Hospital Medicine

Inspired by the ABIM Foundation’s Choosing Wisely^® campaign, the “Things We Do for No Reason^™” (TWDFNR) series reviews practices that have become common parts of hospital care but may provide little value to our patients. Practices reviewed in the TWDFNR series do not represent clear-cut conclusions or clinical practice standards but are meant as a starting place for research and active discussions among hospitalists and patients. We invite you to be part of that discussion.

A 56-year-old woman presents to the emergency department with a 2-week history of abdominal pain associated with nausea and an episode of nonbilious, nonbloody emesis. Her last bowel movement was 2 days prior to her presentation. The patient has tachycardia to 105 beats per minute but otherwise normal vital signs. Findings on her physical examination include dry mucous membranes and increased bowel sounds. A review of systems reveals an unintentional weight loss of 15 kg over the past 4 months and increased fatigue. Computed tomography scan of the abdomen and pelvis with contrast reveals multiple areas of attenuation in the liver and small bowel obstruction. The hospitalist admits the patient to the medicine service for supportive treatment and workup for underlying malignancy. Her admitting team orders serum tumor biomarkers on admission to expedite the diagnosis.

When patients present with unexplained weight loss or with metastasis from an unknown primary location, the initial workup often includes imaging and a tumor biomarker panel (eg, cancer antigen 125 [CA125], carbohydrate antigen 19-9 [CA19-9], carcinoembryonic antigen [CEA]). The CA125, CA19-9, and CEA biomarkers are traditionally associated with ovarian, pancreatic, and colorectal cancer, respectively.¹ While clinicians initially used these serum biomarkers to monitor for cancer recurrence or treatment response, they have since become widely used in multiple clinical stages of oncological evaluation.

Hospitalists routinely order biomarkers as part of the malignancy workup. More than a dozen oncology biomarkers are used in the clinical setting to risk stratify, plan treatment, and monitor for recurrence. For example, studies associate elevated preoperative levels of CEA and CA19-9 with metastatic invasion of colorectal² and gastric³ cancers and with poor prognosis of intrahepatic cholangiocarcinoma. Similarly, CA125 has demonstrated utility in monitoring response to ovarian cancer treatment.⁴ Specific biomarkers, such as alpha-fetoprotein, improve diagnosis of liver and nonseminomatous testicular tumors.⁵ Clinicians often apply the same paradigm to other biomarkers due to their widespread availability, noninvasiveness, reproducibility, and ease of use, particularly in acute settings wherein any new information is perceived to be potentially helpful.

Utilizing these serum biomarkers to diagnose cancer has the potential for diagnostic error and can result in unnecessary patient anxiety and follow-up testing. Since tissue sampling is necessary and remains the gold standard in most cancer diagnoses, obtaining these tumor biomarkers in the early diagnostic stage does not change management and may even lead to harm. Furthermore, due to their poor sensitivity and specificity, these biomarkers cannot rule in or rule out cancer. Elevated CA125, CA19-9, and CEA biomarkers occur in a variety of malignancies, including gastric, gallbladder, hepatocellular, bladder, and breast cancers.^1,3,6 In addition, these biomarkers have a very limited role in the workup of cancer of unknown primary origin.⁷

Even in the setting of a known pelvic mass, the use of CA125 alone has poor sensitivity at a cut-off level of 35 U/mL as a biomarker for the diagnosis of early ovarian cancer.⁸

Serum CA19-9 is not a useful diagnostic biomarker as elevated CA19-9 can occur in benign conditions, including cirrhosis, chronic pancreatitis, and cholangitis. In a systematic review of patients with histologic confirmation of pancreatic malignancy, the median positive predictive value of CA19-9 was 72% (interquartile range, 41%-95%).⁹ Additionally, patients with Lewis-null blood type, which is present in 5% to 10% of the Caucasian population, do not produce CA19-9.¹⁰ Therefore, CA19-9 will be 0% specific for tumors in this population.

The use of CEA in the diagnosis of colorectal cancer is also questionable. In stage I colorectal cancer, CEA was only 38.1% sensitive at a cut-off level of 2.41 ng/mL; it was 78.3% sensitive in stage IV disease.¹¹ The specificity of CEA is limited since elevated CEA occurs in benign conditions, such as inflammatory bowel disease, smoking, hypothyroidism, pancreatitis, biliary obstruction, peptic ulcers, and cirrhosis—though CEA levels in these conditions are rarely >10 ng/mL.¹¹ Regardless of the results of biomarker testing, definitive diagnosis requires tissue biopsy; therefore, biomarker findings are of little utility in the initial workup.

In addition to variable diagnostic utility, overreliance on these biomarkers has the potential for serious patient harm. In a study examining patients with established rectal cancer, combination CEA and CA19-9 testing alone was insufficient to predict the pathologic stage of disease correctly.² A cancer misdiagnosis not only traumatizes patients but also erodes their trust in clinicians and creates anxiety during future clinical encounters. Overutilization of these tumor biomarkers is also costly and contributes to waste in the US healthcare system.

There is a role for tumor biomarker testing in specific cancers after the primary source of malignancy has been determined. When evaluating a known pelvic mass, CA125 testing is performed in conjunction with transvaginal ultrasound and assessment of menopausal status in the risk of ovarian malignancy algorithm for prognostication of disease prior to surgery.¹² This algorithm takes into account levels of CA125 in addition to levels of human epididymis protein 4 and patient age, yielding an area under the curve as high as 0.93 for ovarian cancer risk classification.⁸ Beyond the prognostication process, oncologists follow CA125 to monitor response to first-line ovarian cancer treatment. However, CA125 has a less defined role in surveillance for ovarian cancer recurrence.

CA19-9 has demonstrated utility for pancreatic cancer and cholangiocarcinoma survival estimates. A national cohort analysis of patients with established intrahepatic cholangiocarcinoma found that CA19-9 independently predicted increased mortality. Patients with elevated CA19-9 also had significantly more nodal metastases and positive-margin resections.⁶ A study of 353 patients with pancreatic ductal adenocarcinoma undergoing radical resection further demonstrated the utility of CA19-9. In this study, patients with postoperative CA19-9 normalization had improved survival by almost 12 months when compared to those with consistently elevated CA19-9.¹³

Last, the literature describes CEA biomarker testing in the surveillance of patients after curative treatment of colon and rectal cancer. The American Society of Colon and Rectal Surgeons recommends regularly tracking this biomarker following curative resection, in conjunction with colonoscopy and chest and liver imaging studies.¹⁴ A prospective randomized controlled study that followed this monitoring protocol in cured asymptomatic patients on a bimonthly basis found that early diagnosis of recurrent colorectal cancer improved survival.¹⁵ The use of CEA testing as a monitoring tool should therefore be a point of discussion between providers and patients, as its utility varies based on patient comorbidities, their ability to tolerate surgery or chemotherapy, risk factors for recurrence, performance status, compliance, age, and preference.¹⁴

The use of CA125, CA19-9, and CEA testing alone as initial diagnostic tools for malignancy are problematic due to their poor sensitivities and/or positive predictive value. Multiple studies have demonstrated their utility as markers of metastasis or malignancy progression rather than as clinically useful markers for the detection of any one type of cancer.^1,3,6 In an undiagnosed symptomatic patient with unexplained weight loss or symptoms of a tumor mass, elevated CA125, CA19-9, and CEA add no new information as metastatic pancreatic, colorectal, ovarian, gastric, gallbladder, hepatocellular, bladder, ovarian, and breast cancers all remain in the differential diagnosis. Clinicians should approach the initial diagnosis of cancer in such patients with appropriate imaging studies, a thorough physical examination, and prompt biopsy of abnormal findings, as long as these are consistent with the patient’s goals of care. After establishing a tissue diagnosis, some tumor biomarkers have valid prognostic, staging, and monitoring roles.^6,13,14

• Do not routinely order CA125, CA19-9, and CEA tests for the initial diagnostic workup of visceral malignancy of unknown origin regardless of whether imaging studies have been obtained.
• Use appropriate imaging, perform a thorough physical examination, and obtain tissue biopsy in the initial diagnostic workup of a visceral malignancy of unknown origin.

Clinicians should use serum biomarkers, like any other diagnostic test, to maximize benefit while preventing patient harm. In general, CA125, CA19-9, and CEA do not have a role in cancer diagnosis. The patient described in our clinical scenario would not benefit from a serum tumor biomarker panel at the time of admission. Regardless of findings from these tests, a tissue sample is required to make a definitive diagnosis of underlying malignancy in this patient.

Do you think this is a low-value practice? Is this truly a “Thing We Do for No Reason^™”? Share what you do in your practice and join in the conversation online by retweeting it on Twitter (#TWDFNR) and liking it on Facebook. We invite you to propose ideas for other “Things We Do for No Reason^™” topics by emailing [email protected]

Inspired by the ABIM Foundation’s Choosing Wisely^® campaign, the “Things We Do for No Reason^™” (TWDFNR) series reviews practices that have become common parts of hospital care but may provide little value to our patients. Practices reviewed in the TWDFNR series do not represent clear-cut conclusions or clinical practice standards but are meant as a starting place for research and active discussions among hospitalists and patients. We invite you to be part of that discussion.

A 56-year-old woman presents to the emergency department with a 2-week history of abdominal pain associated with nausea and an episode of nonbilious, nonbloody emesis. Her last bowel movement was 2 days prior to her presentation. The patient has tachycardia to 105 beats per minute but otherwise normal vital signs. Findings on her physical examination include dry mucous membranes and increased bowel sounds. A review of systems reveals an unintentional weight loss of 15 kg over the past 4 months and increased fatigue. Computed tomography scan of the abdomen and pelvis with contrast reveals multiple areas of attenuation in the liver and small bowel obstruction. The hospitalist admits the patient to the medicine service for supportive treatment and workup for underlying malignancy. Her admitting team orders serum tumor biomarkers on admission to expedite the diagnosis.

When patients present with unexplained weight loss or with metastasis from an unknown primary location, the initial workup often includes imaging and a tumor biomarker panel (eg, cancer antigen 125 [CA125], carbohydrate antigen 19-9 [CA19-9], carcinoembryonic antigen [CEA]). The CA125, CA19-9, and CEA biomarkers are traditionally associated with ovarian, pancreatic, and colorectal cancer, respectively.¹ While clinicians initially used these serum biomarkers to monitor for cancer recurrence or treatment response, they have since become widely used in multiple clinical stages of oncological evaluation.

Hospitalists routinely order biomarkers as part of the malignancy workup. More than a dozen oncology biomarkers are used in the clinical setting to risk stratify, plan treatment, and monitor for recurrence. For example, studies associate elevated preoperative levels of CEA and CA19-9 with metastatic invasion of colorectal² and gastric³ cancers and with poor prognosis of intrahepatic cholangiocarcinoma. Similarly, CA125 has demonstrated utility in monitoring response to ovarian cancer treatment.⁴ Specific biomarkers, such as alpha-fetoprotein, improve diagnosis of liver and nonseminomatous testicular tumors.⁵ Clinicians often apply the same paradigm to other biomarkers due to their widespread availability, noninvasiveness, reproducibility, and ease of use, particularly in acute settings wherein any new information is perceived to be potentially helpful.

Utilizing these serum biomarkers to diagnose cancer has the potential for diagnostic error and can result in unnecessary patient anxiety and follow-up testing. Since tissue sampling is necessary and remains the gold standard in most cancer diagnoses, obtaining these tumor biomarkers in the early diagnostic stage does not change management and may even lead to harm. Furthermore, due to their poor sensitivity and specificity, these biomarkers cannot rule in or rule out cancer. Elevated CA125, CA19-9, and CEA biomarkers occur in a variety of malignancies, including gastric, gallbladder, hepatocellular, bladder, and breast cancers.^1,3,6 In addition, these biomarkers have a very limited role in the workup of cancer of unknown primary origin.⁷

Even in the setting of a known pelvic mass, the use of CA125 alone has poor sensitivity at a cut-off level of 35 U/mL as a biomarker for the diagnosis of early ovarian cancer.⁸

Serum CA19-9 is not a useful diagnostic biomarker as elevated CA19-9 can occur in benign conditions, including cirrhosis, chronic pancreatitis, and cholangitis. In a systematic review of patients with histologic confirmation of pancreatic malignancy, the median positive predictive value of CA19-9 was 72% (interquartile range, 41%-95%).⁹ Additionally, patients with Lewis-null blood type, which is present in 5% to 10% of the Caucasian population, do not produce CA19-9.¹⁰ Therefore, CA19-9 will be 0% specific for tumors in this population.

The use of CEA in the diagnosis of colorectal cancer is also questionable. In stage I colorectal cancer, CEA was only 38.1% sensitive at a cut-off level of 2.41 ng/mL; it was 78.3% sensitive in stage IV disease.¹¹ The specificity of CEA is limited since elevated CEA occurs in benign conditions, such as inflammatory bowel disease, smoking, hypothyroidism, pancreatitis, biliary obstruction, peptic ulcers, and cirrhosis—though CEA levels in these conditions are rarely >10 ng/mL.¹¹ Regardless of the results of biomarker testing, definitive diagnosis requires tissue biopsy; therefore, biomarker findings are of little utility in the initial workup.

In addition to variable diagnostic utility, overreliance on these biomarkers has the potential for serious patient harm. In a study examining patients with established rectal cancer, combination CEA and CA19-9 testing alone was insufficient to predict the pathologic stage of disease correctly.² A cancer misdiagnosis not only traumatizes patients but also erodes their trust in clinicians and creates anxiety during future clinical encounters. Overutilization of these tumor biomarkers is also costly and contributes to waste in the US healthcare system.

There is a role for tumor biomarker testing in specific cancers after the primary source of malignancy has been determined. When evaluating a known pelvic mass, CA125 testing is performed in conjunction with transvaginal ultrasound and assessment of menopausal status in the risk of ovarian malignancy algorithm for prognostication of disease prior to surgery.¹² This algorithm takes into account levels of CA125 in addition to levels of human epididymis protein 4 and patient age, yielding an area under the curve as high as 0.93 for ovarian cancer risk classification.⁸ Beyond the prognostication process, oncologists follow CA125 to monitor response to first-line ovarian cancer treatment. However, CA125 has a less defined role in surveillance for ovarian cancer recurrence.

CA19-9 has demonstrated utility for pancreatic cancer and cholangiocarcinoma survival estimates. A national cohort analysis of patients with established intrahepatic cholangiocarcinoma found that CA19-9 independently predicted increased mortality. Patients with elevated CA19-9 also had significantly more nodal metastases and positive-margin resections.⁶ A study of 353 patients with pancreatic ductal adenocarcinoma undergoing radical resection further demonstrated the utility of CA19-9. In this study, patients with postoperative CA19-9 normalization had improved survival by almost 12 months when compared to those with consistently elevated CA19-9.¹³

Last, the literature describes CEA biomarker testing in the surveillance of patients after curative treatment of colon and rectal cancer. The American Society of Colon and Rectal Surgeons recommends regularly tracking this biomarker following curative resection, in conjunction with colonoscopy and chest and liver imaging studies.¹⁴ A prospective randomized controlled study that followed this monitoring protocol in cured asymptomatic patients on a bimonthly basis found that early diagnosis of recurrent colorectal cancer improved survival.¹⁵ The use of CEA testing as a monitoring tool should therefore be a point of discussion between providers and patients, as its utility varies based on patient comorbidities, their ability to tolerate surgery or chemotherapy, risk factors for recurrence, performance status, compliance, age, and preference.¹⁴

The use of CA125, CA19-9, and CEA testing alone as initial diagnostic tools for malignancy are problematic due to their poor sensitivities and/or positive predictive value. Multiple studies have demonstrated their utility as markers of metastasis or malignancy progression rather than as clinically useful markers for the detection of any one type of cancer.^1,3,6 In an undiagnosed symptomatic patient with unexplained weight loss or symptoms of a tumor mass, elevated CA125, CA19-9, and CEA add no new information as metastatic pancreatic, colorectal, ovarian, gastric, gallbladder, hepatocellular, bladder, ovarian, and breast cancers all remain in the differential diagnosis. Clinicians should approach the initial diagnosis of cancer in such patients with appropriate imaging studies, a thorough physical examination, and prompt biopsy of abnormal findings, as long as these are consistent with the patient’s goals of care. After establishing a tissue diagnosis, some tumor biomarkers have valid prognostic, staging, and monitoring roles.^6,13,14

• Do not routinely order CA125, CA19-9, and CEA tests for the initial diagnostic workup of visceral malignancy of unknown origin regardless of whether imaging studies have been obtained.
• Use appropriate imaging, perform a thorough physical examination, and obtain tissue biopsy in the initial diagnostic workup of a visceral malignancy of unknown origin.

