitals or health systems and how long these repercussions may last. In this perspective, we describe key lessons we have learned in leading our hospital medicine group during the COVID-19 pandemic: how to apply emotional intelligence to proactively address the emotional effects of the crisis.

In the early days of the COVID-19 pandemic, the evolving knowledge of the disease process, changing national and local public health guidelines, and instability of the PPE supply chain necessitated rapid change. This pace no longer allowed for our typical time frame of weeks to months for implementation of large-scale operational changes; instead, it demanded adaptation in hours to days. We operated under a strategy of developing new workflows and policies that were logical and reflected the best available information at the time.

For instance, our hospital medicine service cared for some of the earliest-identified COVID-19 patients in the United States in early February 2020. Our initial operational plan for caring for patients with COVID-19 involved grouping these patients on a limited number of direct-care hospitalist teams. The advantages of this approach, which benefitted from low numbers of initial patients, were clear: consolidation of clinical and operational knowledge (including optimal PPE practices) in a few individuals, streamlining communication with infectious diseases specialists and public health departments, and requiring change on only a couple of teams while allowing others to continue their usual workflow. However, we soon learned that providers caring for COVID-19 patients were experiencing an onslaught of negative emotions: fear of contracting the virus themselves or carrying it home to infect loved ones, anxiety of not understanding the clinical disease or having treatments to offer, resentment of having been randomly assigned to the team that would care for these patients, and loneliness of being a sole provider experiencing these emotions. We found ourselves in the position of managing these emotional responses reactively.

To reduce the distress experienced by our hospitalists and to lead more effectively, we realized the need to proactively address the emotional effects that the pandemic was having. Several authors who have written about valuable leadership lessons during this time have noted the importance of acknowledging the emotional tolls of such a crisis and creating venues for hospitalists to share their experiences.^1-4 However, solely adding “wellness” as a checklist item for leaders to address fails to capture the nuances of the complex human emotions that hospitalists may endure at this time and how these emotions influence frontline hospitalists’ responses to operational changes. It is critically important for hospital medicine leaders to employ emotional intelligence, defined as “the ability to monitor one’s own and others’ feelings and emotions, to discriminate among them and to use this information to guide one’s thinking and actions.”^5-7 Integrating emotional intelligence allows hospital medicine leaders to anticipate, identify, articulate, and manage the emotional responses to necessary changes and stresses that occur during a crisis such as the COVID-19 pandemic.

As we applied principles of emotional intelligence to our leadership response to the COVID crisis, we found the following seven techniques effective (Appendix Table):

1. ASK. Leaders should ask individual hospitalists “How are you feeling?” instead of “How are you doing?” or “How can I help?” This question may feel too intimate for some, or leaders may worry that the question feels patronizing; however, in our experience, hospitalists respond positively to this prompt, welcome the opportunity to communicate their feelings, and value being heard. Moreover, when hospitalists feel overwhelmed, they may not be able to determine what help they do or do not need. By understanding the emotions of frontline hospitalists, leaders may be better able to address those emotions directly, find solutions to problems, and anticipate reactions to future policies.⁴

2. SHARE. Leaders should model what they ask of frontline hospitalists and share their own feelings, even if they are experiencing mixed or negative emotions. For instance, a leader who is feeling saddened about the death of a patient can begin a meeting by sharing this sentiment. By allowing themselves to display vulnerability, leaders demonstrate courage and promote a culture of openness, honesty, and mutual trust.

3. INITIATE. Leaders should embrace difficult conversations and be transparent about uncertainty, although they may not have the answers and may need to take local responsibility for consequences of decisions made externally, such as those made by the health system or government. Confronting difficult discussions and being transparent about “unknowns” provides acknowledgement, reassurance, and shared experience that expresses to the hospitalist group that, while the future may be unsettled, they will face it together.

4. ANTICIPATE. Leaders should anticipate the emotional responses to operational changes while designing them and rolling them out. While negative emotions may heavily outweigh positive emotions in times of crisis, we have also found that harnessing positive emotions when designing operational initiatives can assist with successful implementation. For example, by surveying our hospitalists, we found that many felt enthusiastic about caring for patients with COVID-19, curious about new skill sets, and passionate about helping in a time of crisis. By generating a list of these hospitalists up front, we were able to preferentially staff COVID-19 teams with providers who were eager to care for those patients and, thereby, minimize anxiety among those who were more apprehensive.

5. ENCOURAGE. Leaders should provide time and space (including virtually) for hospitalists to discuss their emotions.⁸ We found that creating multiple layers of opportunity for expression allows for engagement with a wider range of hospitalists, some of whom may be reluctant to share feelings openly or to a group, whereas others may enjoy the opportunity to reveal their feelings publicly. These varied venues for emotional expression may range from brief individual check-ins to open “office hours” to dedicated meetings such as “Hospitalist Town Halls.” For instance, spending the first few minutes of a meeting with a smaller group by encouraging each participant to share something personal can build community and mutual understanding, as well as cue leaders in to where participants may be on the emotional landscape.

6. NURTURE. Beyond inviting the expression of emotions, leaders should ensure that hospitalists have access to more formal systems of support, especially for hospitalists who may be experiencing more intense negative emotions. Support may be provided through unit- or team-based debriefing sessions, health-system sponsored support programs, or individual counseling sessions.^4,8

7. APPRECIATE. Leaders should deliberately foster gratitude by sincerely and frequently expressing their appreciation. Because expressing gratitude builds resiliency,⁹ cultivating a culture of gratitude may bolster resilience in the entire hospital medicine group. Opportunities for thankfulness abound as hospitalists volunteer for extra shifts, cover for ill colleagues, participate in new working groups and task forces, and sacrifice their personal safety on the front lines. We often incorporate statements of appreciation into one-on-one conversations with hospitalists, during operational and divisional meetings, and in email. We also built gratitude expressions into the daily work on the Respiratory Isolation Unit at our hospital via daily interdisciplinary huddles for frontline providers to share their experiences and emotions. During huddles, providers are asked to pair negative emotions with suggestions for improvement and to share a moment of gratitude. This helps to engender a spirit of camaraderie, shared mission, and collective optimism.

Hospitalists are experiencing a wide range of emotions related to the COVID-19 pandemic. Hospital medicine leaders must have strategies to understand the emotions providers are experiencing. Being aware of and acknowledging these emotions up front can help leaders plan and implement the operational changes necessary to manage the crisis. Because our health system and city have fortunately been spared the worst of the pandemic so far without large volumes of patients with COVID-19, we recognize that the strategies above may be challenging for leaders in overwhelmed health systems. However, we hope that leaders at all levels can apply the lessons we have learned: to ask hospitalists how they are feeling, share their own feelings, initiate difficult conversations when needed, anticipate the emotional effects of operational changes, encourage expressions of emotion in multiple venues, nurture hospitalists who need more formal support, and appreciate frontline hospitalists. While the emotional needs of hospitalists will undoubtedly change over time as the pandemic evolves, we suspect that these strategies will continue to be important over the coming weeks, months, and longer as we settle into the postpandemic world.

The coronavirus disease of 2019 (COVID-19) pandemic has created innumerable challenges on scales both global and personal while straining health systems and their personnel. Hospitalists and hospital medicine groups are experiencing unique burdens as they confront the pandemic on the frontlines. Hospital medicine groups are being challenged by the rapid operational changes necessary in preparing for and caring for patients with COVID-19. These challenges include drafting new diagnostic and management algorithms, establishing and enacting policies on personal protective equipment (PPE) and patient and provider testing, modifying staffing protocols including deploying staff to new roles or integrating non-hospitalists into hospital medicine roles, and developing capacity for patient surges¹—all in the setting of uncertainty about how the pandemic may affect individual hospitals or health systems and how long these repercussions may last. In this perspective, we describe key lessons we have learned in leading our hospital medicine group during the COVID-19 pandemic: how to apply emotional intelligence to proactively address the emotional effects of the crisis.

In the early days of the COVID-19 pandemic, the evolving knowledge of the disease process, changing national and local public health guidelines, and instability of the PPE supply chain necessitated rapid change. This pace no longer allowed for our typical time frame of weeks to months for implementation of large-scale operational changes; instead, it demanded adaptation in hours to days. We operated under a strategy of developing new workflows and policies that were logical and reflected the best available information at the time.

For instance, our hospital medicine service cared for some of the earliest-identified COVID-19 patients in the United States in early February 2020. Our initial operational plan for caring for patients with COVID-19 involved grouping these patients on a limited number of direct-care hospitalist teams. The advantages of this approach, which benefitted from low numbers of initial patients, were clear: consolidation of clinical and operational knowledge (including optimal PPE practices) in a few individuals, streamlining communication with infectious diseases specialists and public health departments, and requiring change on only a couple of teams while allowing others to continue their usual workflow. However, we soon learned that providers caring for COVID-19 patients were experiencing an onslaught of negative emotions: fear of contracting the virus themselves or carrying it home to infect loved ones, anxiety of not understanding the clinical disease or having treatments to offer, resentment of having been randomly assigned to the team that would care for these patients, and loneliness of being a sole provider experiencing these emotions. We found ourselves in the position of managing these emotional responses reactively.

To reduce the distress experienced by our hospitalists and to lead more effectively, we realized the need to proactively address the emotional effects that the pandemic was having. Several authors who have written about valuable leadership lessons during this time have noted the importance of acknowledging the emotional tolls of such a crisis and creating venues for hospitalists to share their experiences.^1-4 However, solely adding “wellness” as a checklist item for leaders to address fails to capture the nuances of the complex human emotions that hospitalists may endure at this time and how these emotions influence frontline hospitalists’ responses to operational changes. It is critically important for hospital medicine leaders to employ emotional intelligence, defined as “the ability to monitor one’s own and others’ feelings and emotions, to discriminate among them and to use this information to guide one’s thinking and actions.”^5-7 Integrating emotional intelligence allows hospital medicine leaders to anticipate, identify, articulate, and manage the emotional responses to necessary changes and stresses that occur during a crisis such as the COVID-19 pandemic.

As we applied principles of emotional intelligence to our leadership response to the COVID crisis, we found the following seven techniques effective (Appendix Table):

1. ASK. Leaders should ask individual hospitalists “How are you feeling?” instead of “How are you doing?” or “How can I help?” This question may feel too intimate for some, or leaders may worry that the question feels patronizing; however, in our experience, hospitalists respond positively to this prompt, welcome the opportunity to communicate their feelings, and value being heard. Moreover, when hospitalists feel overwhelmed, they may not be able to determine what help they do or do not need. By understanding the emotions of frontline hospitalists, leaders may be better able to address those emotions directly, find solutions to problems, and anticipate reactions to future policies.⁴

2. SHARE. Leaders should model what they ask of frontline hospitalists and share their own feelings, even if they are experiencing mixed or negative emotions. For instance, a leader who is feeling saddened about the death of a patient can begin a meeting by sharing this sentiment. By allowing themselves to display vulnerability, leaders demonstrate courage and promote a culture of openness, honesty, and mutual trust.

3. INITIATE. Leaders should embrace difficult conversations and be transparent about uncertainty, although they may not have the answers and may need to take local responsibility for consequences of decisions made externally, such as those made by the health system or government. Confronting difficult discussions and being transparent about “unknowns” provides acknowledgement, reassurance, and shared experience that expresses to the hospitalist group that, while the future may be unsettled, they will face it together.

4. ANTICIPATE. Leaders should anticipate the emotional responses to operational changes while designing them and rolling them out. While negative emotions may heavily outweigh positive emotions in times of crisis, we have also found that harnessing positive emotions when designing operational initiatives can assist with successful implementation. For example, by surveying our hospitalists, we found that many felt enthusiastic about caring for patients with COVID-19, curious about new skill sets, and passionate about helping in a time of crisis. By generating a list of these hospitalists up front, we were able to preferentially staff COVID-19 teams with providers who were eager to care for those patients and, thereby, minimize anxiety among those who were more apprehensive.

5. ENCOURAGE. Leaders should provide time and space (including virtually) for hospitalists to discuss their emotions.⁸ We found that creating multiple layers of opportunity for expression allows for engagement with a wider range of hospitalists, some of whom may be reluctant to share feelings openly or to a group, whereas others may enjoy the opportunity to reveal their feelings publicly. These varied venues for emotional expression may range from brief individual check-ins to open “office hours” to dedicated meetings such as “Hospitalist Town Halls.” For instance, spending the first few minutes of a meeting with a smaller group by encouraging each participant to share something personal can build community and mutual understanding, as well as cue leaders in to where participants may be on the emotional landscape.

6. NURTURE. Beyond inviting the expression of emotions, leaders should ensure that hospitalists have access to more formal systems of support, especially for hospitalists who may be experiencing more intense negative emotions. Support may be provided through unit- or team-based debriefing sessions, health-system sponsored support programs, or individual counseling sessions.^4,8

7. APPRECIATE. Leaders should deliberately foster gratitude by sincerely and frequently expressing their appreciation. Because expressing gratitude builds resiliency,⁹ cultivating a culture of gratitude may bolster resilience in the entire hospital medicine group. Opportunities for thankfulness abound as hospitalists volunteer for extra shifts, cover for ill colleagues, participate in new working groups and task forces, and sacrifice their personal safety on the front lines. We often incorporate statements of appreciation into one-on-one conversations with hospitalists, during operational and divisional meetings, and in email. We also built gratitude expressions into the daily work on the Respiratory Isolation Unit at our hospital via daily interdisciplinary huddles for frontline providers to share their experiences and emotions. During huddles, providers are asked to pair negative emotions with suggestions for improvement and to share a moment of gratitude. This helps to engender a spirit of camaraderie, shared mission, and collective optimism.

Hospitalists are experiencing a wide range of emotions related to the COVID-19 pandemic. Hospital medicine leaders must have strategies to understand the emotions providers are experiencing. Being aware of and acknowledging these emotions up front can help leaders plan and implement the operational changes necessary to manage the crisis. Because our health system and city have fortunately been spared the worst of the pandemic so far without large volumes of patients with COVID-19, we recognize that the strategies above may be challenging for leaders in overwhelmed health systems. However, we hope that leaders at all levels can apply the lessons we have learned: to ask hospitalists how they are feeling, share their own feelings, initiate difficult conversations when needed, anticipate the emotional effects of operational changes, encourage expressions of emotion in multiple venues, nurture hospitalists who need more formal support, and appreciate frontline hospitalists. While the emotional needs of hospitalists will undoubtedly change over time as the pandemic evolves, we suspect that these strategies will continue to be important over the coming weeks, months, and longer as we settle into the postpandemic world.

Recent experiences in the transportation industry highlight the importance of getting right the regulation of decision-support systems in high-stakes environments. Two tragic plane crashes resulted in 346 deaths and were deemed, in part, to be related to a cockpit alert system that overwhelmed pilots with multiple notifications.¹ Similarly, a driverless car struck and killed a pedestrian in the street, in part because the car was not programmed to look for humans outside of a crosswalk.² These two bellwether events offer poignant lessons for the healthcare industry in which human lives also depend on decision-support systems.