Clinicians should use serum biomarkers, like any other diagnostic test, to maximize benefit while preventing patient harm. In general, CA125, CA19-9, and CEA do not have a role in cancer diagnosis. The patient described in our clinical scenario would not benefit from a serum tumor biomarker panel at the time of admission. Regardless of findings from these tests, a tissue sample is required to make a definitive diagnosis of underlying malignancy in this patient.

Do you think this is a low-value practice? Is this truly a “Thing We Do for No Reason^™”? Share what you do in your practice and join in the conversation online by retweeting it on Twitter (#TWDFNR) and liking it on Facebook. We invite you to propose ideas for other “Things We Do for No Reason^™” topics by emailing [email protected]

Inspired by the ABIM Foundation’s Choosing Wisely^® campaign, the “Things We Do for No Reason^™” (TWDFNR) series reviews practices that have become common parts of hospital care but may provide little value to our patients. Practices reviewed in the TWDFNR series do not represent clear-cut conclusions or clinical practice standards but are meant as a starting place for research and active discussions among hospitalists and patients. We invite you to be part of that discussion.

A 56-year-old woman presents to the emergency department with a 2-week history of abdominal pain associated with nausea and an episode of nonbilious, nonbloody emesis. Her last bowel movement was 2 days prior to her presentation. The patient has tachycardia to 105 beats per minute but otherwise normal vital signs. Findings on her physical examination include dry mucous membranes and increased bowel sounds. A review of systems reveals an unintentional weight loss of 15 kg over the past 4 months and increased fatigue. Computed tomography scan of the abdomen and pelvis with contrast reveals multiple areas of attenuation in the liver and small bowel obstruction. The hospitalist admits the patient to the medicine service for supportive treatment and workup for underlying malignancy. Her admitting team orders serum tumor biomarkers on admission to expedite the diagnosis.

When patients present with unexplained weight loss or with metastasis from an unknown primary location, the initial workup often includes imaging and a tumor biomarker panel (eg, cancer antigen 125 [CA125], carbohydrate antigen 19-9 [CA19-9], carcinoembryonic antigen [CEA]). The CA125, CA19-9, and CEA biomarkers are traditionally associated with ovarian, pancreatic, and colorectal cancer, respectively.¹ While clinicians initially used these serum biomarkers to monitor for cancer recurrence or treatment response, they have since become widely used in multiple clinical stages of oncological evaluation.

Hospitalists routinely order biomarkers as part of the malignancy workup. More than a dozen oncology biomarkers are used in the clinical setting to risk stratify, plan treatment, and monitor for recurrence. For example, studies associate elevated preoperative levels of CEA and CA19-9 with metastatic invasion of colorectal² and gastric³ cancers and with poor prognosis of intrahepatic cholangiocarcinoma. Similarly, CA125 has demonstrated utility in monitoring response to ovarian cancer treatment.⁴ Specific biomarkers, such as alpha-fetoprotein, improve diagnosis of liver and nonseminomatous testicular tumors.⁵ Clinicians often apply the same paradigm to other biomarkers due to their widespread availability, noninvasiveness, reproducibility, and ease of use, particularly in acute settings wherein any new information is perceived to be potentially helpful.

Utilizing these serum biomarkers to diagnose cancer has the potential for diagnostic error and can result in unnecessary patient anxiety and follow-up testing. Since tissue sampling is necessary and remains the gold standard in most cancer diagnoses, obtaining these tumor biomarkers in the early diagnostic stage does not change management and may even lead to harm. Furthermore, due to their poor sensitivity and specificity, these biomarkers cannot rule in or rule out cancer. Elevated CA125, CA19-9, and CEA biomarkers occur in a variety of malignancies, including gastric, gallbladder, hepatocellular, bladder, and breast cancers.^1,3,6 In addition, these biomarkers have a very limited role in the workup of cancer of unknown primary origin.⁷

Even in the setting of a known pelvic mass, the use of CA125 alone has poor sensitivity at a cut-off level of 35 U/mL as a biomarker for the diagnosis of early ovarian cancer.⁸

Serum CA19-9 is not a useful diagnostic biomarker as elevated CA19-9 can occur in benign conditions, including cirrhosis, chronic pancreatitis, and cholangitis. In a systematic review of patients with histologic confirmation of pancreatic malignancy, the median positive predictive value of CA19-9 was 72% (interquartile range, 41%-95%).⁹ Additionally, patients with Lewis-null blood type, which is present in 5% to 10% of the Caucasian population, do not produce CA19-9.¹⁰ Therefore, CA19-9 will be 0% specific for tumors in this population.

The use of CEA in the diagnosis of colorectal cancer is also questionable. In stage I colorectal cancer, CEA was only 38.1% sensitive at a cut-off level of 2.41 ng/mL; it was 78.3% sensitive in stage IV disease.¹¹ The specificity of CEA is limited since elevated CEA occurs in benign conditions, such as inflammatory bowel disease, smoking, hypothyroidism, pancreatitis, biliary obstruction, peptic ulcers, and cirrhosis—though CEA levels in these conditions are rarely >10 ng/mL.¹¹ Regardless of the results of biomarker testing, definitive diagnosis requires tissue biopsy; therefore, biomarker findings are of little utility in the initial workup.

In addition to variable diagnostic utility, overreliance on these biomarkers has the potential for serious patient harm. In a study examining patients with established rectal cancer, combination CEA and CA19-9 testing alone was insufficient to predict the pathologic stage of disease correctly.² A cancer misdiagnosis not only traumatizes patients but also erodes their trust in clinicians and creates anxiety during future clinical encounters. Overutilization of these tumor biomarkers is also costly and contributes to waste in the US healthcare system.

There is a role for tumor biomarker testing in specific cancers after the primary source of malignancy has been determined. When evaluating a known pelvic mass, CA125 testing is performed in conjunction with transvaginal ultrasound and assessment of menopausal status in the risk of ovarian malignancy algorithm for prognostication of disease prior to surgery.¹² This algorithm takes into account levels of CA125 in addition to levels of human epididymis protein 4 and patient age, yielding an area under the curve as high as 0.93 for ovarian cancer risk classification.⁸ Beyond the prognostication process, oncologists follow CA125 to monitor response to first-line ovarian cancer treatment. However, CA125 has a less defined role in surveillance for ovarian cancer recurrence.

CA19-9 has demonstrated utility for pancreatic cancer and cholangiocarcinoma survival estimates. A national cohort analysis of patients with established intrahepatic cholangiocarcinoma found that CA19-9 independently predicted increased mortality. Patients with elevated CA19-9 also had significantly more nodal metastases and positive-margin resections.⁶ A study of 353 patients with pancreatic ductal adenocarcinoma undergoing radical resection further demonstrated the utility of CA19-9. In this study, patients with postoperative CA19-9 normalization had improved survival by almost 12 months when compared to those with consistently elevated CA19-9.¹³

Last, the literature describes CEA biomarker testing in the surveillance of patients after curative treatment of colon and rectal cancer. The American Society of Colon and Rectal Surgeons recommends regularly tracking this biomarker following curative resection, in conjunction with colonoscopy and chest and liver imaging studies.¹⁴ A prospective randomized controlled study that followed this monitoring protocol in cured asymptomatic patients on a bimonthly basis found that early diagnosis of recurrent colorectal cancer improved survival.¹⁵ The use of CEA testing as a monitoring tool should therefore be a point of discussion between providers and patients, as its utility varies based on patient comorbidities, their ability to tolerate surgery or chemotherapy, risk factors for recurrence, performance status, compliance, age, and preference.¹⁴

The use of CA125, CA19-9, and CEA testing alone as initial diagnostic tools for malignancy are problematic due to their poor sensitivities and/or positive predictive value. Multiple studies have demonstrated their utility as markers of metastasis or malignancy progression rather than as clinically useful markers for the detection of any one type of cancer.^1,3,6 In an undiagnosed symptomatic patient with unexplained weight loss or symptoms of a tumor mass, elevated CA125, CA19-9, and CEA add no new information as metastatic pancreatic, colorectal, ovarian, gastric, gallbladder, hepatocellular, bladder, ovarian, and breast cancers all remain in the differential diagnosis. Clinicians should approach the initial diagnosis of cancer in such patients with appropriate imaging studies, a thorough physical examination, and prompt biopsy of abnormal findings, as long as these are consistent with the patient’s goals of care. After establishing a tissue diagnosis, some tumor biomarkers have valid prognostic, staging, and monitoring roles.^6,13,14

• Do not routinely order CA125, CA19-9, and CEA tests for the initial diagnostic workup of visceral malignancy of unknown origin regardless of whether imaging studies have been obtained.
• Use appropriate imaging, perform a thorough physical examination, and obtain tissue biopsy in the initial diagnostic workup of a visceral malignancy of unknown origin.

Clinicians should use serum biomarkers, like any other diagnostic test, to maximize benefit while preventing patient harm. In general, CA125, CA19-9, and CEA do not have a role in cancer diagnosis. The patient described in our clinical scenario would not benefit from a serum tumor biomarker panel at the time of admission. Regardless of findings from these tests, a tissue sample is required to make a definitive diagnosis of underlying malignancy in this patient.

Do you think this is a low-value practice? Is this truly a “Thing We Do for No Reason^™”? Share what you do in your practice and join in the conversation online by retweeting it on Twitter (#TWDFNR) and liking it on Facebook. We invite you to propose ideas for other “Things We Do for No Reason^™” topics by emailing [email protected]

Inspired by the ABIM Foundation’s Choosing Wisely^® campaign, the “Things We Do for No Reason^™” (TWDFNR) series reviews practices that have become common parts of hospital care but may provide little value to our patients. Practices reviewed in the TWDFNR series do not represent clear-cut conclusions or clinical practice standards but are meant as a starting place for research and active discussions among hospitalists and patients. We invite you to be part of that discussion.

A hospitalist admits a 29-year-old man with hypertension, obesity, and type 2 diabetes (type 2 DM) for a posterior neck abscess that failed outpatient oral antibiotic therapy. The patient’s medications include metformin monotherapy. Vital signs taken upon admission include a blood pressure of 136/82 mm Hg, heart rate of 98 beats per minute, respiratory rate 18 of breaths per minute, oxygen saturation of 100% on room air, and temperature of 38.5 oC. Laboratory evaluation revealed a glucose level of 212 mg/dL, with a hemoglobin A1c of 8.0%, lactic acid of 1.4 mmol/L, and normal renal and hepatic function. Based on these findings, the hospitalist holds metformin and starts the patient on sliding-scale insulin therapy.

Metformin, an oral medication used to treat type 2 DM, is a biguanide that increases peripheral glucose utilization and decreases hepatic gluconeogenesis. However, metformin-associated shunting of metabolism toward anaerobic respiration increases the risk of lactic acidosis.¹ Because the kidneys excrete metformin, the risk of developing metformin-associated lactic acidosis (MALA) increases with renal impairment. Disease states common among hospitalized patients, such as hypoperfusion, advanced cirrhosis, alcohol abuse, cardiac failure, muscle ischemia, and severe infection, increase the risk of acute kidney injury (AKI) and elevate blood lactate levels. Therefore, hospitalists regularly hold metformin in the inpatient setting.

Following the introduction of metformin in the United States, the US Food and Drug Administration (FDA) received 47 confirmed reports of nonfatal lactic acidosis associated with the use of metformin, all of which involved cardiac disease (specifically congestive heart failure [CHF]), renal insufficiency, hypoxia, or sepsis.² Consequently, the FDA listed CHF as a contraindication to metformin use; however, it has since changed the use of metformin in CHF from a contraindication to a warning/precaution for lactic acidosis. The FDA also added a warning against the use of metformin in patients with sepsis or in patients older than 80 years who have abnormal creatinine clearance.

Acute kidney injury, a common inpatient condition, occurs in 20% of hospitalized patients and more than 50% of intensive care patients.³ Moreover, a retrospective observational study showed approximately 50% of all patients hospitalized for COVID-19 had AKI.⁴ Iodinated contrast, a diagnostic media commonly used in the hospital, may also increase the risk of renal dysfunction. The FDA recommends providers discontinue metformin at or before initiating imaging studies with iodinated contrast⁵ in patients with an estimated glomerular filtration rate (eGFR) between 30 and 60 mL/min/1.73 m². The FDA also advises that providers not restart metformin until 48 hours after an intra-arterial (IA) or intravenous (IV) contrast study in patients with an eGFR <60 mL/min/1.73 m² (equivalent to chronic kidney disease [CKD] stage 3 or worse).⁵ The American Diabetes Association (ADA) recommends the same eGFR cutoff level in its clinical practice recommendations, as well as withholding metformin 48 hours before patients receive IV contrast.⁶ Given the risk of AKI in hospitalized patients and concerns of increased MALA, clinicians reflexively hold metformin.

Holding metformin is also consistent with professional guidelines. The 2009 American Association of Clinical Endocrinology and ADA Consensus Statement on Inpatient Glycemic Control recommends cautious use of metformin in the inpatient setting “because of the potential development of a contraindication during the hospitalization.”⁷ Similarly, the 2012 Endocrine Society guidelines recommend withholding metformin in almost all hospitalized patients.⁸

Routinely holding metformin in hospitalized patients is unnecessary and potentially harmful. First, MALA is exceedingly rare, and experts question the causal link. Furthermore, iodinated contrast does not place patients with normal renal function at increased risk of MALA. Finally, holding metformin leads to worsened glycemic control and increased use of insulin, both of which may result in adverse patient outcomes.

The concerns about MALA stem from clinical experiences with phenformin, an older and more potent biguanide. Phenformin shares a similar mechanism of action with metformin but causes more lactic acid production. In 1978, following 306 documented cases of phenformin-associated lactic acidosis, the FDA removed this medication from the market.⁹ Since the initial 47 cases of MALA were reported to the FDA, repeated studies and systematic reviews have disputed the link between metformin and lactic acidosis, particularly in the absence of significant risk factors or in patients with an eGFR ≥30 mL/min/1.73 m². In fact, a large observational study showed a reduction in acidosis and mortality in outpatients with stage 3a CKD (eGFR, 45-59 mL/min/1.73 m²) who were taking metformin compared to patients taking insulin or other oral hypoglycemics agents.¹⁰ In patients with stage 3b CKD (eGFR, 30-44 mL/min/1.73 m²), this study found no difference in the same outcomes.¹⁰

Studies show that metformin does not cause elevated lactate levels in patients with stage 4 CKD (eGFR >15mL/min/1.73²) or lower stages of CKD as long as doses are adjusted appropriately to reflect renal function.¹¹ These and other investigations reveal that in the absence of other risk factors, metformin does not cause lactic acidosis (Table).^10-15 Based on these findings, the Endocrine Society changed the strength of its recommendation to withhold metformin in hospitalized patients to “weak,” with “very low-quality evidence.” The FDA similarly revised its warnings⁸ to allow metformin use in all patients with an eGFR ≥30 mL/min/1.73 m². A large community-based cohort study, which demonstrated no association between hospitalization with acidosis and metformin use in patients with stage 3b CKD or lower stages of CKD, supports this change in treatment threshold.¹⁵

Published evidence also does not support the practice of routinely holding metformin before contrast administration, despite concerns regarding contrast-induced nephropathy. Retrospective chart reviews and a direct comparison in human models have not shown any significant difference in the risk of AKI between the IV and IA contrast.¹⁶ Moreover, evidence suggests no interaction between metformin and contrast media in patients with normal renal function.¹⁷ In response, the American College of Radiology, Canadian Association of Radiology, Royal College of Radiologists, and Royal Australian and New Zealand College of Radiologists all recommend continuing metformin in patients with normal renal function (eGFR ≥30 mL/min/1.73m²) receiving IV contrast. They advise holding metformin for 48 hours in patients with renal insufficiency (eGFR <30 mL/min/1.73m²) or those undergoing IA catheter studies that might result in renal artery emboli.¹⁸

Finally, continuing metformin maintains steady blood glucose control. The practice of replacing metformin with sliding-scale insulin monotherapy for hospitalized patients significantly increases the risk of hyperglycemia and is associated with an increased length of stay.¹⁹ Additionally, unlike insulin, metformin does not increase the risk of hypoglycemia. Finally, a recent matched cohort study comparing the use of oral hypoglycemic agents (metformin, thiazolidines, and sulfonylureas) vs insulin monotherapy in patients undergoing emergency abdominal surgery showed that the patients admitted with sepsis and treated with oral agents had a lower 30-day mortality rate and a shorter length of stay.²⁰ Based on the evidence showing that inpatient oral hypoglycemic agents improve quality metrics and mitigate safety events, the ADA advocates resuming oral antihyperglycemic medications (most commonly metformin) 1 to 2 days before discharge.⁷

Clinicians should continue metformin in all hospitalized patients who are not at significant risk of developing lactic acidosis. Risk factors for MALA include severe sepsis (in the setting of end-organ damage as defined by systemic inflammatory response syndrome criteria), hypoxia requiring oxygen supplementation, hypoperfusion (as from CHF), AKI, CKD (eGFR <30 mL/min/1.73 m²), and advanced cirrhosis. Given the high rates of hypoxia and AKI in admitted patients with COVID-19, clinicians should hold metformin on admission. Continue metformin for patients receiving IV contrast media with an eGFR >30 mL/min/1.73 m². For patients undergoing IA catheter studies associated with a risk for renal artery emboli, or in patients with renal insufficiency (eGFR <30 mL/min/1.73 m²), temporarily hold metformin for 48 hours. When held, restart metformin as soon as risk factors resolve.