Clinical decision-support (CDS) systems are computerized applications, often embedded in an electronic health record (EHR), that provide information to clinicians to inform care. Although CDS systems have been used for many years,³ they have never been subjected to any enforcement of formal testing requirements. However, a draft guidance document released in 2019 from the Food and Drug Administration (FDA) outlined new directions for the regulation of CDS systems.⁴ Although the FDA has thus far focused regulatory efforts on predictive systems developed by private manufacturers,^5,6 this new document provides examples of software that would require regulation for CDS systems that hospitals are already using. Thus, this new guidance raises critical questions—will hospitals themselves be evaluated like private manufacturers, be exempted from federal regulation, or require their own specialized regulation? The FDA has not yet clarified its approach to hospitals or hospital-developed CDS systems, which leaves open numerous possibilities in a rapidly evolving regulatory environment.

Although the FDA has officially regulated CDS systems under section 201(h) of the Federal Food, Drug, and Cosmetic Act (1938), only recently has the FDA begun to sketch the shape of its regulatory efforts. This trend to actually regulate CDS systems began with the 21st Century Cures Act (2016) that amended the definition of software systems that qualify as medical devices and outlined criteria under which a system may be exempt from FDA oversight. For example, regulation would not apply to systems that support “population health” or a “healthy lifestyle” or to ones that qualify as “electronic patient records” as long as they do not “interpret or analyze” data within them.⁷ Following the rapid proliferation of many machine learning and other predictive technologies with medical applications, the FDA began the voluntary Digital Health Software Precertification (Pre-Cert) Program in 2017. Through this program, the FDA selected nine companies from more than 100 applicants and certified them across five domains of excellence. Notably, the Pre-Cert Program currently allows for certification of software manufacturers themselves and does not approve or test actual software devices directly. This regulatory pathway will eventually allow manufacturers to apply under a modified premarket review process for individual software as a medical device (SaMD) that use artificial intelligence (AI) and machine learning. In the meantime, however, many hospitals have developed and deployed their own predictive CDS systems that cross the boundaries into the FDA’s purview and, indeed, do “interpret or analyze” data for real-time EHR alerts, population health management, and other applications.

Regulatory oversight for hospitals could provide quality or safety standards where currently there are none. However, such regulations could also interfere with existing local care practices, hinder rapid development of new CDS systems, and may be perceived as interfering in hospital operations. With the current enthusiasm for AI-based technologies and the concurrent lack of evidence to suggest their effectiveness in practice, regulation could also prompt necessary scrutiny of potential harms of CDS systems, an area with even less evidence. At the same time, CDS developers—private or hospital based—may be able to avoid regulation for some devices with well-placed disclaimers about the intended use of the CDS, one of the FDA criteria for determining the degree of oversight. If the FDA were to regulate hospitals or hospital-developed CDS systems, there are several unanswered questions to consider so that such regulations have their intended impact.

First, does the FDA intend to regulate hospitals and hospital-developed software at all? The framework for determining whether a CDS system will be regulated depends on the severity of the clinical scenario, the ability to independently evaluate the model output, and the intended user (Table). Notably, many types of CDS systems that would require regulation under this framework are already commonplace. For example, the FDA intends to regulate software that “identifies patients who may exhibit signs of opioid addiction,” a scenario similar to prediction models already developed at academic hospitals.⁸ The FDA also plans to regulate a software device even if it is not a CDS system if it is “intended to generate an alarm or an alert to notify a caregiver of a life-threatening condition, such as stroke, and the caregiver relies primarily on this alarm or alert to make a treatment decision.” Although there are no published reports of stroke-specific early warning systems in use, analogous nonspecific and sepsis-specific early warning systems to prompt urgent clinical care have been deployed by hospitals directly⁹ and developed for embedding in commercial EHRs.¹⁰ Hospitals need clarification on the FDA’s regulatory intentions for such CDS systems. FDA regulation of hospitals and hospital-developed CDS systems would fill a critical oversight need and potentially strengthen processes to improve safety and effectiveness. But burdensome regulations may also restrain hospitals from tackling complex problems in medicine for which they are uniquely suited.

Such a regulatory environment may be especially prohibitive for safety-net hospitals that could find themselves at a disadvantage in developing their own CDS systems relative to large academic medical centers that are typically endowed with greater resources. Additionally, CDS systems developed at academic medical centers may not generalize well to populations in the community setting, which could further deepen disparities in access to cutting-edge technologies. For example, racial bias in treatment and referral patterns could bias training labels for CDS systems focused on population health management.¹¹ Similarly, the composition of patient skin color in one population may distort predictions of a model in another with a different distribution of skin color, even when the primary outcome of a prediction model is gender.¹² Additional regulatory steps may apply for models that are adapted to new populations or recalibrated across locations and time.¹³ Until there is more data on the clinical impact of such CDS systems, it is unknown how potential differences in evaluation and approval would actually affect clinical outcomes.

Second, would hospitals be eligible for the Pre-Cert program, and if so, would they be held to the same standards as a private technology manufacturer? The domains of excellence required for precertification approval such as “patient safety,” “clinical responsibility,” and “proactive culture” are aligned with the efforts of hospitals that are already overseen and accredited by organizations like the Joint Commission on Accreditation of Healthcare Organizations and the American Nurses Credentialing Center. There is limited motivation for the FDA to be in the business of regulating these aspects of hospital functions. However, while domains like “product quality” and “cybersecurity” may be less familiar to some hospitals, these existing credentialing bodies may be better suited than the FDA to set and enforce standards for hospitals. In contrast, private manufacturers may have deep expertise in these latter domains. Therefore, as with public-private partnerships for the development of predictive radiology applications,¹⁴ synergies between hospitals and manufacturers may also prove useful for obtaining approvals in a competitive marketplace. Simultaneously, such collaborations would continue to raise questions about conflicts of interest and data privacy.

Finally, regardless of how the FDA will regulate hospitals, what will become of predictive CDS systems that fall outside of the FDA’s scope? Hospitals will continue to find themselves in the position of self-regulation without clear guidance. Although the FDA suggests that developers of unregulated CDS systems still follow best practices for software validation and cybersecurity, existing guidance documents in these domains do not cover the full range of concerns relevant to the development, deployment, and oversight of AI-based CDS systems in the clinical domain. Nor do most hospitals have the infrastructure or expertise to oversee their own CDS systems. Disparate recommendations for development, training, and oversight of AI-based medical systems have emerged but have yet to be endorsed by a federal regulatory body or become part of the hospital accreditation process.¹⁵ Optimal local oversight would require a collaboration between clinical experts, hospital operations leaders, statisticians, data scientists, and ethics experts to ensure effectiveness, safety, and fairness.

Hospitals will remain at the forefront of developing and implementing predictive CDS systems. The proposed FDA regulatory framework would mark an important step toward realizing benefit from such systems, but the FDA needs to clarify the requirements for hospitals and hospital-developed CDS systems to ensure reasonable standards that account for their differences from private software manufacturers. Should the FDA choose to focus regulation on private manufacturers only, hospitals leaders may both feel more empowered to develop their own local CDS tools and feel more comfortable buying CDS systems from vendors that have been precertified. This strategy would provide an optimal balance of assurance and flexibility while maintaining quality standards that ultimately improve patient care.

Recent experiences in the transportation industry highlight the importance of getting right the regulation of decision-support systems in high-stakes environments. Two tragic plane crashes resulted in 346 deaths and were deemed, in part, to be related to a cockpit alert system that overwhelmed pilots with multiple notifications.¹ Similarly, a driverless car struck and killed a pedestrian in the street, in part because the car was not programmed to look for humans outside of a crosswalk.² These two bellwether events offer poignant lessons for the healthcare industry in which human lives also depend on decision-support systems.

Clinical decision-support (CDS) systems are computerized applications, often embedded in an electronic health record (EHR), that provide information to clinicians to inform care. Although CDS systems have been used for many years,³ they have never been subjected to any enforcement of formal testing requirements. However, a draft guidance document released in 2019 from the Food and Drug Administration (FDA) outlined new directions for the regulation of CDS systems.⁴ Although the FDA has thus far focused regulatory efforts on predictive systems developed by private manufacturers,^5,6 this new document provides examples of software that would require regulation for CDS systems that hospitals are already using. Thus, this new guidance raises critical questions—will hospitals themselves be evaluated like private manufacturers, be exempted from federal regulation, or require their own specialized regulation? The FDA has not yet clarified its approach to hospitals or hospital-developed CDS systems, which leaves open numerous possibilities in a rapidly evolving regulatory environment.

Although the FDA has officially regulated CDS systems under section 201(h) of the Federal Food, Drug, and Cosmetic Act (1938), only recently has the FDA begun to sketch the shape of its regulatory efforts. This trend to actually regulate CDS systems began with the 21st Century Cures Act (2016) that amended the definition of software systems that qualify as medical devices and outlined criteria under which a system may be exempt from FDA oversight. For example, regulation would not apply to systems that support “population health” or a “healthy lifestyle” or to ones that qualify as “electronic patient records” as long as they do not “interpret or analyze” data within them.⁷ Following the rapid proliferation of many machine learning and other predictive technologies with medical applications, the FDA began the voluntary Digital Health Software Precertification (Pre-Cert) Program in 2017. Through this program, the FDA selected nine companies from more than 100 applicants and certified them across five domains of excellence. Notably, the Pre-Cert Program currently allows for certification of software manufacturers themselves and does not approve or test actual software devices directly. This regulatory pathway will eventually allow manufacturers to apply under a modified premarket review process for individual software as a medical device (SaMD) that use artificial intelligence (AI) and machine learning. In the meantime, however, many hospitals have developed and deployed their own predictive CDS systems that cross the boundaries into the FDA’s purview and, indeed, do “interpret or analyze” data for real-time EHR alerts, population health management, and other applications.

Regulatory oversight for hospitals could provide quality or safety standards where currently there are none. However, such regulations could also interfere with existing local care practices, hinder rapid development of new CDS systems, and may be perceived as interfering in hospital operations. With the current enthusiasm for AI-based technologies and the concurrent lack of evidence to suggest their effectiveness in practice, regulation could also prompt necessary scrutiny of potential harms of CDS systems, an area with even less evidence. At the same time, CDS developers—private or hospital based—may be able to avoid regulation for some devices with well-placed disclaimers about the intended use of the CDS, one of the FDA criteria for determining the degree of oversight. If the FDA were to regulate hospitals or hospital-developed CDS systems, there are several unanswered questions to consider so that such regulations have their intended impact.

First, does the FDA intend to regulate hospitals and hospital-developed software at all? The framework for determining whether a CDS system will be regulated depends on the severity of the clinical scenario, the ability to independently evaluate the model output, and the intended user (Table). Notably, many types of CDS systems that would require regulation under this framework are already commonplace. For example, the FDA intends to regulate software that “identifies patients who may exhibit signs of opioid addiction,” a scenario similar to prediction models already developed at academic hospitals.⁸ The FDA also plans to regulate a software device even if it is not a CDS system if it is “intended to generate an alarm or an alert to notify a caregiver of a life-threatening condition, such as stroke, and the caregiver relies primarily on this alarm or alert to make a treatment decision.” Although there are no published reports of stroke-specific early warning systems in use, analogous nonspecific and sepsis-specific early warning systems to prompt urgent clinical care have been deployed by hospitals directly⁹ and developed for embedding in commercial EHRs.¹⁰ Hospitals need clarification on the FDA’s regulatory intentions for such CDS systems. FDA regulation of hospitals and hospital-developed CDS systems would fill a critical oversight need and potentially strengthen processes to improve safety and effectiveness. But burdensome regulations may also restrain hospitals from tackling complex problems in medicine for which they are uniquely suited.

Such a regulatory environment may be especially prohibitive for safety-net hospitals that could find themselves at a disadvantage in developing their own CDS systems relative to large academic medical centers that are typically endowed with greater resources. Additionally, CDS systems developed at academic medical centers may not generalize well to populations in the community setting, which could further deepen disparities in access to cutting-edge technologies. For example, racial bias in treatment and referral patterns could bias training labels for CDS systems focused on population health management.¹¹ Similarly, the composition of patient skin color in one population may distort predictions of a model in another with a different distribution of skin color, even when the primary outcome of a prediction model is gender.¹² Additional regulatory steps may apply for models that are adapted to new populations or recalibrated across locations and time.¹³ Until there is more data on the clinical impact of such CDS systems, it is unknown how potential differences in evaluation and approval would actually affect clinical outcomes.

Second, would hospitals be eligible for the Pre-Cert program, and if so, would they be held to the same standards as a private technology manufacturer? The domains of excellence required for precertification approval such as “patient safety,” “clinical responsibility,” and “proactive culture” are aligned with the efforts of hospitals that are already overseen and accredited by organizations like the Joint Commission on Accreditation of Healthcare Organizations and the American Nurses Credentialing Center. There is limited motivation for the FDA to be in the business of regulating these aspects of hospital functions. However, while domains like “product quality” and “cybersecurity” may be less familiar to some hospitals, these existing credentialing bodies may be better suited than the FDA to set and enforce standards for hospitals. In contrast, private manufacturers may have deep expertise in these latter domains. Therefore, as with public-private partnerships for the development of predictive radiology applications,¹⁴ synergies between hospitals and manufacturers may also prove useful for obtaining approvals in a competitive marketplace. Simultaneously, such collaborations would continue to raise questions about conflicts of interest and data privacy.

Finally, regardless of how the FDA will regulate hospitals, what will become of predictive CDS systems that fall outside of the FDA’s scope? Hospitals will continue to find themselves in the position of self-regulation without clear guidance. Although the FDA suggests that developers of unregulated CDS systems still follow best practices for software validation and cybersecurity, existing guidance documents in these domains do not cover the full range of concerns relevant to the development, deployment, and oversight of AI-based CDS systems in the clinical domain. Nor do most hospitals have the infrastructure or expertise to oversee their own CDS systems. Disparate recommendations for development, training, and oversight of AI-based medical systems have emerged but have yet to be endorsed by a federal regulatory body or become part of the hospital accreditation process.¹⁵ Optimal local oversight would require a collaboration between clinical experts, hospital operations leaders, statisticians, data scientists, and ethics experts to ensure effectiveness, safety, and fairness.

Hospitals will remain at the forefront of developing and implementing predictive CDS systems. The proposed FDA regulatory framework would mark an important step toward realizing benefit from such systems, but the FDA needs to clarify the requirements for hospitals and hospital-developed CDS systems to ensure reasonable standards that account for their differences from private software manufacturers. Should the FDA choose to focus regulation on private manufacturers only, hospitals leaders may both feel more empowered to develop their own local CDS tools and feel more comfortable buying CDS systems from vendors that have been precertified. This strategy would provide an optimal balance of assurance and flexibility while maintaining quality standards that ultimately improve patient care.

Recent experiences in the transportation industry highlight the importance of getting right the regulation of decision-support systems in high-stakes environments. Two tragic plane crashes resulted in 346 deaths and were deemed, in part, to be related to a cockpit alert system that overwhelmed pilots with multiple notifications.¹ Similarly, a driverless car struck and killed a pedestrian in the street, in part because the car was not programmed to look for humans outside of a crosswalk.² These two bellwether events offer poignant lessons for the healthcare industry in which human lives also depend on decision-support systems.

Clinical decision-support (CDS) systems are computerized applications, often embedded in an electronic health record (EHR), that provide information to clinicians to inform care. Although CDS systems have been used for many years,³ they have never been subjected to any enforcement of formal testing requirements. However, a draft guidance document released in 2019 from the Food and Drug Administration (FDA) outlined new directions for the regulation of CDS systems.⁴ Although the FDA has thus far focused regulatory efforts on predictive systems developed by private manufacturers,^5,6 this new document provides examples of software that would require regulation for CDS systems that hospitals are already using. Thus, this new guidance raises critical questions—will hospitals themselves be evaluated like private manufacturers, be exempted from federal regulation, or require their own specialized regulation? The FDA has not yet clarified its approach to hospitals or hospital-developed CDS systems, which leaves open numerous possibilities in a rapidly evolving regulatory environment.