• Hold metformin in patients with or undergoing the following:
  • High risk for or currently suffering from decompensated heart failure, severe sepsis, or other disease states resulting in hypoxia or tissue hypoperfusion;
  • An eGFR <30 mL/min/1.73 m² or AKI; resume metformin when the AKI resolves;
  • COVID-19 infection, until the risk of hypoxia has resolved;
  • IV contrast study in the presence of acute renal failure or an eGFR <30 mL/min/1.73 m²; resume metformin 48 hours after contrast administration;
  • Intra-arterial catheter study that might result in renal artery emboli; resume metformin when renal function normalizes.
• Continue metformin in all hospitalized patients in the absence of the aforementioned disease states or contrast-related indications.

Returning to the patient in our clinical scenario, we recommend continuing metformin given the lack of risk factors or disease states associated with increased lactic acidosis. The practice of withholding metformin in hospitalized patients for fear of MALA is based on minimal evidence. Clinicians should, however, hold metformin in patients who have true contraindications, including existing acidosis, hypoperfusion, renal insufficiency, CHF, severe sepsis, hypoxia, advanced cirrhosis, and COVID-19. With regard to iodinated contrast studies, temporarily withhold metformin for 48 hours in patients with an eGFR <30 mL/min/1.73 m², acute kidney injury, or in patients undergoing an IA catheter study at risk for renal artery emboli. Patients should be restarted on metformin 48 hours after these studies and as renal function normalizes. When withholding metformin during a hospitalization, restart it once risk factors have resolved.

Do you think this is a low-value practice? Is this truly a “Thing We Do for No Reason^™”? Share what you do in your practice and join in the conversation online by retweeting it on Twitter  (#TWDFNR) and liking it on Facebook. We invite you to propose ideas for other “Things We Do for No Reason^™” topics by emailing [email protected]

Inspired by the ABIM Foundation’s Choosing Wisely^® campaign, the “Things We Do for No Reason^™” (TWDFNR) series reviews practices that have become common parts of hospital care but may provide little value to our patients. Practices reviewed in the TWDFNR series do not represent clear-cut conclusions or clinical practice standards but are meant as a starting place for research and active discussions among hospitalists and patients. We invite you to be part of that discussion.

A hospitalist admits a 29-year-old man with hypertension, obesity, and type 2 diabetes (type 2 DM) for a posterior neck abscess that failed outpatient oral antibiotic therapy. The patient’s medications include metformin monotherapy. Vital signs taken upon admission include a blood pressure of 136/82 mm Hg, heart rate of 98 beats per minute, respiratory rate 18 of breaths per minute, oxygen saturation of 100% on room air, and temperature of 38.5 oC. Laboratory evaluation revealed a glucose level of 212 mg/dL, with a hemoglobin A1c of 8.0%, lactic acid of 1.4 mmol/L, and normal renal and hepatic function. Based on these findings, the hospitalist holds metformin and starts the patient on sliding-scale insulin therapy.

Metformin, an oral medication used to treat type 2 DM, is a biguanide that increases peripheral glucose utilization and decreases hepatic gluconeogenesis. However, metformin-associated shunting of metabolism toward anaerobic respiration increases the risk of lactic acidosis.¹ Because the kidneys excrete metformin, the risk of developing metformin-associated lactic acidosis (MALA) increases with renal impairment. Disease states common among hospitalized patients, such as hypoperfusion, advanced cirrhosis, alcohol abuse, cardiac failure, muscle ischemia, and severe infection, increase the risk of acute kidney injury (AKI) and elevate blood lactate levels. Therefore, hospitalists regularly hold metformin in the inpatient setting.

Following the introduction of metformin in the United States, the US Food and Drug Administration (FDA) received 47 confirmed reports of nonfatal lactic acidosis associated with the use of metformin, all of which involved cardiac disease (specifically congestive heart failure [CHF]), renal insufficiency, hypoxia, or sepsis.² Consequently, the FDA listed CHF as a contraindication to metformin use; however, it has since changed the use of metformin in CHF from a contraindication to a warning/precaution for lactic acidosis. The FDA also added a warning against the use of metformin in patients with sepsis or in patients older than 80 years who have abnormal creatinine clearance.

Acute kidney injury, a common inpatient condition, occurs in 20% of hospitalized patients and more than 50% of intensive care patients.³ Moreover, a retrospective observational study showed approximately 50% of all patients hospitalized for COVID-19 had AKI.⁴ Iodinated contrast, a diagnostic media commonly used in the hospital, may also increase the risk of renal dysfunction. The FDA recommends providers discontinue metformin at or before initiating imaging studies with iodinated contrast⁵ in patients with an estimated glomerular filtration rate (eGFR) between 30 and 60 mL/min/1.73 m². The FDA also advises that providers not restart metformin until 48 hours after an intra-arterial (IA) or intravenous (IV) contrast study in patients with an eGFR <60 mL/min/1.73 m² (equivalent to chronic kidney disease [CKD] stage 3 or worse).⁵ The American Diabetes Association (ADA) recommends the same eGFR cutoff level in its clinical practice recommendations, as well as withholding metformin 48 hours before patients receive IV contrast.⁶ Given the risk of AKI in hospitalized patients and concerns of increased MALA, clinicians reflexively hold metformin.

Holding metformin is also consistent with professional guidelines. The 2009 American Association of Clinical Endocrinology and ADA Consensus Statement on Inpatient Glycemic Control recommends cautious use of metformin in the inpatient setting “because of the potential development of a contraindication during the hospitalization.”⁷ Similarly, the 2012 Endocrine Society guidelines recommend withholding metformin in almost all hospitalized patients.⁸

Routinely holding metformin in hospitalized patients is unnecessary and potentially harmful. First, MALA is exceedingly rare, and experts question the causal link. Furthermore, iodinated contrast does not place patients with normal renal function at increased risk of MALA. Finally, holding metformin leads to worsened glycemic control and increased use of insulin, both of which may result in adverse patient outcomes.

The concerns about MALA stem from clinical experiences with phenformin, an older and more potent biguanide. Phenformin shares a similar mechanism of action with metformin but causes more lactic acid production. In 1978, following 306 documented cases of phenformin-associated lactic acidosis, the FDA removed this medication from the market.⁹ Since the initial 47 cases of MALA were reported to the FDA, repeated studies and systematic reviews have disputed the link between metformin and lactic acidosis, particularly in the absence of significant risk factors or in patients with an eGFR ≥30 mL/min/1.73 m². In fact, a large observational study showed a reduction in acidosis and mortality in outpatients with stage 3a CKD (eGFR, 45-59 mL/min/1.73 m²) who were taking metformin compared to patients taking insulin or other oral hypoglycemics agents.¹⁰ In patients with stage 3b CKD (eGFR, 30-44 mL/min/1.73 m²), this study found no difference in the same outcomes.¹⁰

Studies show that metformin does not cause elevated lactate levels in patients with stage 4 CKD (eGFR >15mL/min/1.73²) or lower stages of CKD as long as doses are adjusted appropriately to reflect renal function.¹¹ These and other investigations reveal that in the absence of other risk factors, metformin does not cause lactic acidosis (Table).^10-15 Based on these findings, the Endocrine Society changed the strength of its recommendation to withhold metformin in hospitalized patients to “weak,” with “very low-quality evidence.” The FDA similarly revised its warnings⁸ to allow metformin use in all patients with an eGFR ≥30 mL/min/1.73 m². A large community-based cohort study, which demonstrated no association between hospitalization with acidosis and metformin use in patients with stage 3b CKD or lower stages of CKD, supports this change in treatment threshold.¹⁵

Published evidence also does not support the practice of routinely holding metformin before contrast administration, despite concerns regarding contrast-induced nephropathy. Retrospective chart reviews and a direct comparison in human models have not shown any significant difference in the risk of AKI between the IV and IA contrast.¹⁶ Moreover, evidence suggests no interaction between metformin and contrast media in patients with normal renal function.¹⁷ In response, the American College of Radiology, Canadian Association of Radiology, Royal College of Radiologists, and Royal Australian and New Zealand College of Radiologists all recommend continuing metformin in patients with normal renal function (eGFR ≥30 mL/min/1.73m²) receiving IV contrast. They advise holding metformin for 48 hours in patients with renal insufficiency (eGFR <30 mL/min/1.73m²) or those undergoing IA catheter studies that might result in renal artery emboli.¹⁸

Finally, continuing metformin maintains steady blood glucose control. The practice of replacing metformin with sliding-scale insulin monotherapy for hospitalized patients significantly increases the risk of hyperglycemia and is associated with an increased length of stay.¹⁹ Additionally, unlike insulin, metformin does not increase the risk of hypoglycemia. Finally, a recent matched cohort study comparing the use of oral hypoglycemic agents (metformin, thiazolidines, and sulfonylureas) vs insulin monotherapy in patients undergoing emergency abdominal surgery showed that the patients admitted with sepsis and treated with oral agents had a lower 30-day mortality rate and a shorter length of stay.²⁰ Based on the evidence showing that inpatient oral hypoglycemic agents improve quality metrics and mitigate safety events, the ADA advocates resuming oral antihyperglycemic medications (most commonly metformin) 1 to 2 days before discharge.⁷

Clinicians should continue metformin in all hospitalized patients who are not at significant risk of developing lactic acidosis. Risk factors for MALA include severe sepsis (in the setting of end-organ damage as defined by systemic inflammatory response syndrome criteria), hypoxia requiring oxygen supplementation, hypoperfusion (as from CHF), AKI, CKD (eGFR <30 mL/min/1.73 m²), and advanced cirrhosis. Given the high rates of hypoxia and AKI in admitted patients with COVID-19, clinicians should hold metformin on admission. Continue metformin for patients receiving IV contrast media with an eGFR >30 mL/min/1.73 m². For patients undergoing IA catheter studies associated with a risk for renal artery emboli, or in patients with renal insufficiency (eGFR <30 mL/min/1.73 m²), temporarily hold metformin for 48 hours. When held, restart metformin as soon as risk factors resolve.

• Hold metformin in patients with or undergoing the following:
  • High risk for or currently suffering from decompensated heart failure, severe sepsis, or other disease states resulting in hypoxia or tissue hypoperfusion;
  • An eGFR <30 mL/min/1.73 m² or AKI; resume metformin when the AKI resolves;
  • COVID-19 infection, until the risk of hypoxia has resolved;
  • IV contrast study in the presence of acute renal failure or an eGFR <30 mL/min/1.73 m²; resume metformin 48 hours after contrast administration;
  • Intra-arterial catheter study that might result in renal artery emboli; resume metformin when renal function normalizes.
• Continue metformin in all hospitalized patients in the absence of the aforementioned disease states or contrast-related indications.

Returning to the patient in our clinical scenario, we recommend continuing metformin given the lack of risk factors or disease states associated with increased lactic acidosis. The practice of withholding metformin in hospitalized patients for fear of MALA is based on minimal evidence. Clinicians should, however, hold metformin in patients who have true contraindications, including existing acidosis, hypoperfusion, renal insufficiency, CHF, severe sepsis, hypoxia, advanced cirrhosis, and COVID-19. With regard to iodinated contrast studies, temporarily withhold metformin for 48 hours in patients with an eGFR <30 mL/min/1.73 m², acute kidney injury, or in patients undergoing an IA catheter study at risk for renal artery emboli. Patients should be restarted on metformin 48 hours after these studies and as renal function normalizes. When withholding metformin during a hospitalization, restart it once risk factors have resolved.

Do you think this is a low-value practice? Is this truly a “Thing We Do for No Reason^™”? Share what you do in your practice and join in the conversation online by retweeting it on Twitter  (#TWDFNR) and liking it on Facebook. We invite you to propose ideas for other “Things We Do for No Reason^™” topics by emailing [email protected]

Inspired by the ABIM Foundation’s Choosing Wisely^® campaign, the “Things We Do for No Reason^™” (TWDFNR) series reviews practices that have become common parts of hospital care but may provide little value to our patients. Practices reviewed in the TWDFNR series do not represent clear-cut conclusions or clinical practice standards but are meant as a starting place for research and active discussions among hospitalists and patients. We invite you to be part of that discussion.

A hospitalist admits a 29-year-old man with hypertension, obesity, and type 2 diabetes (type 2 DM) for a posterior neck abscess that failed outpatient oral antibiotic therapy. The patient’s medications include metformin monotherapy. Vital signs taken upon admission include a blood pressure of 136/82 mm Hg, heart rate of 98 beats per minute, respiratory rate 18 of breaths per minute, oxygen saturation of 100% on room air, and temperature of 38.5 oC. Laboratory evaluation revealed a glucose level of 212 mg/dL, with a hemoglobin A1c of 8.0%, lactic acid of 1.4 mmol/L, and normal renal and hepatic function. Based on these findings, the hospitalist holds metformin and starts the patient on sliding-scale insulin therapy.

Metformin, an oral medication used to treat type 2 DM, is a biguanide that increases peripheral glucose utilization and decreases hepatic gluconeogenesis. However, metformin-associated shunting of metabolism toward anaerobic respiration increases the risk of lactic acidosis.¹ Because the kidneys excrete metformin, the risk of developing metformin-associated lactic acidosis (MALA) increases with renal impairment. Disease states common among hospitalized patients, such as hypoperfusion, advanced cirrhosis, alcohol abuse, cardiac failure, muscle ischemia, and severe infection, increase the risk of acute kidney injury (AKI) and elevate blood lactate levels. Therefore, hospitalists regularly hold metformin in the inpatient setting.

Following the introduction of metformin in the United States, the US Food and Drug Administration (FDA) received 47 confirmed reports of nonfatal lactic acidosis associated with the use of metformin, all of which involved cardiac disease (specifically congestive heart failure [CHF]), renal insufficiency, hypoxia, or sepsis.² Consequently, the FDA listed CHF as a contraindication to metformin use; however, it has since changed the use of metformin in CHF from a contraindication to a warning/precaution for lactic acidosis. The FDA also added a warning against the use of metformin in patients with sepsis or in patients older than 80 years who have abnormal creatinine clearance.

Acute kidney injury, a common inpatient condition, occurs in 20% of hospitalized patients and more than 50% of intensive care patients.³ Moreover, a retrospective observational study showed approximately 50% of all patients hospitalized for COVID-19 had AKI.⁴ Iodinated contrast, a diagnostic media commonly used in the hospital, may also increase the risk of renal dysfunction. The FDA recommends providers discontinue metformin at or before initiating imaging studies with iodinated contrast⁵ in patients with an estimated glomerular filtration rate (eGFR) between 30 and 60 mL/min/1.73 m². The FDA also advises that providers not restart metformin until 48 hours after an intra-arterial (IA) or intravenous (IV) contrast study in patients with an eGFR <60 mL/min/1.73 m² (equivalent to chronic kidney disease [CKD] stage 3 or worse).⁵ The American Diabetes Association (ADA) recommends the same eGFR cutoff level in its clinical practice recommendations, as well as withholding metformin 48 hours before patients receive IV contrast.⁶ Given the risk of AKI in hospitalized patients and concerns of increased MALA, clinicians reflexively hold metformin.