Although the FDA has officially regulated CDS systems under section 201(h) of the Federal Food, Drug, and Cosmetic Act (1938), only recently has the FDA begun to sketch the shape of its regulatory efforts. This trend to actually regulate CDS systems began with the 21st Century Cures Act (2016) that amended the definition of software systems that qualify as medical devices and outlined criteria under which a system may be exempt from FDA oversight. For example, regulation would not apply to systems that support “population health” or a “healthy lifestyle” or to ones that qualify as “electronic patient records” as long as they do not “interpret or analyze” data within them.⁷ Following the rapid proliferation of many machine learning and other predictive technologies with medical applications, the FDA began the voluntary Digital Health Software Precertification (Pre-Cert) Program in 2017. Through this program, the FDA selected nine companies from more than 100 applicants and certified them across five domains of excellence. Notably, the Pre-Cert Program currently allows for certification of software manufacturers themselves and does not approve or test actual software devices directly. This regulatory pathway will eventually allow manufacturers to apply under a modified premarket review process for individual software as a medical device (SaMD) that use artificial intelligence (AI) and machine learning. In the meantime, however, many hospitals have developed and deployed their own predictive CDS systems that cross the boundaries into the FDA’s purview and, indeed, do “interpret or analyze” data for real-time EHR alerts, population health management, and other applications.

Regulatory oversight for hospitals could provide quality or safety standards where currently there are none. However, such regulations could also interfere with existing local care practices, hinder rapid development of new CDS systems, and may be perceived as interfering in hospital operations. With the current enthusiasm for AI-based technologies and the concurrent lack of evidence to suggest their effectiveness in practice, regulation could also prompt necessary scrutiny of potential harms of CDS systems, an area with even less evidence. At the same time, CDS developers—private or hospital based—may be able to avoid regulation for some devices with well-placed disclaimers about the intended use of the CDS, one of the FDA criteria for determining the degree of oversight. If the FDA were to regulate hospitals or hospital-developed CDS systems, there are several unanswered questions to consider so that such regulations have their intended impact.

First, does the FDA intend to regulate hospitals and hospital-developed software at all? The framework for determining whether a CDS system will be regulated depends on the severity of the clinical scenario, the ability to independently evaluate the model output, and the intended user (Table). Notably, many types of CDS systems that would require regulation under this framework are already commonplace. For example, the FDA intends to regulate software that “identifies patients who may exhibit signs of opioid addiction,” a scenario similar to prediction models already developed at academic hospitals.⁸ The FDA also plans to regulate a software device even if it is not a CDS system if it is “intended to generate an alarm or an alert to notify a caregiver of a life-threatening condition, such as stroke, and the caregiver relies primarily on this alarm or alert to make a treatment decision.” Although there are no published reports of stroke-specific early warning systems in use, analogous nonspecific and sepsis-specific early warning systems to prompt urgent clinical care have been deployed by hospitals directly⁹ and developed for embedding in commercial EHRs.¹⁰ Hospitals need clarification on the FDA’s regulatory intentions for such CDS systems. FDA regulation of hospitals and hospital-developed CDS systems would fill a critical oversight need and potentially strengthen processes to improve safety and effectiveness. But burdensome regulations may also restrain hospitals from tackling complex problems in medicine for which they are uniquely suited.

Such a regulatory environment may be especially prohibitive for safety-net hospitals that could find themselves at a disadvantage in developing their own CDS systems relative to large academic medical centers that are typically endowed with greater resources. Additionally, CDS systems developed at academic medical centers may not generalize well to populations in the community setting, which could further deepen disparities in access to cutting-edge technologies. For example, racial bias in treatment and referral patterns could bias training labels for CDS systems focused on population health management.¹¹ Similarly, the composition of patient skin color in one population may distort predictions of a model in another with a different distribution of skin color, even when the primary outcome of a prediction model is gender.¹² Additional regulatory steps may apply for models that are adapted to new populations or recalibrated across locations and time.¹³ Until there is more data on the clinical impact of such CDS systems, it is unknown how potential differences in evaluation and approval would actually affect clinical outcomes.

Second, would hospitals be eligible for the Pre-Cert program, and if so, would they be held to the same standards as a private technology manufacturer? The domains of excellence required for precertification approval such as “patient safety,” “clinical responsibility,” and “proactive culture” are aligned with the efforts of hospitals that are already overseen and accredited by organizations like the Joint Commission on Accreditation of Healthcare Organizations and the American Nurses Credentialing Center. There is limited motivation for the FDA to be in the business of regulating these aspects of hospital functions. However, while domains like “product quality” and “cybersecurity” may be less familiar to some hospitals, these existing credentialing bodies may be better suited than the FDA to set and enforce standards for hospitals. In contrast, private manufacturers may have deep expertise in these latter domains. Therefore, as with public-private partnerships for the development of predictive radiology applications,¹⁴ synergies between hospitals and manufacturers may also prove useful for obtaining approvals in a competitive marketplace. Simultaneously, such collaborations would continue to raise questions about conflicts of interest and data privacy.

Finally, regardless of how the FDA will regulate hospitals, what will become of predictive CDS systems that fall outside of the FDA’s scope? Hospitals will continue to find themselves in the position of self-regulation without clear guidance. Although the FDA suggests that developers of unregulated CDS systems still follow best practices for software validation and cybersecurity, existing guidance documents in these domains do not cover the full range of concerns relevant to the development, deployment, and oversight of AI-based CDS systems in the clinical domain. Nor do most hospitals have the infrastructure or expertise to oversee their own CDS systems. Disparate recommendations for development, training, and oversight of AI-based medical systems have emerged but have yet to be endorsed by a federal regulatory body or become part of the hospital accreditation process.¹⁵ Optimal local oversight would require a collaboration between clinical experts, hospital operations leaders, statisticians, data scientists, and ethics experts to ensure effectiveness, safety, and fairness.

Hospitals will remain at the forefront of developing and implementing predictive CDS systems. The proposed FDA regulatory framework would mark an important step toward realizing benefit from such systems, but the FDA needs to clarify the requirements for hospitals and hospital-developed CDS systems to ensure reasonable standards that account for their differences from private software manufacturers. Should the FDA choose to focus regulation on private manufacturers only, hospitals leaders may both feel more empowered to develop their own local CDS tools and feel more comfortable buying CDS systems from vendors that have been precertified. This strategy would provide an optimal balance of assurance and flexibility while maintaining quality standards that ultimately improve patient care.

There is growing momentum to hold drug manufacturers accountable for the more than 400,000 US opioid overdose deaths that have occurred since 1999.¹ As state lawsuits against pharmaceutical manufacturers and distributors wind their way through the legal system, hospitals—which may benefit from settlement funds—have been paying close attention. Recently, former Governor John Kasich (R-Ohio), West Virginia University president E. Gordon Gee, and America’s Essential Hospitals argued that adequately compensating hospitals for the costs of being on the crisis’ “front lines” requires prioritizing them as settlement fund recipients.²

Hospitals should be laying the groundwork for how settlement funds might be used. They may consider enhancing some of the most promising, evidence-based services for individuals with opioid use disorders (OUDs), including improving treatment for commonly associated health conditions such as HIV and hepatitis C virus (HCV); expanding ambulatory long-term antibiotic treatment for endocarditis and other intravenous drug use–associated infections; more broadly adopting harm-reduction practices such as naloxone coprescribing; and applying best practices to caring for substance-exposed infants. They could also develop clinical services not already provided, including creating programs for OUD management during pregnancy and initiating medication for OUD in inpatient, emergency department, and ambulatory settings. In short, hospitals play a critical role in engaging people with OUD in treatment at every possible opportunity.³

When considering how to most effectively use opioid settlement funding, hospitals may consider adding or expanding these much-needed clinical services to address opioid-­related harms; however, their efforts should not stop there. Investments made outside hospital walls could have a significant effect on the public’s health, especially if they target social determinants of health. By tackling factors in the pathway to developing OUD, such as lack of meaningful employment, affordable housing, and mental health care, hospitals can move beyond treating the downstream consequences of addiction and toward preventing community-level opioid-related harms. To accomplish this daunting goal, hospitals will need to strengthen existing relationships with community partners and build new ones. Yet in a 2015 study, only 54% of nonprofit hospitals proposed a strategy to address the overdose crisis that involved community partnering.⁴

In this Perspective, we describe the following three strategies hospitals can use to multiply the reach of their opioid settlement funding by addressing root causes of opioid use through primary prevention: (1) supporting economic opportunities in their communities, (2) expanding affordable housing options in surrounding neighborhoods, and (3) building capacity in ambulatory practices and pharmacies to prevent OUD (Table).

Lack of economic opportunity is one of many root causes of opioid use. For example, a recent study found that automotive assembly plant closures were associated with increases in opioid overdose mortality.⁵ To tackle this complex issue, hospitals can play a crucial role in expanding employment and career advancement options for members of their local communities. Specifically, hospitals can do the following:

• Create jobs within the healthcare system and preferentially recruit and hire from surrounding neighborhoods
• Establish structured career development programs to build skills among entry-level healthcare employees
• Award contracts of varying sizes to locally owned businesses
• Employ individuals with lived experience with substance use disorders, such as peer recovery coaches⁶

To illustrate how health systems are investing in enhancing career opportunities for members of their communities, hundreds of institutions have implemented “School at Work,” a 6-month career development program for entry-level healthcare employees.⁷ The hospitals’ Human Resources department trains participants in communication skills, reading and writing, patient safety and satisfaction, medical terminology, and strategies for success and career advancement. Evaluations of this program have demonstrated improved employee outcomes and a favorable return on investment for hospitals.⁸

As “anchor institutions” and large employers in many communities, hospitals can simultaneously enhance their own workforce and offer employment opportunities that can help break the cycle of addiction that commonly traps individuals and families in communities affected by the overdose crisis.

Hospitals are increasingly supporting interventions that fall outside their traditional purview as they seek to improve population health, such as developing safe green outdoor spaces and increasing access to healthy food options by supporting local farmers markets and grocers.⁹ Stable, decent, and affordable housing is critically important to health and well-being,¹⁰ and there is a well-documented association of opioid use disorder and opioid misuse with housing instability.¹¹ Given evidence of improved outcomes with hospital-led housing interventions,¹² a growing number of hospitals are partnering with housing authorities and community groups to help do the following¹³:

• Contribute to supportive housing options
• Provide environmental health assessments, repairs, and renovations
• Buy or develop affordable housing units

Boston Medical Center, where one in four inpatients are experiencing homelessnes and one in three pediatric emergency department patients are housing insecure, provides an example of how a hospital has invested in housing.¹⁴ In 2017, the hospital began a 5-year, $6.5 million investment in community partnerships in surrounding neighborhoods. Instead of building housing units or acting as a landlord, the hospital chose to invest funding in creative ways to increase the pool of affordable housing. It invested $1 million to rehabilitate permanent, supportive housing units for individuals with mental health conditions in a nearby Boston neighborhood and in a housing stabilization program for people with complex medical issues including substance use disorder. It provided resources to a homeless shelter near the hospital and to the Boston Health Care for the Homeless Program, which provides healthcare to individuals with housing instability. It also funded a community wellness advocate based at the hospital, who received training in substance use disorders and served as a liaison between the hospital and the Boston Housing Authority.

Housing instability is just one of the social determinants of health that hospitals have the capacity to address as they consider where to invest their opioid settlement funds.

Finally, hospitals can partner with community ambulatory practices and pharmacies to prevent the progression to problematic opioid use and OUD. Specifically, hospitals can do as follows:

• Provide evidence-based training to community providers on safe prescribing practices for acute and chronic pain management, as well as postoperative, postprocedural, and postpartum pain management
• Support ambulatory providers in expanding office-based mental health treatment through direct care via telemedicine and in building mental health treatment capacity through consultation, continuing medical education, and telementorship (eg, Project ECHO¹⁵)
• Support ambulatory providers to implement risk reduction strategies to prevent initiation of problematic opioid use, particularly among adolescents and young adults
• Partner with local pharmacies to promote point-of-­prescription counseling on the risks and benefits of opioids

Hospitals bring key strengths and resources to these prevention-­oriented partnerships. First, they may have resources available for clinical research, implementation support, program evaluation, and quality improvement, bringing such expertise to partnerships with ambulatory practices and pharmacies. They likely have specific expertise among their staff, including areas such as pain management, obstetric care, pediatrics, and adolescent medicine, and can provide specialists for consultation services or telementoring initiatives. They also can organize continuing medical education and can offer in-service training at local practices and pharmacies.

Project ECHO is one example of telementoring to build capacity among community providers to manage chronic pain and address addiction and other related harms.¹⁶ The Project ECHO model includes virtual sessions with didactic content and case presentations during which specialists mentor community clinicians. Specific to primary prevention, telementoring has been shown to improve access to evidence-based treatment of chronic pain and mental health conditions,^17,18 which could prevent the development of OUD. By equipping community clinicians with tools to prevent the development of problematic opioid use, hospitals can help reduce the downstream burden of OUD and its associated morbidity, mortality, and costs.

The opioid crisis has devastated families, reduced life expectancy in certain communities,¹⁹ and had a substantial financial impact on hospitals—resulting in an estimated $11 billion in costs to US hospitals each year.²⁰ This ongoing crisis is only going to be compounded by the recent emergence of the SARS-CoV-2 virus. Hospital resources are being strained in unprecedented ways, which has required unprecedented responses in order to continue to serve their communities. Supporting economic opportunity, stable housing, and mental health treatment will be challenging in this new environment but has never been more urgently needed. If opioid settlement funds are targeted to US hospitals, they should be held accountable for where funds are spent because they have a unique opportunity to focus on primary prevention in their communities—confronting OUD before it begins.²¹ However, if hospitals use opioid settlement funding only to continue to provide services already offered, or fail to make bold investments in their communities, this public health crisis will continue to strain the resources of those providing clinical care on the front lines.

The authors wish to thank Hilary Peterson of the RAND Corporation for preparing the paper for submission. She was not compensated for her contribution.