Holding metformin is also consistent with professional guidelines. The 2009 American Association of Clinical Endocrinology and ADA Consensus Statement on Inpatient Glycemic Control recommends cautious use of metformin in the inpatient setting “because of the potential development of a contraindication during the hospitalization.”⁷ Similarly, the 2012 Endocrine Society guidelines recommend withholding metformin in almost all hospitalized patients.⁸

Routinely holding metformin in hospitalized patients is unnecessary and potentially harmful. First, MALA is exceedingly rare, and experts question the causal link. Furthermore, iodinated contrast does not place patients with normal renal function at increased risk of MALA. Finally, holding metformin leads to worsened glycemic control and increased use of insulin, both of which may result in adverse patient outcomes.

The concerns about MALA stem from clinical experiences with phenformin, an older and more potent biguanide. Phenformin shares a similar mechanism of action with metformin but causes more lactic acid production. In 1978, following 306 documented cases of phenformin-associated lactic acidosis, the FDA removed this medication from the market.⁹ Since the initial 47 cases of MALA were reported to the FDA, repeated studies and systematic reviews have disputed the link between metformin and lactic acidosis, particularly in the absence of significant risk factors or in patients with an eGFR ≥30 mL/min/1.73 m². In fact, a large observational study showed a reduction in acidosis and mortality in outpatients with stage 3a CKD (eGFR, 45-59 mL/min/1.73 m²) who were taking metformin compared to patients taking insulin or other oral hypoglycemics agents.¹⁰ In patients with stage 3b CKD (eGFR, 30-44 mL/min/1.73 m²), this study found no difference in the same outcomes.¹⁰

Studies show that metformin does not cause elevated lactate levels in patients with stage 4 CKD (eGFR >15mL/min/1.73²) or lower stages of CKD as long as doses are adjusted appropriately to reflect renal function.¹¹ These and other investigations reveal that in the absence of other risk factors, metformin does not cause lactic acidosis (Table).^10-15 Based on these findings, the Endocrine Society changed the strength of its recommendation to withhold metformin in hospitalized patients to “weak,” with “very low-quality evidence.” The FDA similarly revised its warnings⁸ to allow metformin use in all patients with an eGFR ≥30 mL/min/1.73 m². A large community-based cohort study, which demonstrated no association between hospitalization with acidosis and metformin use in patients with stage 3b CKD or lower stages of CKD, supports this change in treatment threshold.¹⁵

Published evidence also does not support the practice of routinely holding metformin before contrast administration, despite concerns regarding contrast-induced nephropathy. Retrospective chart reviews and a direct comparison in human models have not shown any significant difference in the risk of AKI between the IV and IA contrast.¹⁶ Moreover, evidence suggests no interaction between metformin and contrast media in patients with normal renal function.¹⁷ In response, the American College of Radiology, Canadian Association of Radiology, Royal College of Radiologists, and Royal Australian and New Zealand College of Radiologists all recommend continuing metformin in patients with normal renal function (eGFR ≥30 mL/min/1.73m²) receiving IV contrast. They advise holding metformin for 48 hours in patients with renal insufficiency (eGFR <30 mL/min/1.73m²) or those undergoing IA catheter studies that might result in renal artery emboli.¹⁸

Finally, continuing metformin maintains steady blood glucose control. The practice of replacing metformin with sliding-scale insulin monotherapy for hospitalized patients significantly increases the risk of hyperglycemia and is associated with an increased length of stay.¹⁹ Additionally, unlike insulin, metformin does not increase the risk of hypoglycemia. Finally, a recent matched cohort study comparing the use of oral hypoglycemic agents (metformin, thiazolidines, and sulfonylureas) vs insulin monotherapy in patients undergoing emergency abdominal surgery showed that the patients admitted with sepsis and treated with oral agents had a lower 30-day mortality rate and a shorter length of stay.²⁰ Based on the evidence showing that inpatient oral hypoglycemic agents improve quality metrics and mitigate safety events, the ADA advocates resuming oral antihyperglycemic medications (most commonly metformin) 1 to 2 days before discharge.⁷

Clinicians should continue metformin in all hospitalized patients who are not at significant risk of developing lactic acidosis. Risk factors for MALA include severe sepsis (in the setting of end-organ damage as defined by systemic inflammatory response syndrome criteria), hypoxia requiring oxygen supplementation, hypoperfusion (as from CHF), AKI, CKD (eGFR <30 mL/min/1.73 m²), and advanced cirrhosis. Given the high rates of hypoxia and AKI in admitted patients with COVID-19, clinicians should hold metformin on admission. Continue metformin for patients receiving IV contrast media with an eGFR >30 mL/min/1.73 m². For patients undergoing IA catheter studies associated with a risk for renal artery emboli, or in patients with renal insufficiency (eGFR <30 mL/min/1.73 m²), temporarily hold metformin for 48 hours. When held, restart metformin as soon as risk factors resolve.

• Hold metformin in patients with or undergoing the following:
  • High risk for or currently suffering from decompensated heart failure, severe sepsis, or other disease states resulting in hypoxia or tissue hypoperfusion;
  • An eGFR <30 mL/min/1.73 m² or AKI; resume metformin when the AKI resolves;
  • COVID-19 infection, until the risk of hypoxia has resolved;
  • IV contrast study in the presence of acute renal failure or an eGFR <30 mL/min/1.73 m²; resume metformin 48 hours after contrast administration;
  • Intra-arterial catheter study that might result in renal artery emboli; resume metformin when renal function normalizes.
• Continue metformin in all hospitalized patients in the absence of the aforementioned disease states or contrast-related indications.

Returning to the patient in our clinical scenario, we recommend continuing metformin given the lack of risk factors or disease states associated with increased lactic acidosis. The practice of withholding metformin in hospitalized patients for fear of MALA is based on minimal evidence. Clinicians should, however, hold metformin in patients who have true contraindications, including existing acidosis, hypoperfusion, renal insufficiency, CHF, severe sepsis, hypoxia, advanced cirrhosis, and COVID-19. With regard to iodinated contrast studies, temporarily withhold metformin for 48 hours in patients with an eGFR <30 mL/min/1.73 m², acute kidney injury, or in patients undergoing an IA catheter study at risk for renal artery emboli. Patients should be restarted on metformin 48 hours after these studies and as renal function normalizes. When withholding metformin during a hospitalization, restart it once risk factors have resolved.

Do you think this is a low-value practice? Is this truly a “Thing We Do for No Reason^™”? Share what you do in your practice and join in the conversation online by retweeting it on Twitter  (#TWDFNR) and liking it on Facebook. We invite you to propose ideas for other “Things We Do for No Reason^™” topics by emailing [email protected]

Hospital readmissions are a significant problem in the United States, affecting 15% to 30% of discharges and incurring costs of more than $17 billion annually.¹ Timely posthospitalization follow-up visits are critical to ensure the effective transfer of patients to the outpatient setting; such visits reduce readmission rates as well as hospital length of stay and overall health care resource utilization.^2-4 Patients who receive inadequate follow-up care (ie, within 4 weeks of discharge) are significantly more likely to be readmitted than those who receive close follow-up care^.5

Due to the large clinical and financial consequences associated with hospital readmission, a variety of interventions have been studied, including home visits, telemonitoring, medication management, telephone calls, and postdischarge clinics.^6,7 While studies have not shown postdischarge clinics to be universally efficacious in reducing readmission rates, there is increasing evidence of reduced readmission rates in clinics that target high-risk patients (eg, patients with congestive heart failure [CHF]) rather than the total population.² A study by Hernandez et al that evaluated the relationship between early physician follow-up and 30-day readmissions showed a significantly lower readmission rate among hospitals with higher follow-up rates.⁸ Similarly, patients with CHF in a large, integrated health system who were seen within 7 days of discharge had an odds ratio (OR) of 0.81 (95% CI, 0.70-0.94) for 30-day readmissions.⁹

Transitions-of-care clinics (TOCC), designed to provide early postdischarge follow-up to high-risk patients, have been shown to reduce 30-day readmission rates,^3,4,10,11 especially in clinics that have same-physician follow-up visits rather than follow-up visits with a community primary care physician (PCP).¹² The most pronounced impact of postdischarge follow-up is seen in high-risk patients with high complexity or high severity of disease; however, complex rural patients are less likely to have access to specialty care.¹³ As a result, since rural residents must travel farther for specialty care, they are seen less frequently than their urban counterparts.^14,15

Prior to our TOCC initiative, the Iowa City VA (ICVA) ranked in the fifth quintile of the Veterans Health Administration (VHA) Strategic Analytics for Improvement and Learning model for hospital-wide readmissions (HWR), meaning that HWR at ICVA were higher than 80% of the other VHA healthcare centers. The low score in this metric was in part due to readmission rates in high-risk populations, including patients with CHF and those with high Care Assessment Need (CAN) scores. One concern was that the ICVA system serves many veterans from rural areas, some of whom must travel up to 200 miles to access inpatient and subspecialty care.

To meet these challenges, we implemented a TOCC to deliver timely postdischarge care focusing on high-risk and high-complexity patients. To address access-to-care issues of patients living in rural areas within the ICVA, we included virtual follow-up visits as a key component of our intervention.^16,17 The aim of this project was to decrease 30-day readmission rates of ICVA patients by 20% within 12 months of implementation.

Setting/Study Population

The ICVA serves 184,000 veterans stretched over 50 counties in eastern Iowa, western Illinois, and northern Missouri, with more than 60% of these patients residing in rural areas. Patients were initially eligible for the TOCC if they had an admission diagnosis of CHF and a CAN score > 85 at the time of discharge. The CAN score, developed by the VA to assess the risk of hospital readmission in individual patients, factors in several variables, including demographics, coexisting conditions, vital signs, utilization of services, pharmacy visits, and laboratory results. Patients in the top 5% (95-99) have a readmission rate of 20% at 90 days. Since the CAN is a proprietary tool, it may not be published in full; however, this assessment tool is commonly used and frequently cited in VA research.^18-22 The CAN score is expressed as a percentile ranging from 0 (lowest risk) to 99 (highest risk). Patient eligibility was expanded during subsequent Plan-Do-Study-Act (PDSA) cycles, as outlined below. Patient eligibility was expanded during subsequent PDSA cycles (also outlined below). A review by a local institutional review board was obtained, and the study was classified as exempt due to the use of deidentified data. Standards for Quality Improvement Reporting Excellence 2.0 guidelines were used to construct the manuscript.

Magnitude Assessment

The numbers of discharges, readmissions within 30 days, emergency department (ED) visits by all discharged veterans, and veterans discharged with a CHF hospital diagnosis were recorded from February 2017 to February 2018, which were the 12 months immediately preceding the pilot implementation.

Intervention

The primary intervention was referral to the newly formed ICVA TOCC. The multidisciplinary TOCC team consisted of hospitalists, pharmacists, schedulers, and discharge planners/care managers. Patients were identified by the hospitalist team during admission; prior to hospital discharge, these patients were referred to TOCC discharge planners to schedule appropriate follow-up appointments. Virtual follow-up visits were conducted using a patient’s home technology; in cases where a patient lacked adequate technology capabilities (eg, no computer or internet access), the ICVA provided a tablet device with cellular internet capability for temporary use. Specific clinical activities included medication reconciliation by a pharmacist, follow-up of pending laboratory studies, imaging studies, pathology results, medical diagnosis education, counseling regarding dietary restrictions, and contingency planning outside of an ED visit in the event of a change in clinical status. In addition, the TOCC aimed to facilitate a smooth transition of care back to the PCP by arranging follow-up appointments, providing visit summaries, and scheduling consults with specialty care, as appropriate.

Measures

The primary objective measure was the 30-day readmission rate in the ICVA hospital. Secondary measures included the number of VHA ED visits within 30 days of discharge. The main process measures were the number of hospital discharges per month, the number of TOCC referrals, the number of TOCC appointments made, the number of virtual and in-person visits, and the percentage of appointment “no-shows.”

Implementation

The TOCC was piloted from April 2018 to October 2018. During the pilot phase, TOCC enrollment was limited to virtual appointments and to patients with an admission diagnosis of CHF and a CAN score of > 85. The TOCC had staff on-site 2 days a week; this included pharmacists to reconcile medications and hospitalists to address follow-up care needs.

The TOCC clinic was temporarily closed at the end of October 2018 to analyze pilot results. Based on stakeholder feedback, changes made as part of the second PDSA cycle included expanding eligibility criteria to any hospital admission diagnosis and to patients with a CAN score < 85 if the hospitalist team felt the patient was likely to benefit from TOCC follow-up. In addition, on-site clinic staffing was expanded from 2 to 5 days per week to improve access, and the option for an in-person visit was added based on concerns some veterans expressed regarding the use of the technology at home. Finally, a formal resident program was added, and the order set for referrals was simplified. The TOCC was restarted in February 2019, and TOCC metrics were reviewed monthly. By July 2019, we identified issues with TOCC referrals and appointment creation that required additional modifications to the intervention.

A third PDSA cycle was initiated in July 2019 and included major changes, notably the formation of a designated TOCC committee. The committee appointed a dedicated TOCC scheduler whose role was to reduce confusion regarding scheduling, to update the discharge instructions/orders template to lower incidences of “double-booking” that occurred with PCP and TOCC appointments, to modify discharge educational instruction regarding virtual visits and tablet use, to adjust the TOCC-PCP handoff, and to formalize interactions between discharge coordinators and residents to review possible referrals every morning (Appendix Figure 1).

Statistical Analysis

Run charts were constructed by plotting monthly primary outcome values and monthly process metrics (Figure, Appendix Figure 2, Appendix Figure 3). Chi-square tests were used to compare 30-day readmission rates before and after the intervention. Statistical modeling was used to determine differences in outcomes between referred patients seen and referred patients not seen by the TOCC. In these statistical models, the outcome measures were 30-day readmissions, 30-day ED visits, and 6-month mortality. Covariates included in the final analysis were age, gender, race, CAN score, rural-urban commuting area code, referral service (resident vs nonresident), and admission diagnosis. Admission diagnoses were sorted by the investigators into one of the following seven categories: cardiac, infectious, pulmonary, gastrointestinal, neurologic, renal, and other.

Mean (SD) or counts and percentages were used to describe the distribution of continuous and categorical variables, respectively. Kruskal-Wallis test, t test, or chi-square tests were used, as appropriate, across categories. Generalized linear models with a logistic link function were used to test for differences between patients who kept their appointment at the TOCC and those who did not keep their TOCC appointment (both unadjusted and adjusted for all of the covariates previously mentioned). In addition, generalized linear models were also used to compare outcomes between TOCC patients seen virtually vs those seen in-person (both unadjusted and adjusted for all the covariates previously mentioned). All statistical tests were considered significant at a two-sided P < .05. All analyses were performed using SAS software version 9.4 (SAS Institute Inc).

Magnitude Assessment

During the preimplementation period (February 2017-February 2018), there were 3014 patient discharges from ICVA and 343 readmissions, resulting in a readmission rate of 11.4%. Among patients with a hospital-admission diagnosis of cardiorespiratory disease, which included patients with CHF, there were 381 discharges and 46 readmissions, resulting in a readmission rate of 12.1%.

Primary Outcome

During the pilot phase, which was conducted from April 2018 to October 2018, 142 patients who met inclusion criteria (CHF diagnosis and a CAN score > 85) were discharged from ICVA, and 56 referrals to the TOCC were placed. The readmission rate among the cardiorespiratory cohort of veterans was 9.5%.

During the expansion of the intervention from February 2019 to February 2020, there were 2844 discharges from the ICVA and 291 readmissions, resulting in a readmission rate of 10.2%. However, there was a further decrease in the readmission rate after the third PDSA cycle was initiated in July 2019 (Appendix Figure 1). The readmission rate was 9.2% in the final 6 months of the intervention period, and 7.9% in the final 3 months. Of note, in the group of 1948 patients who did not meet the eligibility criteria to participate in our study, the readmission rate during the same time period was 8.6% (161 readmissions).

When comparing the 6 months following the third PDSA cycle to the magnitude assessment period, there was a relative readmission reduction of 19.3% (P = .04), and an absolute reduction of 2.2%. If the final 3 months of the intervention period are included, there was an absolute reduction of 3.5% and a relative reduction of 30.7% (P = .01). Notably, before the pilot phase, ICVA was in the fifth quintile for HWR among VA hospitals but improved to the second quintile by the end of the expansion phase.

Process Outcomes

Process metrics for TOCC referrals, the number of patients seen, and the number of virtual and in-person visits over time are shown in Appendix Figure 3. Rates of TOCC referrals and the number of TOCC visits were lower than anticipated during the first 5 months of the intervention. However, TOCC referrals increased significantly after we implemented the previously described changes as part of the third PDSA cycle. As a result, total, virtual, and in-person visits also significantly increased from July 2019 to February 2020. The proportion of patients choosing virtual vs in-person visits fluctuated over time, but virtual visits were generally chosen more often than in-person visits.

Statistical Modeling

Baseline Data

Cohort characteristics are shown in Table 1. The cohort, which reflected the ICVA population, was predominantly male (96%) and White (93%), with a mean age of 67 years. The population was approximately half urban and half rural in composition, and the most common reason for hospital admission was cardiac. Other than a small but statistically significant difference in CAN scores, there were no significant differences between patients who kept their TOCC appointment and those who did not. There were also no differences in baseline characteristics between patients who chose virtual follow-up and patients who chose in-person follow-up, including the proportion of urban and rural patients.

Outcomes

Patients who kept their TOCC appointments had a 30-day readmission rate of 9.6%, which was significantly lower than the 30-day readmission rate of 27% in the group that did not keep their TOCC appointment (P < .001). Similarly, the percentage of patients treated in the ED was 15% in the TOCC group compared to 31.2% in the group that canceled their appointment (P < .001) (Table 1). In the multivariable analysis, patients who were seen in the TOCC group had an OR for 30-day readmission of 0.35 (95% CI, 0.19-0.62, P < .001), and an OR for ED visits of 0.39 (95% CI, 0.23-0.65; P < .001) (Table 2). There was no statistically significant difference in 6-month mortality between the two groups. In the virtual group compared to the in-person group, there were no statistically significant differences in outcomes between the two groups in the unadjusted or adjusted analysis (Table 2).

In this quality improvement initiative, we found that a TOCC targeting high-risk patients and offering virtual follow-up visits significantly decreased the 30-day readmission rates among veterans at ICVA. Statistical comparisons of patients seen at the TOCC vs those not seen at the TOCC showed a dramatic reduction in 30-day readmissions and ED visits. Finally, virtual follow-ups were more popular than in-person visits, and patients who followed up virtually had equivalent outcomes to those with the more traditional follow-up.

In the expansion phase, eligibility was expanded to include any hospital indication but continued to focus on high-risk patients. Existing literature suggests that providing postdischarge care to all patients, including low- or medium-risk patients, may not be as impactful as enrolling high-risk patients only. For instance, a postdischarge clinic offered to all patients at a VA system in Colorado did not reduce readmission rates compared to PCP follow-up.²³ In contrast, a study of more than 10,000 high-risk urban patients demonstrated that postdischarge care resulted in a 9.3% reduction in readmission risk.²⁴ Our data are consistent with the previously published studies, as the average CAN score of patients seen in TOCC was 90, suggesting a high risk of readmission. In the final 12 months of the intervention, 15% of discharged patients were seen at the TOCC clinic, suggesting that targeted intervention within the small subset of high-risk patients was sufficient to achieve our primary aim. Of note, among patients who did not meet the inclusion criteria for TOCC referral (ie, patients not considered high risk [CAN score ≤ 85]), the rate of readmissions was 8.6%.

Most of the available research on the efficacy of postdischarge clinics was conducted in urban environments. Our ICVA population sees a large proportion of rural veterans, who account for just over 50% of the discharge population. In a study of more than 2 million Medicare patients discharged from US hospitals, the 30-day readmission rates and adjusted mortality rates were higher among patients in rural counties, and post–acute care seemed to have a greater impact in rural rather than urban settings.²⁵ Previous studies have demonstrated that virtual visits have the potential to improve readmission rates, especially in patients with CHF²⁶ and in patients at the highest risk for readmission.²⁷ In our study, the pilot phase offered only virtual visits, but we subsequently added an in-person option based on veteran feedback. Interestingly, over the next 12 months, virtual visits were more popular with both urban and rural veterans, and there were no differences in the number of rural patients in the in-person vs the virtual group. These findings suggest factors other than rurality influenced the decision to choose virtual follow-up visits over in-person visits. Future studies should seek to determine the extent to which factors such as age, race, educational level, and socioeconomic circumstances impact veterans’ follow-up decisions. Not only were outcomes among patients who chose virtual visits the same as those of patients who chose in-person visits, but both of these groups had better outcomes compared to the non-TOCC group (Table 2). This finding demonstrating the efficacy of virtual visits among rural and urban patients has taken on increased significance due to the COVID-19 pandemic, as virtual visits offer a safer option, one that minimizes physical contact.

Our quality improvement analysis included a statistical comparison of patients seen vs those not seen at the TOCC. Patients who were referred to the TOCC but chose not to keep their appointment were similar to those seen in TOCC in terms of age, CAN score, rurality, and hospital diagnosis, but readmission rates were substantially higher in this group even after adjustments for covariates (Table 2). Evaluating causality in interventions aimed to reduce hospital readmission rates is complicated.²⁸ Our findings add greater plausibility to the utility of TOCC in accounting for at least a portion of the reported decrease in ICVA 30-day readmissions.

Our study has several strengths, including an observation period longer than 2 years, a large population of discharged veterans within an integrated healthcare system, and a large proportion of patients living in rural areas. Another strength of our study is the innovative nature of the intervention, which features a multidisciplinary team and the option of virtual or in-person visits. Nevertheless, this study also has several important limitations. As a single-center study, our findings may not be generalizable to other institutions, especially those outside the VHA system. Similarly, our study population reflected that of the ICVA, which may limit generalizability to a more diverse population. While we attempted to account in our statistical modeling for baseline differences between referred patients seen by the TOCC and those referred but not seen, we cannot exclude residual confounding between the groups. Specifically, the comparison of patients who did and did not choose TOCC follow-up introduces the possibility of selection bias. Future randomized/controlled studies will need to evaluate whether TOCC is more effective than the standard of care to reduce readmissions. Finally, since the analysis period following the final PDSA cycle was compressed due to the onset of the COVID-19 pandemic in the United States, no data are available regarding the sustained impacts of changes made during this cycle.

A multidisciplinary TOCC within the ICVA, featuring both virtual and in-person visits, reduced 30-day readmission rates by 19.3%; this approach to care was especially effective in patients with CHF. Virtual visits were the follow-up mode of choice for both urban and rural veterans, and there was no difference in outcomes between these two follow-up options. Future studies will focus on additional quality metrics, including cost-effectiveness and patient satisfaction.

Hospital readmissions are a significant problem in the United States, affecting 15% to 30% of discharges and incurring costs of more than $17 billion annually.¹ Timely posthospitalization follow-up visits are critical to ensure the effective transfer of patients to the outpatient setting; such visits reduce readmission rates as well as hospital length of stay and overall health care resource utilization.^2-4 Patients who receive inadequate follow-up care (ie, within 4 weeks of discharge) are significantly more likely to be readmitted than those who receive close follow-up care^.5

Due to the large clinical and financial consequences associated with hospital readmission, a variety of interventions have been studied, including home visits, telemonitoring, medication management, telephone calls, and postdischarge clinics.^6,7 While studies have not shown postdischarge clinics to be universally efficacious in reducing readmission rates, there is increasing evidence of reduced readmission rates in clinics that target high-risk patients (eg, patients with congestive heart failure [CHF]) rather than the total population.² A study by Hernandez et al that evaluated the relationship between early physician follow-up and 30-day readmissions showed a significantly lower readmission rate among hospitals with higher follow-up rates.⁸ Similarly, patients with CHF in a large, integrated health system who were seen within 7 days of discharge had an odds ratio (OR) of 0.81 (95% CI, 0.70-0.94) for 30-day readmissions.⁹

Transitions-of-care clinics (TOCC), designed to provide early postdischarge follow-up to high-risk patients, have been shown to reduce 30-day readmission rates,^3,4,10,11 especially in clinics that have same-physician follow-up visits rather than follow-up visits with a community primary care physician (PCP).¹² The most pronounced impact of postdischarge follow-up is seen in high-risk patients with high complexity or high severity of disease; however, complex rural patients are less likely to have access to specialty care.¹³ As a result, since rural residents must travel farther for specialty care, they are seen less frequently than their urban counterparts.^14,15

Prior to our TOCC initiative, the Iowa City VA (ICVA) ranked in the fifth quintile of the Veterans Health Administration (VHA) Strategic Analytics for Improvement and Learning model for hospital-wide readmissions (HWR), meaning that HWR at ICVA were higher than 80% of the other VHA healthcare centers. The low score in this metric was in part due to readmission rates in high-risk populations, including patients with CHF and those with high Care Assessment Need (CAN) scores. One concern was that the ICVA system serves many veterans from rural areas, some of whom must travel up to 200 miles to access inpatient and subspecialty care.

To meet these challenges, we implemented a TOCC to deliver timely postdischarge care focusing on high-risk and high-complexity patients. To address access-to-care issues of patients living in rural areas within the ICVA, we included virtual follow-up visits as a key component of our intervention.^16,17 The aim of this project was to decrease 30-day readmission rates of ICVA patients by 20% within 12 months of implementation.

Setting/Study Population

The ICVA serves 184,000 veterans stretched over 50 counties in eastern Iowa, western Illinois, and northern Missouri, with more than 60% of these patients residing in rural areas. Patients were initially eligible for the TOCC if they had an admission diagnosis of CHF and a CAN score > 85 at the time of discharge. The CAN score, developed by the VA to assess the risk of hospital readmission in individual patients, factors in several variables, including demographics, coexisting conditions, vital signs, utilization of services, pharmacy visits, and laboratory results. Patients in the top 5% (95-99) have a readmission rate of 20% at 90 days. Since the CAN is a proprietary tool, it may not be published in full; however, this assessment tool is commonly used and frequently cited in VA research.^18-22 The CAN score is expressed as a percentile ranging from 0 (lowest risk) to 99 (highest risk). Patient eligibility was expanded during subsequent Plan-Do-Study-Act (PDSA) cycles, as outlined below. Patient eligibility was expanded during subsequent PDSA cycles (also outlined below). A review by a local institutional review board was obtained, and the study was classified as exempt due to the use of deidentified data. Standards for Quality Improvement Reporting Excellence 2.0 guidelines were used to construct the manuscript.

Magnitude Assessment

The numbers of discharges, readmissions within 30 days, emergency department (ED) visits by all discharged veterans, and veterans discharged with a CHF hospital diagnosis were recorded from February 2017 to February 2018, which were the 12 months immediately preceding the pilot implementation.

Intervention

The primary intervention was referral to the newly formed ICVA TOCC. The multidisciplinary TOCC team consisted of hospitalists, pharmacists, schedulers, and discharge planners/care managers. Patients were identified by the hospitalist team during admission; prior to hospital discharge, these patients were referred to TOCC discharge planners to schedule appropriate follow-up appointments. Virtual follow-up visits were conducted using a patient’s home technology; in cases where a patient lacked adequate technology capabilities (eg, no computer or internet access), the ICVA provided a tablet device with cellular internet capability for temporary use. Specific clinical activities included medication reconciliation by a pharmacist, follow-up of pending laboratory studies, imaging studies, pathology results, medical diagnosis education, counseling regarding dietary restrictions, and contingency planning outside of an ED visit in the event of a change in clinical status. In addition, the TOCC aimed to facilitate a smooth transition of care back to the PCP by arranging follow-up appointments, providing visit summaries, and scheduling consults with specialty care, as appropriate.

Measures

The primary objective measure was the 30-day readmission rate in the ICVA hospital. Secondary measures included the number of VHA ED visits within 30 days of discharge. The main process measures were the number of hospital discharges per month, the number of TOCC referrals, the number of TOCC appointments made, the number of virtual and in-person visits, and the percentage of appointment “no-shows.”

Implementation

The TOCC was piloted from April 2018 to October 2018. During the pilot phase, TOCC enrollment was limited to virtual appointments and to patients with an admission diagnosis of CHF and a CAN score of > 85. The TOCC had staff on-site 2 days a week; this included pharmacists to reconcile medications and hospitalists to address follow-up care needs.

The TOCC clinic was temporarily closed at the end of October 2018 to analyze pilot results. Based on stakeholder feedback, changes made as part of the second PDSA cycle included expanding eligibility criteria to any hospital admission diagnosis and to patients with a CAN score < 85 if the hospitalist team felt the patient was likely to benefit from TOCC follow-up. In addition, on-site clinic staffing was expanded from 2 to 5 days per week to improve access, and the option for an in-person visit was added based on concerns some veterans expressed regarding the use of the technology at home. Finally, a formal resident program was added, and the order set for referrals was simplified. The TOCC was restarted in February 2019, and TOCC metrics were reviewed monthly. By July 2019, we identified issues with TOCC referrals and appointment creation that required additional modifications to the intervention.

A third PDSA cycle was initiated in July 2019 and included major changes, notably the formation of a designated TOCC committee. The committee appointed a dedicated TOCC scheduler whose role was to reduce confusion regarding scheduling, to update the discharge instructions/orders template to lower incidences of “double-booking” that occurred with PCP and TOCC appointments, to modify discharge educational instruction regarding virtual visits and tablet use, to adjust the TOCC-PCP handoff, and to formalize interactions between discharge coordinators and residents to review possible referrals every morning (Appendix Figure 1).

Statistical Analysis

Run charts were constructed by plotting monthly primary outcome values and monthly process metrics (Figure, Appendix Figure 2, Appendix Figure 3). Chi-square tests were used to compare 30-day readmission rates before and after the intervention. Statistical modeling was used to determine differences in outcomes between referred patients seen and referred patients not seen by the TOCC. In these statistical models, the outcome measures were 30-day readmissions, 30-day ED visits, and 6-month mortality. Covariates included in the final analysis were age, gender, race, CAN score, rural-urban commuting area code, referral service (resident vs nonresident), and admission diagnosis. Admission diagnoses were sorted by the investigators into one of the following seven categories: cardiac, infectious, pulmonary, gastrointestinal, neurologic, renal, and other.

Mean (SD) or counts and percentages were used to describe the distribution of continuous and categorical variables, respectively. Kruskal-Wallis test, t test, or chi-square tests were used, as appropriate, across categories. Generalized linear models with a logistic link function were used to test for differences between patients who kept their appointment at the TOCC and those who did not keep their TOCC appointment (both unadjusted and adjusted for all of the covariates previously mentioned). In addition, generalized linear models were also used to compare outcomes between TOCC patients seen virtually vs those seen in-person (both unadjusted and adjusted for all the covariates previously mentioned). All statistical tests were considered significant at a two-sided P < .05. All analyses were performed using SAS software version 9.4 (SAS Institute Inc).

Magnitude Assessment

During the preimplementation period (February 2017-February 2018), there were 3014 patient discharges from ICVA and 343 readmissions, resulting in a readmission rate of 11.4%. Among patients with a hospital-admission diagnosis of cardiorespiratory disease, which included patients with CHF, there were 381 discharges and 46 readmissions, resulting in a readmission rate of 12.1%.

Primary Outcome

During the pilot phase, which was conducted from April 2018 to October 2018, 142 patients who met inclusion criteria (CHF diagnosis and a CAN score > 85) were discharged from ICVA, and 56 referrals to the TOCC were placed. The readmission rate among the cardiorespiratory cohort of veterans was 9.5%.

During the expansion of the intervention from February 2019 to February 2020, there were 2844 discharges from the ICVA and 291 readmissions, resulting in a readmission rate of 10.2%. However, there was a further decrease in the readmission rate after the third PDSA cycle was initiated in July 2019 (Appendix Figure 1). The readmission rate was 9.2% in the final 6 months of the intervention period, and 7.9% in the final 3 months. Of note, in the group of 1948 patients who did not meet the eligibility criteria to participate in our study, the readmission rate during the same time period was 8.6% (161 readmissions).

When comparing the 6 months following the third PDSA cycle to the magnitude assessment period, there was a relative readmission reduction of 19.3% (P = .04), and an absolute reduction of 2.2%. If the final 3 months of the intervention period are included, there was an absolute reduction of 3.5% and a relative reduction of 30.7% (P = .01). Notably, before the pilot phase, ICVA was in the fifth quintile for HWR among VA hospitals but improved to the second quintile by the end of the expansion phase.