The authors report being supported by grants from the National Institute on Drug Abuse of the National Institutes of Health under awards R21DA045212 (Dr Faherty), K23DA045085 (Dr Hadland), L40DA042434 (Dr Hadland), K23DA038720 (Dr Patrick), R01DA045729 (Dr Patrick), and P50DA046351 (Dr Stein). Dr Hadland also reports grant support from the Thrasher Research Fund and the Academic Pediatric Association. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

There is growing momentum to hold drug manufacturers accountable for the more than 400,000 US opioid overdose deaths that have occurred since 1999.¹ As state lawsuits against pharmaceutical manufacturers and distributors wind their way through the legal system, hospitals—which may benefit from settlement funds—have been paying close attention. Recently, former Governor John Kasich (R-Ohio), West Virginia University president E. Gordon Gee, and America’s Essential Hospitals argued that adequately compensating hospitals for the costs of being on the crisis’ “front lines” requires prioritizing them as settlement fund recipients.²

Hospitals should be laying the groundwork for how settlement funds might be used. They may consider enhancing some of the most promising, evidence-based services for individuals with opioid use disorders (OUDs), including improving treatment for commonly associated health conditions such as HIV and hepatitis C virus (HCV); expanding ambulatory long-term antibiotic treatment for endocarditis and other intravenous drug use–associated infections; more broadly adopting harm-reduction practices such as naloxone coprescribing; and applying best practices to caring for substance-exposed infants. They could also develop clinical services not already provided, including creating programs for OUD management during pregnancy and initiating medication for OUD in inpatient, emergency department, and ambulatory settings. In short, hospitals play a critical role in engaging people with OUD in treatment at every possible opportunity.³

When considering how to most effectively use opioid settlement funding, hospitals may consider adding or expanding these much-needed clinical services to address opioid-­related harms; however, their efforts should not stop there. Investments made outside hospital walls could have a significant effect on the public’s health, especially if they target social determinants of health. By tackling factors in the pathway to developing OUD, such as lack of meaningful employment, affordable housing, and mental health care, hospitals can move beyond treating the downstream consequences of addiction and toward preventing community-level opioid-related harms. To accomplish this daunting goal, hospitals will need to strengthen existing relationships with community partners and build new ones. Yet in a 2015 study, only 54% of nonprofit hospitals proposed a strategy to address the overdose crisis that involved community partnering.⁴

In this Perspective, we describe the following three strategies hospitals can use to multiply the reach of their opioid settlement funding by addressing root causes of opioid use through primary prevention: (1) supporting economic opportunities in their communities, (2) expanding affordable housing options in surrounding neighborhoods, and (3) building capacity in ambulatory practices and pharmacies to prevent OUD (Table).

Lack of economic opportunity is one of many root causes of opioid use. For example, a recent study found that automotive assembly plant closures were associated with increases in opioid overdose mortality.⁵ To tackle this complex issue, hospitals can play a crucial role in expanding employment and career advancement options for members of their local communities. Specifically, hospitals can do the following:

• Create jobs within the healthcare system and preferentially recruit and hire from surrounding neighborhoods
• Establish structured career development programs to build skills among entry-level healthcare employees
• Award contracts of varying sizes to locally owned businesses
• Employ individuals with lived experience with substance use disorders, such as peer recovery coaches⁶

To illustrate how health systems are investing in enhancing career opportunities for members of their communities, hundreds of institutions have implemented “School at Work,” a 6-month career development program for entry-level healthcare employees.⁷ The hospitals’ Human Resources department trains participants in communication skills, reading and writing, patient safety and satisfaction, medical terminology, and strategies for success and career advancement. Evaluations of this program have demonstrated improved employee outcomes and a favorable return on investment for hospitals.⁸

As “anchor institutions” and large employers in many communities, hospitals can simultaneously enhance their own workforce and offer employment opportunities that can help break the cycle of addiction that commonly traps individuals and families in communities affected by the overdose crisis.

Hospitals are increasingly supporting interventions that fall outside their traditional purview as they seek to improve population health, such as developing safe green outdoor spaces and increasing access to healthy food options by supporting local farmers markets and grocers.⁹ Stable, decent, and affordable housing is critically important to health and well-being,¹⁰ and there is a well-documented association of opioid use disorder and opioid misuse with housing instability.¹¹ Given evidence of improved outcomes with hospital-led housing interventions,¹² a growing number of hospitals are partnering with housing authorities and community groups to help do the following¹³:

• Contribute to supportive housing options
• Provide environmental health assessments, repairs, and renovations
• Buy or develop affordable housing units

Boston Medical Center, where one in four inpatients are experiencing homelessnes and one in three pediatric emergency department patients are housing insecure, provides an example of how a hospital has invested in housing.¹⁴ In 2017, the hospital began a 5-year, $6.5 million investment in community partnerships in surrounding neighborhoods. Instead of building housing units or acting as a landlord, the hospital chose to invest funding in creative ways to increase the pool of affordable housing. It invested $1 million to rehabilitate permanent, supportive housing units for individuals with mental health conditions in a nearby Boston neighborhood and in a housing stabilization program for people with complex medical issues including substance use disorder. It provided resources to a homeless shelter near the hospital and to the Boston Health Care for the Homeless Program, which provides healthcare to individuals with housing instability. It also funded a community wellness advocate based at the hospital, who received training in substance use disorders and served as a liaison between the hospital and the Boston Housing Authority.

Housing instability is just one of the social determinants of health that hospitals have the capacity to address as they consider where to invest their opioid settlement funds.

Finally, hospitals can partner with community ambulatory practices and pharmacies to prevent the progression to problematic opioid use and OUD. Specifically, hospitals can do as follows:

• Provide evidence-based training to community providers on safe prescribing practices for acute and chronic pain management, as well as postoperative, postprocedural, and postpartum pain management
• Support ambulatory providers in expanding office-based mental health treatment through direct care via telemedicine and in building mental health treatment capacity through consultation, continuing medical education, and telementorship (eg, Project ECHO¹⁵)
• Support ambulatory providers to implement risk reduction strategies to prevent initiation of problematic opioid use, particularly among adolescents and young adults
• Partner with local pharmacies to promote point-of-­prescription counseling on the risks and benefits of opioids

Hospitals bring key strengths and resources to these prevention-­oriented partnerships. First, they may have resources available for clinical research, implementation support, program evaluation, and quality improvement, bringing such expertise to partnerships with ambulatory practices and pharmacies. They likely have specific expertise among their staff, including areas such as pain management, obstetric care, pediatrics, and adolescent medicine, and can provide specialists for consultation services or telementoring initiatives. They also can organize continuing medical education and can offer in-service training at local practices and pharmacies.

Project ECHO is one example of telementoring to build capacity among community providers to manage chronic pain and address addiction and other related harms.¹⁶ The Project ECHO model includes virtual sessions with didactic content and case presentations during which specialists mentor community clinicians. Specific to primary prevention, telementoring has been shown to improve access to evidence-based treatment of chronic pain and mental health conditions,^17,18 which could prevent the development of OUD. By equipping community clinicians with tools to prevent the development of problematic opioid use, hospitals can help reduce the downstream burden of OUD and its associated morbidity, mortality, and costs.

The opioid crisis has devastated families, reduced life expectancy in certain communities,¹⁹ and had a substantial financial impact on hospitals—resulting in an estimated $11 billion in costs to US hospitals each year.²⁰ This ongoing crisis is only going to be compounded by the recent emergence of the SARS-CoV-2 virus. Hospital resources are being strained in unprecedented ways, which has required unprecedented responses in order to continue to serve their communities. Supporting economic opportunity, stable housing, and mental health treatment will be challenging in this new environment but has never been more urgently needed. If opioid settlement funds are targeted to US hospitals, they should be held accountable for where funds are spent because they have a unique opportunity to focus on primary prevention in their communities—confronting OUD before it begins.²¹ However, if hospitals use opioid settlement funding only to continue to provide services already offered, or fail to make bold investments in their communities, this public health crisis will continue to strain the resources of those providing clinical care on the front lines.

The authors wish to thank Hilary Peterson of the RAND Corporation for preparing the paper for submission. She was not compensated for her contribution.