Process Outcomes

Process metrics for TOCC referrals, the number of patients seen, and the number of virtual and in-person visits over time are shown in Appendix Figure 3. Rates of TOCC referrals and the number of TOCC visits were lower than anticipated during the first 5 months of the intervention. However, TOCC referrals increased significantly after we implemented the previously described changes as part of the third PDSA cycle. As a result, total, virtual, and in-person visits also significantly increased from July 2019 to February 2020. The proportion of patients choosing virtual vs in-person visits fluctuated over time, but virtual visits were generally chosen more often than in-person visits.

Statistical Modeling

Baseline Data

Cohort characteristics are shown in Table 1. The cohort, which reflected the ICVA population, was predominantly male (96%) and White (93%), with a mean age of 67 years. The population was approximately half urban and half rural in composition, and the most common reason for hospital admission was cardiac. Other than a small but statistically significant difference in CAN scores, there were no significant differences between patients who kept their TOCC appointment and those who did not. There were also no differences in baseline characteristics between patients who chose virtual follow-up and patients who chose in-person follow-up, including the proportion of urban and rural patients.

Outcomes

Patients who kept their TOCC appointments had a 30-day readmission rate of 9.6%, which was significantly lower than the 30-day readmission rate of 27% in the group that did not keep their TOCC appointment (P < .001). Similarly, the percentage of patients treated in the ED was 15% in the TOCC group compared to 31.2% in the group that canceled their appointment (P < .001) (Table 1). In the multivariable analysis, patients who were seen in the TOCC group had an OR for 30-day readmission of 0.35 (95% CI, 0.19-0.62, P < .001), and an OR for ED visits of 0.39 (95% CI, 0.23-0.65; P < .001) (Table 2). There was no statistically significant difference in 6-month mortality between the two groups. In the virtual group compared to the in-person group, there were no statistically significant differences in outcomes between the two groups in the unadjusted or adjusted analysis (Table 2).

In this quality improvement initiative, we found that a TOCC targeting high-risk patients and offering virtual follow-up visits significantly decreased the 30-day readmission rates among veterans at ICVA. Statistical comparisons of patients seen at the TOCC vs those not seen at the TOCC showed a dramatic reduction in 30-day readmissions and ED visits. Finally, virtual follow-ups were more popular than in-person visits, and patients who followed up virtually had equivalent outcomes to those with the more traditional follow-up.

In the expansion phase, eligibility was expanded to include any hospital indication but continued to focus on high-risk patients. Existing literature suggests that providing postdischarge care to all patients, including low- or medium-risk patients, may not be as impactful as enrolling high-risk patients only. For instance, a postdischarge clinic offered to all patients at a VA system in Colorado did not reduce readmission rates compared to PCP follow-up.²³ In contrast, a study of more than 10,000 high-risk urban patients demonstrated that postdischarge care resulted in a 9.3% reduction in readmission risk.²⁴ Our data are consistent with the previously published studies, as the average CAN score of patients seen in TOCC was 90, suggesting a high risk of readmission. In the final 12 months of the intervention, 15% of discharged patients were seen at the TOCC clinic, suggesting that targeted intervention within the small subset of high-risk patients was sufficient to achieve our primary aim. Of note, among patients who did not meet the inclusion criteria for TOCC referral (ie, patients not considered high risk [CAN score ≤ 85]), the rate of readmissions was 8.6%.

Most of the available research on the efficacy of postdischarge clinics was conducted in urban environments. Our ICVA population sees a large proportion of rural veterans, who account for just over 50% of the discharge population. In a study of more than 2 million Medicare patients discharged from US hospitals, the 30-day readmission rates and adjusted mortality rates were higher among patients in rural counties, and post–acute care seemed to have a greater impact in rural rather than urban settings.²⁵ Previous studies have demonstrated that virtual visits have the potential to improve readmission rates, especially in patients with CHF²⁶ and in patients at the highest risk for readmission.²⁷ In our study, the pilot phase offered only virtual visits, but we subsequently added an in-person option based on veteran feedback. Interestingly, over the next 12 months, virtual visits were more popular with both urban and rural veterans, and there were no differences in the number of rural patients in the in-person vs the virtual group. These findings suggest factors other than rurality influenced the decision to choose virtual follow-up visits over in-person visits. Future studies should seek to determine the extent to which factors such as age, race, educational level, and socioeconomic circumstances impact veterans’ follow-up decisions. Not only were outcomes among patients who chose virtual visits the same as those of patients who chose in-person visits, but both of these groups had better outcomes compared to the non-TOCC group (Table 2). This finding demonstrating the efficacy of virtual visits among rural and urban patients has taken on increased significance due to the COVID-19 pandemic, as virtual visits offer a safer option, one that minimizes physical contact.

Our quality improvement analysis included a statistical comparison of patients seen vs those not seen at the TOCC. Patients who were referred to the TOCC but chose not to keep their appointment were similar to those seen in TOCC in terms of age, CAN score, rurality, and hospital diagnosis, but readmission rates were substantially higher in this group even after adjustments for covariates (Table 2). Evaluating causality in interventions aimed to reduce hospital readmission rates is complicated.²⁸ Our findings add greater plausibility to the utility of TOCC in accounting for at least a portion of the reported decrease in ICVA 30-day readmissions.

Our study has several strengths, including an observation period longer than 2 years, a large population of discharged veterans within an integrated healthcare system, and a large proportion of patients living in rural areas. Another strength of our study is the innovative nature of the intervention, which features a multidisciplinary team and the option of virtual or in-person visits. Nevertheless, this study also has several important limitations. As a single-center study, our findings may not be generalizable to other institutions, especially those outside the VHA system. Similarly, our study population reflected that of the ICVA, which may limit generalizability to a more diverse population. While we attempted to account in our statistical modeling for baseline differences between referred patients seen by the TOCC and those referred but not seen, we cannot exclude residual confounding between the groups. Specifically, the comparison of patients who did and did not choose TOCC follow-up introduces the possibility of selection bias. Future randomized/controlled studies will need to evaluate whether TOCC is more effective than the standard of care to reduce readmissions. Finally, since the analysis period following the final PDSA cycle was compressed due to the onset of the COVID-19 pandemic in the United States, no data are available regarding the sustained impacts of changes made during this cycle.

A multidisciplinary TOCC within the ICVA, featuring both virtual and in-person visits, reduced 30-day readmission rates by 19.3%; this approach to care was especially effective in patients with CHF. Virtual visits were the follow-up mode of choice for both urban and rural veterans, and there was no difference in outcomes between these two follow-up options. Future studies will focus on additional quality metrics, including cost-effectiveness and patient satisfaction.

Hospital readmissions are a significant problem in the United States, affecting 15% to 30% of discharges and incurring costs of more than $17 billion annually.¹ Timely posthospitalization follow-up visits are critical to ensure the effective transfer of patients to the outpatient setting; such visits reduce readmission rates as well as hospital length of stay and overall health care resource utilization.^2-4 Patients who receive inadequate follow-up care (ie, within 4 weeks of discharge) are significantly more likely to be readmitted than those who receive close follow-up care^.5

Due to the large clinical and financial consequences associated with hospital readmission, a variety of interventions have been studied, including home visits, telemonitoring, medication management, telephone calls, and postdischarge clinics.^6,7 While studies have not shown postdischarge clinics to be universally efficacious in reducing readmission rates, there is increasing evidence of reduced readmission rates in clinics that target high-risk patients (eg, patients with congestive heart failure [CHF]) rather than the total population.² A study by Hernandez et al that evaluated the relationship between early physician follow-up and 30-day readmissions showed a significantly lower readmission rate among hospitals with higher follow-up rates.⁸ Similarly, patients with CHF in a large, integrated health system who were seen within 7 days of discharge had an odds ratio (OR) of 0.81 (95% CI, 0.70-0.94) for 30-day readmissions.⁹

Transitions-of-care clinics (TOCC), designed to provide early postdischarge follow-up to high-risk patients, have been shown to reduce 30-day readmission rates,^3,4,10,11 especially in clinics that have same-physician follow-up visits rather than follow-up visits with a community primary care physician (PCP).¹² The most pronounced impact of postdischarge follow-up is seen in high-risk patients with high complexity or high severity of disease; however, complex rural patients are less likely to have access to specialty care.¹³ As a result, since rural residents must travel farther for specialty care, they are seen less frequently than their urban counterparts.^14,15

Prior to our TOCC initiative, the Iowa City VA (ICVA) ranked in the fifth quintile of the Veterans Health Administration (VHA) Strategic Analytics for Improvement and Learning model for hospital-wide readmissions (HWR), meaning that HWR at ICVA were higher than 80% of the other VHA healthcare centers. The low score in this metric was in part due to readmission rates in high-risk populations, including patients with CHF and those with high Care Assessment Need (CAN) scores. One concern was that the ICVA system serves many veterans from rural areas, some of whom must travel up to 200 miles to access inpatient and subspecialty care.

To meet these challenges, we implemented a TOCC to deliver timely postdischarge care focusing on high-risk and high-complexity patients. To address access-to-care issues of patients living in rural areas within the ICVA, we included virtual follow-up visits as a key component of our intervention.^16,17 The aim of this project was to decrease 30-day readmission rates of ICVA patients by 20% within 12 months of implementation.

Setting/Study Population

The ICVA serves 184,000 veterans stretched over 50 counties in eastern Iowa, western Illinois, and northern Missouri, with more than 60% of these patients residing in rural areas. Patients were initially eligible for the TOCC if they had an admission diagnosis of CHF and a CAN score > 85 at the time of discharge. The CAN score, developed by the VA to assess the risk of hospital readmission in individual patients, factors in several variables, including demographics, coexisting conditions, vital signs, utilization of services, pharmacy visits, and laboratory results. Patients in the top 5% (95-99) have a readmission rate of 20% at 90 days. Since the CAN is a proprietary tool, it may not be published in full; however, this assessment tool is commonly used and frequently cited in VA research.^18-22 The CAN score is expressed as a percentile ranging from 0 (lowest risk) to 99 (highest risk). Patient eligibility was expanded during subsequent Plan-Do-Study-Act (PDSA) cycles, as outlined below. Patient eligibility was expanded during subsequent PDSA cycles (also outlined below). A review by a local institutional review board was obtained, and the study was classified as exempt due to the use of deidentified data. Standards for Quality Improvement Reporting Excellence 2.0 guidelines were used to construct the manuscript.

Magnitude Assessment

The numbers of discharges, readmissions within 30 days, emergency department (ED) visits by all discharged veterans, and veterans discharged with a CHF hospital diagnosis were recorded from February 2017 to February 2018, which were the 12 months immediately preceding the pilot implementation.

Intervention

The primary intervention was referral to the newly formed ICVA TOCC. The multidisciplinary TOCC team consisted of hospitalists, pharmacists, schedulers, and discharge planners/care managers. Patients were identified by the hospitalist team during admission; prior to hospital discharge, these patients were referred to TOCC discharge planners to schedule appropriate follow-up appointments. Virtual follow-up visits were conducted using a patient’s home technology; in cases where a patient lacked adequate technology capabilities (eg, no computer or internet access), the ICVA provided a tablet device with cellular internet capability for temporary use. Specific clinical activities included medication reconciliation by a pharmacist, follow-up of pending laboratory studies, imaging studies, pathology results, medical diagnosis education, counseling regarding dietary restrictions, and contingency planning outside of an ED visit in the event of a change in clinical status. In addition, the TOCC aimed to facilitate a smooth transition of care back to the PCP by arranging follow-up appointments, providing visit summaries, and scheduling consults with specialty care, as appropriate.

Measures

The primary objective measure was the 30-day readmission rate in the ICVA hospital. Secondary measures included the number of VHA ED visits within 30 days of discharge. The main process measures were the number of hospital discharges per month, the number of TOCC referrals, the number of TOCC appointments made, the number of virtual and in-person visits, and the percentage of appointment “no-shows.”

Implementation

The TOCC was piloted from April 2018 to October 2018. During the pilot phase, TOCC enrollment was limited to virtual appointments and to patients with an admission diagnosis of CHF and a CAN score of > 85. The TOCC had staff on-site 2 days a week; this included pharmacists to reconcile medications and hospitalists to address follow-up care needs.

The TOCC clinic was temporarily closed at the end of October 2018 to analyze pilot results. Based on stakeholder feedback, changes made as part of the second PDSA cycle included expanding eligibility criteria to any hospital admission diagnosis and to patients with a CAN score < 85 if the hospitalist team felt the patient was likely to benefit from TOCC follow-up. In addition, on-site clinic staffing was expanded from 2 to 5 days per week to improve access, and the option for an in-person visit was added based on concerns some veterans expressed regarding the use of the technology at home. Finally, a formal resident program was added, and the order set for referrals was simplified. The TOCC was restarted in February 2019, and TOCC metrics were reviewed monthly. By July 2019, we identified issues with TOCC referrals and appointment creation that required additional modifications to the intervention.

A third PDSA cycle was initiated in July 2019 and included major changes, notably the formation of a designated TOCC committee. The committee appointed a dedicated TOCC scheduler whose role was to reduce confusion regarding scheduling, to update the discharge instructions/orders template to lower incidences of “double-booking” that occurred with PCP and TOCC appointments, to modify discharge educational instruction regarding virtual visits and tablet use, to adjust the TOCC-PCP handoff, and to formalize interactions between discharge coordinators and residents to review possible referrals every morning (Appendix Figure 1).

Statistical Analysis

Run charts were constructed by plotting monthly primary outcome values and monthly process metrics (Figure, Appendix Figure 2, Appendix Figure 3). Chi-square tests were used to compare 30-day readmission rates before and after the intervention. Statistical modeling was used to determine differences in outcomes between referred patients seen and referred patients not seen by the TOCC. In these statistical models, the outcome measures were 30-day readmissions, 30-day ED visits, and 6-month mortality. Covariates included in the final analysis were age, gender, race, CAN score, rural-urban commuting area code, referral service (resident vs nonresident), and admission diagnosis. Admission diagnoses were sorted by the investigators into one of the following seven categories: cardiac, infectious, pulmonary, gastrointestinal, neurologic, renal, and other.

Mean (SD) or counts and percentages were used to describe the distribution of continuous and categorical variables, respectively. Kruskal-Wallis test, t test, or chi-square tests were used, as appropriate, across categories. Generalized linear models with a logistic link function were used to test for differences between patients who kept their appointment at the TOCC and those who did not keep their TOCC appointment (both unadjusted and adjusted for all of the covariates previously mentioned). In addition, generalized linear models were also used to compare outcomes between TOCC patients seen virtually vs those seen in-person (both unadjusted and adjusted for all the covariates previously mentioned). All statistical tests were considered significant at a two-sided P < .05. All analyses were performed using SAS software version 9.4 (SAS Institute Inc).

Magnitude Assessment

During the preimplementation period (February 2017-February 2018), there were 3014 patient discharges from ICVA and 343 readmissions, resulting in a readmission rate of 11.4%. Among patients with a hospital-admission diagnosis of cardiorespiratory disease, which included patients with CHF, there were 381 discharges and 46 readmissions, resulting in a readmission rate of 12.1%.

Primary Outcome

During the pilot phase, which was conducted from April 2018 to October 2018, 142 patients who met inclusion criteria (CHF diagnosis and a CAN score > 85) were discharged from ICVA, and 56 referrals to the TOCC were placed. The readmission rate among the cardiorespiratory cohort of veterans was 9.5%.

During the expansion of the intervention from February 2019 to February 2020, there were 2844 discharges from the ICVA and 291 readmissions, resulting in a readmission rate of 10.2%. However, there was a further decrease in the readmission rate after the third PDSA cycle was initiated in July 2019 (Appendix Figure 1). The readmission rate was 9.2% in the final 6 months of the intervention period, and 7.9% in the final 3 months. Of note, in the group of 1948 patients who did not meet the eligibility criteria to participate in our study, the readmission rate during the same time period was 8.6% (161 readmissions).

When comparing the 6 months following the third PDSA cycle to the magnitude assessment period, there was a relative readmission reduction of 19.3% (P = .04), and an absolute reduction of 2.2%. If the final 3 months of the intervention period are included, there was an absolute reduction of 3.5% and a relative reduction of 30.7% (P = .01). Notably, before the pilot phase, ICVA was in the fifth quintile for HWR among VA hospitals but improved to the second quintile by the end of the expansion phase.

Process Outcomes

Process metrics for TOCC referrals, the number of patients seen, and the number of virtual and in-person visits over time are shown in Appendix Figure 3. Rates of TOCC referrals and the number of TOCC visits were lower than anticipated during the first 5 months of the intervention. However, TOCC referrals increased significantly after we implemented the previously described changes as part of the third PDSA cycle. As a result, total, virtual, and in-person visits also significantly increased from July 2019 to February 2020. The proportion of patients choosing virtual vs in-person visits fluctuated over time, but virtual visits were generally chosen more often than in-person visits.

Statistical Modeling

Baseline Data

Cohort characteristics are shown in Table 1. The cohort, which reflected the ICVA population, was predominantly male (96%) and White (93%), with a mean age of 67 years. The population was approximately half urban and half rural in composition, and the most common reason for hospital admission was cardiac. Other than a small but statistically significant difference in CAN scores, there were no significant differences between patients who kept their TOCC appointment and those who did not. There were also no differences in baseline characteristics between patients who chose virtual follow-up and patients who chose in-person follow-up, including the proportion of urban and rural patients.

Outcomes

Patients who kept their TOCC appointments had a 30-day readmission rate of 9.6%, which was significantly lower than the 30-day readmission rate of 27% in the group that did not keep their TOCC appointment (P < .001). Similarly, the percentage of patients treated in the ED was 15% in the TOCC group compared to 31.2% in the group that canceled their appointment (P < .001) (Table 1). In the multivariable analysis, patients who were seen in the TOCC group had an OR for 30-day readmission of 0.35 (95% CI, 0.19-0.62, P < .001), and an OR for ED visits of 0.39 (95% CI, 0.23-0.65; P < .001) (Table 2). There was no statistically significant difference in 6-month mortality between the two groups. In the virtual group compared to the in-person group, there were no statistically significant differences in outcomes between the two groups in the unadjusted or adjusted analysis (Table 2).

In this quality improvement initiative, we found that a TOCC targeting high-risk patients and offering virtual follow-up visits significantly decreased the 30-day readmission rates among veterans at ICVA. Statistical comparisons of patients seen at the TOCC vs those not seen at the TOCC showed a dramatic reduction in 30-day readmissions and ED visits. Finally, virtual follow-ups were more popular than in-person visits, and patients who followed up virtually had equivalent outcomes to those with the more traditional follow-up.

In the expansion phase, eligibility was expanded to include any hospital indication but continued to focus on high-risk patients. Existing literature suggests that providing postdischarge care to all patients, including low- or medium-risk patients, may not be as impactful as enrolling high-risk patients only. For instance, a postdischarge clinic offered to all patients at a VA system in Colorado did not reduce readmission rates compared to PCP follow-up.²³ In contrast, a study of more than 10,000 high-risk urban patients demonstrated that postdischarge care resulted in a 9.3% reduction in readmission risk.²⁴ Our data are consistent with the previously published studies, as the average CAN score of patients seen in TOCC was 90, suggesting a high risk of readmission. In the final 12 months of the intervention, 15% of discharged patients were seen at the TOCC clinic, suggesting that targeted intervention within the small subset of high-risk patients was sufficient to achieve our primary aim. Of note, among patients who did not meet the inclusion criteria for TOCC referral (ie, patients not considered high risk [CAN score ≤ 85]), the rate of readmissions was 8.6%.

Most of the available research on the efficacy of postdischarge clinics was conducted in urban environments. Our ICVA population sees a large proportion of rural veterans, who account for just over 50% of the discharge population. In a study of more than 2 million Medicare patients discharged from US hospitals, the 30-day readmission rates and adjusted mortality rates were higher among patients in rural counties, and post–acute care seemed to have a greater impact in rural rather than urban settings.²⁵ Previous studies have demonstrated that virtual visits have the potential to improve readmission rates, especially in patients with CHF²⁶ and in patients at the highest risk for readmission.²⁷ In our study, the pilot phase offered only virtual visits, but we subsequently added an in-person option based on veteran feedback. Interestingly, over the next 12 months, virtual visits were more popular with both urban and rural veterans, and there were no differences in the number of rural patients in the in-person vs the virtual group. These findings suggest factors other than rurality influenced the decision to choose virtual follow-up visits over in-person visits. Future studies should seek to determine the extent to which factors such as age, race, educational level, and socioeconomic circumstances impact veterans’ follow-up decisions. Not only were outcomes among patients who chose virtual visits the same as those of patients who chose in-person visits, but both of these groups had better outcomes compared to the non-TOCC group (Table 2). This finding demonstrating the efficacy of virtual visits among rural and urban patients has taken on increased significance due to the COVID-19 pandemic, as virtual visits offer a safer option, one that minimizes physical contact.

Our quality improvement analysis included a statistical comparison of patients seen vs those not seen at the TOCC. Patients who were referred to the TOCC but chose not to keep their appointment were similar to those seen in TOCC in terms of age, CAN score, rurality, and hospital diagnosis, but readmission rates were substantially higher in this group even after adjustments for covariates (Table 2). Evaluating causality in interventions aimed to reduce hospital readmission rates is complicated.²⁸ Our findings add greater plausibility to the utility of TOCC in accounting for at least a portion of the reported decrease in ICVA 30-day readmissions.

Our study has several strengths, including an observation period longer than 2 years, a large population of discharged veterans within an integrated healthcare system, and a large proportion of patients living in rural areas. Another strength of our study is the innovative nature of the intervention, which features a multidisciplinary team and the option of virtual or in-person visits. Nevertheless, this study also has several important limitations. As a single-center study, our findings may not be generalizable to other institutions, especially those outside the VHA system. Similarly, our study population reflected that of the ICVA, which may limit generalizability to a more diverse population. While we attempted to account in our statistical modeling for baseline differences between referred patients seen by the TOCC and those referred but not seen, we cannot exclude residual confounding between the groups. Specifically, the comparison of patients who did and did not choose TOCC follow-up introduces the possibility of selection bias. Future randomized/controlled studies will need to evaluate whether TOCC is more effective than the standard of care to reduce readmissions. Finally, since the analysis period following the final PDSA cycle was compressed due to the onset of the COVID-19 pandemic in the United States, no data are available regarding the sustained impacts of changes made during this cycle.

A multidisciplinary TOCC within the ICVA, featuring both virtual and in-person visits, reduced 30-day readmission rates by 19.3%; this approach to care was especially effective in patients with CHF. Virtual visits were the follow-up mode of choice for both urban and rural veterans, and there was no difference in outcomes between these two follow-up options. Future studies will focus on additional quality metrics, including cost-effectiveness and patient satisfaction.

While we agree that a well-honed physical exam is one of the most important diagnostic tools than an internist can use, we have several responses to Drs Kanjee and McNamara’s point that a routine physical exam is essential for hospitalized patients.¹ They argue that this exam might be helpful as a deliberate practice to improve skills for effective diagnostic exams. To this, we have two responses: the first is that the typical “routine” exam—a brief auscultation of the chest and abdomen—is performed frequently enough that additional practice should not be necessary for any practicing hospitalist. Performing a true full exam that would hone infrequently used skills, such as a full neurological exam, an orthopedic knee exam, or fundoscopy, comes at the expense of time spent talking to patients, as well as the potential harm of downstream testing cascades leading to adverse events. Second, we would argue that the real skill being developed is not “recognizing normal” but instead learning how to appropriately use physical diagnostic skills. Knowing precisely what exam maneuvers might be beneficial in a given hospitalized patient is incredibly complex, far more so than charts in evidence-based exam textbooks would suggest. It is this skill, not “recognizing normal,” that requires deliberate practice.

We agree that even during routine hospitalizations, daily exams may help detect complications of therapy, such as a patient with cellulitis on intravenous fluids developing volume overload. We are not against performing physical exams for diagnostic or monitoring purposes. In fact, it may be that most hospitalized patients would benefit from some sort of daily exam. However, rarely performed maneuvers, such as walking with patients or performing a validated delirium screen, are likely to have a higher yield than routine lung auscultation. It may also be true that hospitalized patients would benefit from certain screening exam maneuvers, but again, evidence is lacking, and decades of experience in the outpatient world would suggest the contrary.

Finally, and most ardently, we disagree that performing a routine daily physical exam can somehow inoculate against burnout. That is a view wholly unsupported by any evidence. The physical exam was originally developed as a diagnostic tool, not as a method to connect with patients. However, this traditional “routine” exam has been taught in medical schools as normal ever since, with very little serious interrogation of its utility or downstream effects. Increased cynicism about the exam’s usefulness, in our opinion, reflects physician cognizance of actual disutility of routine exams, rather than pining for a halcyon era that never existed. In fact, we believe a more hypothesis-driven diagnostic use of exams enriches physical diagnosis. For instance, listening to the chest of a patient with cellulitis on intravenous fluids is no longer “just listening,” it is an exercise specifically looking for a finding that affects management. Patient-centered care means tailoring all of our care—including the physical exam—to the needs of the patient. Doing a cursory, routine exam day after day for every patient with the goal of “recognizing normal” is not patient-centered, but rather physician-centered.

We do not doubt the importance of ritual, especially in such a stressful situation as a modern hospitalization. But rather than use a diagnostic procedure with downstream effects, we urge hospitalists to consider instead a ritual dating back to the time of Hippocrates—the compassionate physician sitting at the bedside, laying a hand on the shoulder, and listening to the patient’s concerns. That is authentic human connection rather than performance.

The authors of this point-counterpoint to thank Chris Smith, MD, and the members of the BIDMC Internal Medicine Residency Clinician Educator Track for thoughtful discussion around these topics.

While we agree that a well-honed physical exam is one of the most important diagnostic tools than an internist can use, we have several responses to Drs Kanjee and McNamara’s point that a routine physical exam is essential for hospitalized patients.¹ They argue that this exam might be helpful as a deliberate practice to improve skills for effective diagnostic exams. To this, we have two responses: the first is that the typical “routine” exam—a brief auscultation of the chest and abdomen—is performed frequently enough that additional practice should not be necessary for any practicing hospitalist. Performing a true full exam that would hone infrequently used skills, such as a full neurological exam, an orthopedic knee exam, or fundoscopy, comes at the expense of time spent talking to patients, as well as the potential harm of downstream testing cascades leading to adverse events. Second, we would argue that the real skill being developed is not “recognizing normal” but instead learning how to appropriately use physical diagnostic skills. Knowing precisely what exam maneuvers might be beneficial in a given hospitalized patient is incredibly complex, far more so than charts in evidence-based exam textbooks would suggest. It is this skill, not “recognizing normal,” that requires deliberate practice.

We agree that even during routine hospitalizations, daily exams may help detect complications of therapy, such as a patient with cellulitis on intravenous fluids developing volume overload. We are not against performing physical exams for diagnostic or monitoring purposes. In fact, it may be that most hospitalized patients would benefit from some sort of daily exam. However, rarely performed maneuvers, such as walking with patients or performing a validated delirium screen, are likely to have a higher yield than routine lung auscultation. It may also be true that hospitalized patients would benefit from certain screening exam maneuvers, but again, evidence is lacking, and decades of experience in the outpatient world would suggest the contrary.

Finally, and most ardently, we disagree that performing a routine daily physical exam can somehow inoculate against burnout. That is a view wholly unsupported by any evidence. The physical exam was originally developed as a diagnostic tool, not as a method to connect with patients. However, this traditional “routine” exam has been taught in medical schools as normal ever since, with very little serious interrogation of its utility or downstream effects. Increased cynicism about the exam’s usefulness, in our opinion, reflects physician cognizance of actual disutility of routine exams, rather than pining for a halcyon era that never existed. In fact, we believe a more hypothesis-driven diagnostic use of exams enriches physical diagnosis. For instance, listening to the chest of a patient with cellulitis on intravenous fluids is no longer “just listening,” it is an exercise specifically looking for a finding that affects management. Patient-centered care means tailoring all of our care—including the physical exam—to the needs of the patient. Doing a cursory, routine exam day after day for every patient with the goal of “recognizing normal” is not patient-centered, but rather physician-centered.

We do not doubt the importance of ritual, especially in such a stressful situation as a modern hospitalization. But rather than use a diagnostic procedure with downstream effects, we urge hospitalists to consider instead a ritual dating back to the time of Hippocrates—the compassionate physician sitting at the bedside, laying a hand on the shoulder, and listening to the patient’s concerns. That is authentic human connection rather than performance.

The authors of this point-counterpoint to thank Chris Smith, MD, and the members of the BIDMC Internal Medicine Residency Clinician Educator Track for thoughtful discussion around these topics.

While we agree that a well-honed physical exam is one of the most important diagnostic tools than an internist can use, we have several responses to Drs Kanjee and McNamara’s point that a routine physical exam is essential for hospitalized patients.¹ They argue that this exam might be helpful as a deliberate practice to improve skills for effective diagnostic exams. To this, we have two responses: the first is that the typical “routine” exam—a brief auscultation of the chest and abdomen—is performed frequently enough that additional practice should not be necessary for any practicing hospitalist. Performing a true full exam that would hone infrequently used skills, such as a full neurological exam, an orthopedic knee exam, or fundoscopy, comes at the expense of time spent talking to patients, as well as the potential harm of downstream testing cascades leading to adverse events. Second, we would argue that the real skill being developed is not “recognizing normal” but instead learning how to appropriately use physical diagnostic skills. Knowing precisely what exam maneuvers might be beneficial in a given hospitalized patient is incredibly complex, far more so than charts in evidence-based exam textbooks would suggest. It is this skill, not “recognizing normal,” that requires deliberate practice.

We agree that even during routine hospitalizations, daily exams may help detect complications of therapy, such as a patient with cellulitis on intravenous fluids developing volume overload. We are not against performing physical exams for diagnostic or monitoring purposes. In fact, it may be that most hospitalized patients would benefit from some sort of daily exam. However, rarely performed maneuvers, such as walking with patients or performing a validated delirium screen, are likely to have a higher yield than routine lung auscultation. It may also be true that hospitalized patients would benefit from certain screening exam maneuvers, but again, evidence is lacking, and decades of experience in the outpatient world would suggest the contrary.

Finally, and most ardently, we disagree that performing a routine daily physical exam can somehow inoculate against burnout. That is a view wholly unsupported by any evidence. The physical exam was originally developed as a diagnostic tool, not as a method to connect with patients. However, this traditional “routine” exam has been taught in medical schools as normal ever since, with very little serious interrogation of its utility or downstream effects. Increased cynicism about the exam’s usefulness, in our opinion, reflects physician cognizance of actual disutility of routine exams, rather than pining for a halcyon era that never existed. In fact, we believe a more hypothesis-driven diagnostic use of exams enriches physical diagnosis. For instance, listening to the chest of a patient with cellulitis on intravenous fluids is no longer “just listening,” it is an exercise specifically looking for a finding that affects management. Patient-centered care means tailoring all of our care—including the physical exam—to the needs of the patient. Doing a cursory, routine exam day after day for every patient with the goal of “recognizing normal” is not patient-centered, but rather physician-centered.

We do not doubt the importance of ritual, especially in such a stressful situation as a modern hospitalization. But rather than use a diagnostic procedure with downstream effects, we urge hospitalists to consider instead a ritual dating back to the time of Hippocrates—the compassionate physician sitting at the bedside, laying a hand on the shoulder, and listening to the patient’s concerns. That is authentic human connection rather than performance.

The authors of this point-counterpoint to thank Chris Smith, MD, and the members of the BIDMC Internal Medicine Residency Clinician Educator Track for thoughtful discussion around these topics.

We read with interest the perspective of Drs Rodman and Warnock,¹ but disagree with the authors on several points. We find that the routine daily physical examination, often neglected or unappreciated amidst technological advances and new diagnostic testing, provides significant value to both hospitalist and patient in terms of diagnosis, treatment, patient-centered care, and maintaining the patient-doctor relationship.