The authors report being supported by grants from the National Institute on Drug Abuse of the National Institutes of Health under awards R21DA045212 (Dr Faherty), K23DA045085 (Dr Hadland), L40DA042434 (Dr Hadland), K23DA038720 (Dr Patrick), R01DA045729 (Dr Patrick), and P50DA046351 (Dr Stein). Dr Hadland also reports grant support from the Thrasher Research Fund and the Academic Pediatric Association. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

There is growing momentum to hold drug manufacturers accountable for the more than 400,000 US opioid overdose deaths that have occurred since 1999.¹ As state lawsuits against pharmaceutical manufacturers and distributors wind their way through the legal system, hospitals—which may benefit from settlement funds—have been paying close attention. Recently, former Governor John Kasich (R-Ohio), West Virginia University president E. Gordon Gee, and America’s Essential Hospitals argued that adequately compensating hospitals for the costs of being on the crisis’ “front lines” requires prioritizing them as settlement fund recipients.²

Hospitals should be laying the groundwork for how settlement funds might be used. They may consider enhancing some of the most promising, evidence-based services for individuals with opioid use disorders (OUDs), including improving treatment for commonly associated health conditions such as HIV and hepatitis C virus (HCV); expanding ambulatory long-term antibiotic treatment for endocarditis and other intravenous drug use–associated infections; more broadly adopting harm-reduction practices such as naloxone coprescribing; and applying best practices to caring for substance-exposed infants. They could also develop clinical services not already provided, including creating programs for OUD management during pregnancy and initiating medication for OUD in inpatient, emergency department, and ambulatory settings. In short, hospitals play a critical role in engaging people with OUD in treatment at every possible opportunity.³

When considering how to most effectively use opioid settlement funding, hospitals may consider adding or expanding these much-needed clinical services to address opioid-­related harms; however, their efforts should not stop there. Investments made outside hospital walls could have a significant effect on the public’s health, especially if they target social determinants of health. By tackling factors in the pathway to developing OUD, such as lack of meaningful employment, affordable housing, and mental health care, hospitals can move beyond treating the downstream consequences of addiction and toward preventing community-level opioid-related harms. To accomplish this daunting goal, hospitals will need to strengthen existing relationships with community partners and build new ones. Yet in a 2015 study, only 54% of nonprofit hospitals proposed a strategy to address the overdose crisis that involved community partnering.⁴

In this Perspective, we describe the following three strategies hospitals can use to multiply the reach of their opioid settlement funding by addressing root causes of opioid use through primary prevention: (1) supporting economic opportunities in their communities, (2) expanding affordable housing options in surrounding neighborhoods, and (3) building capacity in ambulatory practices and pharmacies to prevent OUD (Table).

Lack of economic opportunity is one of many root causes of opioid use. For example, a recent study found that automotive assembly plant closures were associated with increases in opioid overdose mortality.⁵ To tackle this complex issue, hospitals can play a crucial role in expanding employment and career advancement options for members of their local communities. Specifically, hospitals can do the following:

• Create jobs within the healthcare system and preferentially recruit and hire from surrounding neighborhoods
• Establish structured career development programs to build skills among entry-level healthcare employees
• Award contracts of varying sizes to locally owned businesses
• Employ individuals with lived experience with substance use disorders, such as peer recovery coaches⁶

To illustrate how health systems are investing in enhancing career opportunities for members of their communities, hundreds of institutions have implemented “School at Work,” a 6-month career development program for entry-level healthcare employees.⁷ The hospitals’ Human Resources department trains participants in communication skills, reading and writing, patient safety and satisfaction, medical terminology, and strategies for success and career advancement. Evaluations of this program have demonstrated improved employee outcomes and a favorable return on investment for hospitals.⁸

As “anchor institutions” and large employers in many communities, hospitals can simultaneously enhance their own workforce and offer employment opportunities that can help break the cycle of addiction that commonly traps individuals and families in communities affected by the overdose crisis.

Hospitals are increasingly supporting interventions that fall outside their traditional purview as they seek to improve population health, such as developing safe green outdoor spaces and increasing access to healthy food options by supporting local farmers markets and grocers.⁹ Stable, decent, and affordable housing is critically important to health and well-being,¹⁰ and there is a well-documented association of opioid use disorder and opioid misuse with housing instability.¹¹ Given evidence of improved outcomes with hospital-led housing interventions,¹² a growing number of hospitals are partnering with housing authorities and community groups to help do the following¹³:

• Contribute to supportive housing options
• Provide environmental health assessments, repairs, and renovations
• Buy or develop affordable housing units

Boston Medical Center, where one in four inpatients are experiencing homelessnes and one in three pediatric emergency department patients are housing insecure, provides an example of how a hospital has invested in housing.¹⁴ In 2017, the hospital began a 5-year, $6.5 million investment in community partnerships in surrounding neighborhoods. Instead of building housing units or acting as a landlord, the hospital chose to invest funding in creative ways to increase the pool of affordable housing. It invested $1 million to rehabilitate permanent, supportive housing units for individuals with mental health conditions in a nearby Boston neighborhood and in a housing stabilization program for people with complex medical issues including substance use disorder. It provided resources to a homeless shelter near the hospital and to the Boston Health Care for the Homeless Program, which provides healthcare to individuals with housing instability. It also funded a community wellness advocate based at the hospital, who received training in substance use disorders and served as a liaison between the hospital and the Boston Housing Authority.

Housing instability is just one of the social determinants of health that hospitals have the capacity to address as they consider where to invest their opioid settlement funds.

Finally, hospitals can partner with community ambulatory practices and pharmacies to prevent the progression to problematic opioid use and OUD. Specifically, hospitals can do as follows:

• Provide evidence-based training to community providers on safe prescribing practices for acute and chronic pain management, as well as postoperative, postprocedural, and postpartum pain management
• Support ambulatory providers in expanding office-based mental health treatment through direct care via telemedicine and in building mental health treatment capacity through consultation, continuing medical education, and telementorship (eg, Project ECHO¹⁵)
• Support ambulatory providers to implement risk reduction strategies to prevent initiation of problematic opioid use, particularly among adolescents and young adults
• Partner with local pharmacies to promote point-of-­prescription counseling on the risks and benefits of opioids

Hospitals bring key strengths and resources to these prevention-­oriented partnerships. First, they may have resources available for clinical research, implementation support, program evaluation, and quality improvement, bringing such expertise to partnerships with ambulatory practices and pharmacies. They likely have specific expertise among their staff, including areas such as pain management, obstetric care, pediatrics, and adolescent medicine, and can provide specialists for consultation services or telementoring initiatives. They also can organize continuing medical education and can offer in-service training at local practices and pharmacies.

Project ECHO is one example of telementoring to build capacity among community providers to manage chronic pain and address addiction and other related harms.¹⁶ The Project ECHO model includes virtual sessions with didactic content and case presentations during which specialists mentor community clinicians. Specific to primary prevention, telementoring has been shown to improve access to evidence-based treatment of chronic pain and mental health conditions,^17,18 which could prevent the development of OUD. By equipping community clinicians with tools to prevent the development of problematic opioid use, hospitals can help reduce the downstream burden of OUD and its associated morbidity, mortality, and costs.

The opioid crisis has devastated families, reduced life expectancy in certain communities,¹⁹ and had a substantial financial impact on hospitals—resulting in an estimated $11 billion in costs to US hospitals each year.²⁰ This ongoing crisis is only going to be compounded by the recent emergence of the SARS-CoV-2 virus. Hospital resources are being strained in unprecedented ways, which has required unprecedented responses in order to continue to serve their communities. Supporting economic opportunity, stable housing, and mental health treatment will be challenging in this new environment but has never been more urgently needed. If opioid settlement funds are targeted to US hospitals, they should be held accountable for where funds are spent because they have a unique opportunity to focus on primary prevention in their communities—confronting OUD before it begins.²¹ However, if hospitals use opioid settlement funding only to continue to provide services already offered, or fail to make bold investments in their communities, this public health crisis will continue to strain the resources of those providing clinical care on the front lines.

The authors wish to thank Hilary Peterson of the RAND Corporation for preparing the paper for submission. She was not compensated for her contribution.

The authors report being supported by grants from the National Institute on Drug Abuse of the National Institutes of Health under awards R21DA045212 (Dr Faherty), K23DA045085 (Dr Hadland), L40DA042434 (Dr Hadland), K23DA038720 (Dr Patrick), R01DA045729 (Dr Patrick), and P50DA046351 (Dr Stein). Dr Hadland also reports grant support from the Thrasher Research Fund and the Academic Pediatric Association. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.