First, the daily physical exam provides the practice that hospitalists need to develop and maintain necessary diagnostic finesse. We are taught fundamental physical exam maneuvers in medical school, but these skills often atrophy during residency and throughout our careers.² This leaves us, as practicing physicians, potentially worse at these competencies than when we were students. The daily physical exam, on the other hand, provides frequent and effective practice to keep up and build upon these skills. We gain in part from our repeated normal exams, which help us to skillfully recognize the rare abnormal finding; a hospitalist must likely feel hundreds of normal abdomens to reliably discover a furtive abdominal mass. Our exams also benefit from several forms of prompt and relevant feedback, including that which is provided by subspecialist consultants (like a cardiologist agreeing with your assessment of jugular venous pressure) and other diagnostic tests (like the echocardiogram without the valvulopathy thought to be detected at the bedside). The physical examination is best learned at the bedside, and the daily exam offers an unparalleled opportunity to do so. Such continual skill improvement is necessary for hospitalists to accurately apply data from the evidence-based physical diagnosis literature. Many studies of the utility of various physical exam findings³ involve maneuvers performed by experts; to truly apply their results to our generalist practice, we are required to push ourselves to obtain the diagnostic expertise of the specialist. The daily physical examination, being the most concrete way for hospitalists to do so, is therefore essential to practicing better evidence-based physical diagnosis.

Beyond these larger benefits of the daily physical examination on our own practice and skills, patients in our care benefit diagnostically from these exams as well. Time and again, we see an inadequate or incomplete physical exam leading to errors or adverse patient outcomes.^4,5 Even after completion of initial laboratory and imaging tests, laying of hands and stethoscope can lead to dramatic changes in inpatient diagnosis and management.⁶ The subsequent routine daily physical provides fresh opportunities to reexamine the evidence for or against our own working diagnoses and management plans. The adage that “you don’t know what you don’t know” is especially fitting here. We often do not know to look for and rule a disease process in or out if it is not on our initial differential on hospital day one or two; the daily physical exam allows us to be on the lookout for diagnoses we have not yet considered. The two of us have more than once made a serendipitous discovery of a new rash or other physical finding on hospital day three or four that helps suggest another, and ultimately correct, final diagnosis. Particularly in a setting in which so many inpatient diagnoses are wrong and can lead to patient harm,⁷ the daily physical examination provides an important check on our own diagnostic reasoning.

Even if we are right about the diagnosis, the daily exam also allows for timely recognition of complications from our management. Listening to each patient’s lungs every day, including those of patients with seemingly unrelated lower-extremity cellulitis, means we will more promptly notice when they retain fluid due to as yet unknown underlying heart failure. Those subtle bibasilar crackles not only become diagnostically useful, but also allow us the possibility of intervening and changing course even before the patient reports shortness of breath or the nurse notes hypoxemia on routine vital signs a day or two later. In an era when our treatment regimens are more complex, with frequent off-target results and side effects, the daily exam is a key screening tool for adverse outcomes in an increasingly ill population. Having a frequently updated and accurate baseline exam is also exceptionally important in the event of sudden neurologic deficits; an inpatient with new facial droop and left-arm weakness at 10 am has much better treatment options if their hospitalist has conducted a routine basic neurologic exam that morning and can confidently provide a time when they were “last known well.”

Finally, the daily physical examination is important to patient-centered care and potentially preventing physician burnout.⁸ Patients have more confidence in us when we conduct a thorough exam. The ritual of the physical exam is also an important contributor to the patient-doctor relationship, and a daily exam can help strengthen that bond each morning.⁹ Such benefits also extend to physicians. Hospitalists are spending less and less time at the bedside,¹⁰ a reality at least partially responsible for rising rates of burnout.¹¹ We all went into clinical medicine to take care of and connect with people. The daily physical examination offers valuable time to show our patients we care about them while also giving us the opportunity to spend time with them, rather than with the “iPatient” that can otherwise become our focus.¹² In this way, the daily physical examination can be immensely satisfying and may not only inoculate against burnout but also contribute to a stronger patient-doctor relationship.

For so many reasons, the daily physical exam is of great benefit to hospitalists looking to develop and maintain diagnostic skills, to our patients as we stay on the lookout for unexpected diagnoses and complications, and to the relationships we have with those for whom we care. It is a practice worth not only continuing but celebrating.

We read with interest the perspective of Drs Rodman and Warnock,¹ but disagree with the authors on several points. We find that the routine daily physical examination, often neglected or unappreciated amidst technological advances and new diagnostic testing, provides significant value to both hospitalist and patient in terms of diagnosis, treatment, patient-centered care, and maintaining the patient-doctor relationship.

First, the daily physical exam provides the practice that hospitalists need to develop and maintain necessary diagnostic finesse. We are taught fundamental physical exam maneuvers in medical school, but these skills often atrophy during residency and throughout our careers.² This leaves us, as practicing physicians, potentially worse at these competencies than when we were students. The daily physical exam, on the other hand, provides frequent and effective practice to keep up and build upon these skills. We gain in part from our repeated normal exams, which help us to skillfully recognize the rare abnormal finding; a hospitalist must likely feel hundreds of normal abdomens to reliably discover a furtive abdominal mass. Our exams also benefit from several forms of prompt and relevant feedback, including that which is provided by subspecialist consultants (like a cardiologist agreeing with your assessment of jugular venous pressure) and other diagnostic tests (like the echocardiogram without the valvulopathy thought to be detected at the bedside). The physical examination is best learned at the bedside, and the daily exam offers an unparalleled opportunity to do so. Such continual skill improvement is necessary for hospitalists to accurately apply data from the evidence-based physical diagnosis literature. Many studies of the utility of various physical exam findings³ involve maneuvers performed by experts; to truly apply their results to our generalist practice, we are required to push ourselves to obtain the diagnostic expertise of the specialist. The daily physical examination, being the most concrete way for hospitalists to do so, is therefore essential to practicing better evidence-based physical diagnosis.

Beyond these larger benefits of the daily physical examination on our own practice and skills, patients in our care benefit diagnostically from these exams as well. Time and again, we see an inadequate or incomplete physical exam leading to errors or adverse patient outcomes.^4,5 Even after completion of initial laboratory and imaging tests, laying of hands and stethoscope can lead to dramatic changes in inpatient diagnosis and management.⁶ The subsequent routine daily physical provides fresh opportunities to reexamine the evidence for or against our own working diagnoses and management plans. The adage that “you don’t know what you don’t know” is especially fitting here. We often do not know to look for and rule a disease process in or out if it is not on our initial differential on hospital day one or two; the daily physical exam allows us to be on the lookout for diagnoses we have not yet considered. The two of us have more than once made a serendipitous discovery of a new rash or other physical finding on hospital day three or four that helps suggest another, and ultimately correct, final diagnosis. Particularly in a setting in which so many inpatient diagnoses are wrong and can lead to patient harm,⁷ the daily physical examination provides an important check on our own diagnostic reasoning.

Even if we are right about the diagnosis, the daily exam also allows for timely recognition of complications from our management. Listening to each patient’s lungs every day, including those of patients with seemingly unrelated lower-extremity cellulitis, means we will more promptly notice when they retain fluid due to as yet unknown underlying heart failure. Those subtle bibasilar crackles not only become diagnostically useful, but also allow us the possibility of intervening and changing course even before the patient reports shortness of breath or the nurse notes hypoxemia on routine vital signs a day or two later. In an era when our treatment regimens are more complex, with frequent off-target results and side effects, the daily exam is a key screening tool for adverse outcomes in an increasingly ill population. Having a frequently updated and accurate baseline exam is also exceptionally important in the event of sudden neurologic deficits; an inpatient with new facial droop and left-arm weakness at 10 am has much better treatment options if their hospitalist has conducted a routine basic neurologic exam that morning and can confidently provide a time when they were “last known well.”

Finally, the daily physical examination is important to patient-centered care and potentially preventing physician burnout.⁸ Patients have more confidence in us when we conduct a thorough exam. The ritual of the physical exam is also an important contributor to the patient-doctor relationship, and a daily exam can help strengthen that bond each morning.⁹ Such benefits also extend to physicians. Hospitalists are spending less and less time at the bedside,¹⁰ a reality at least partially responsible for rising rates of burnout.¹¹ We all went into clinical medicine to take care of and connect with people. The daily physical examination offers valuable time to show our patients we care about them while also giving us the opportunity to spend time with them, rather than with the “iPatient” that can otherwise become our focus.¹² In this way, the daily physical examination can be immensely satisfying and may not only inoculate against burnout but also contribute to a stronger patient-doctor relationship.

For so many reasons, the daily physical exam is of great benefit to hospitalists looking to develop and maintain diagnostic skills, to our patients as we stay on the lookout for unexpected diagnoses and complications, and to the relationships we have with those for whom we care. It is a practice worth not only continuing but celebrating.

We read with interest the perspective of Drs Rodman and Warnock,¹ but disagree with the authors on several points. We find that the routine daily physical examination, often neglected or unappreciated amidst technological advances and new diagnostic testing, provides significant value to both hospitalist and patient in terms of diagnosis, treatment, patient-centered care, and maintaining the patient-doctor relationship.

First, the daily physical exam provides the practice that hospitalists need to develop and maintain necessary diagnostic finesse. We are taught fundamental physical exam maneuvers in medical school, but these skills often atrophy during residency and throughout our careers.² This leaves us, as practicing physicians, potentially worse at these competencies than when we were students. The daily physical exam, on the other hand, provides frequent and effective practice to keep up and build upon these skills. We gain in part from our repeated normal exams, which help us to skillfully recognize the rare abnormal finding; a hospitalist must likely feel hundreds of normal abdomens to reliably discover a furtive abdominal mass. Our exams also benefit from several forms of prompt and relevant feedback, including that which is provided by subspecialist consultants (like a cardiologist agreeing with your assessment of jugular venous pressure) and other diagnostic tests (like the echocardiogram without the valvulopathy thought to be detected at the bedside). The physical examination is best learned at the bedside, and the daily exam offers an unparalleled opportunity to do so. Such continual skill improvement is necessary for hospitalists to accurately apply data from the evidence-based physical diagnosis literature. Many studies of the utility of various physical exam findings³ involve maneuvers performed by experts; to truly apply their results to our generalist practice, we are required to push ourselves to obtain the diagnostic expertise of the specialist. The daily physical examination, being the most concrete way for hospitalists to do so, is therefore essential to practicing better evidence-based physical diagnosis.

Beyond these larger benefits of the daily physical examination on our own practice and skills, patients in our care benefit diagnostically from these exams as well. Time and again, we see an inadequate or incomplete physical exam leading to errors or adverse patient outcomes.^4,5 Even after completion of initial laboratory and imaging tests, laying of hands and stethoscope can lead to dramatic changes in inpatient diagnosis and management.⁶ The subsequent routine daily physical provides fresh opportunities to reexamine the evidence for or against our own working diagnoses and management plans. The adage that “you don’t know what you don’t know” is especially fitting here. We often do not know to look for and rule a disease process in or out if it is not on our initial differential on hospital day one or two; the daily physical exam allows us to be on the lookout for diagnoses we have not yet considered. The two of us have more than once made a serendipitous discovery of a new rash or other physical finding on hospital day three or four that helps suggest another, and ultimately correct, final diagnosis. Particularly in a setting in which so many inpatient diagnoses are wrong and can lead to patient harm,⁷ the daily physical examination provides an important check on our own diagnostic reasoning.

Even if we are right about the diagnosis, the daily exam also allows for timely recognition of complications from our management. Listening to each patient’s lungs every day, including those of patients with seemingly unrelated lower-extremity cellulitis, means we will more promptly notice when they retain fluid due to as yet unknown underlying heart failure. Those subtle bibasilar crackles not only become diagnostically useful, but also allow us the possibility of intervening and changing course even before the patient reports shortness of breath or the nurse notes hypoxemia on routine vital signs a day or two later. In an era when our treatment regimens are more complex, with frequent off-target results and side effects, the daily exam is a key screening tool for adverse outcomes in an increasingly ill population. Having a frequently updated and accurate baseline exam is also exceptionally important in the event of sudden neurologic deficits; an inpatient with new facial droop and left-arm weakness at 10 am has much better treatment options if their hospitalist has conducted a routine basic neurologic exam that morning and can confidently provide a time when they were “last known well.”

Finally, the daily physical examination is important to patient-centered care and potentially preventing physician burnout.⁸ Patients have more confidence in us when we conduct a thorough exam. The ritual of the physical exam is also an important contributor to the patient-doctor relationship, and a daily exam can help strengthen that bond each morning.⁹ Such benefits also extend to physicians. Hospitalists are spending less and less time at the bedside,¹⁰ a reality at least partially responsible for rising rates of burnout.¹¹ We all went into clinical medicine to take care of and connect with people. The daily physical examination offers valuable time to show our patients we care about them while also giving us the opportunity to spend time with them, rather than with the “iPatient” that can otherwise become our focus.¹² In this way, the daily physical examination can be immensely satisfying and may not only inoculate against burnout but also contribute to a stronger patient-doctor relationship.

For so many reasons, the daily physical exam is of great benefit to hospitalists looking to develop and maintain diagnostic skills, to our patients as we stay on the lookout for unexpected diagnoses and complications, and to the relationships we have with those for whom we care. It is a practice worth not only continuing but celebrating.

Every day, physicians engage in an elaborate performance with their patients—the routine complete physical exam. We argue that this purportedly time-tested ritual is at best a waste of time, and at worst potentially harmful.

The modern physical exam evolved throughout the 19th century as the first diagnostic tool in a medical field that was rapidly transforming from its traditional roots to a modern scientific discipline.¹ Despite the vast increase in diagnostic tools since then, the physical exam remains one of the most predictive. Several decades of investigation into the “evidence-based” physical exam have attempted to calculate the test characteristics of individual exam findings, confirming that the exam remains as useful a diagnostic tool today as it was for Laënnec or Osler.²

Performing a physical exam for the purposes of diagnosis and prognosis—not only on admission, but also on a daily basis to assess treatment response—remains a fundamental part of a hospitalist’s job. For example, a daily volume assessment, including cardiac auscultation for an S₃, evaluation of the jugular venous pulse, and measurement of edema, is essential in managing patients with decompensated heart failure. However, when we stray from these diagnostic purposes, we are no longer using the exam as intended.

The physical exam most frequently performed in the hospital today is the so-called routine daily exam. Generally, this involves passing a stethoscope fleetingly across the chest and abdomen, perhaps with some additional palpation of the abdomen. Cranial nerves II through XII may also occasionally be checked. This routine exam—and by extension, the templated physical exams that fill hospitalists’ documentation—not only lack an evidence base, but also are arguably harmful to patients. Such exams should not be part of a hospitalist’s daily practice.

The most concerning aspect of a routine daily exam is that examination of an asymptomatic patient—for example, auscultation of the lungs of a patient admitted with lower extremity cellulitis—is fundamentally a screening rather than a diagnostic test. While little work has been done in the inpatient setting, decades of studies on outpatient screening exams demonstrate that very few of them are effective.³ For example, a review of commonly used exam maneuvers in wellness visits concluded that “for the asymptomatic, nonpregnant adult of any age, no evidence supports the need for a complete physical exam as traditionally defined,” recommending against such popular maneuvers as lung and heart auscultation and peripheral pulse palpation.⁴ While the inpatient hospital medicine population has different characteristics that may warrant a routine exam, there is no evidence to support such practice.

It is often argued that the routine physical exam is “cheap” and “quick” and, therefore, should be performed regardless of evidence. While this is certainly true for many diagnostic physical exams, the literature suggests that there is no reason to think that a routine physical exam would be cost-effective.⁵ Even cost-effective screening physical exam tests, such as an outpatient nurse performing a 1-minute pulse palpation starting at age 55, have incremental costs measuring in the thousands of dollars.⁶ Furthermore, screening tests can have unexpected downstream effects that are both costly and associated with morbidity and mortality.⁷ For example, abdominal palpation of a “prominent” aorta can lead to imaging, where incidental findings can trigger procedures that may involve complications.

In addition to potentially adding more risk, the routine daily physical exam represents time that can be better allocated. Medical residents spend the vast majority of their day at the computer, while spending less than 10% of their time at the patient’s bedside.⁸ Anything that takes up that valuable time, including a “routine exam,” is time spent not talking to the patient, learning about their symptoms, their fears, and who they are as human beings.

It is also true that patients expect a physical exam to be performed, and that additional exam maneuvers, including potentially invasive exams, are associated with increasing patient satisfaction.⁹ However, these arguments miss much of the nuance of why patients have these expectations. Qualitative research suggests that much of a patient’s desire for unnecessary tests or exams is actually their concern about a lack of validation or empathy from the physician, as well as general skepticism about evidence-based medical decision-making.¹⁰ Perhaps spending more face time with patients discussing their issues, rather than idle time performing routine maneuvers, would lead to even greater patient satisfaction.

Finally, one of the most popular arguments in defense of a routine physical exam is that the exam is a “sacred ritual” essential to the patient-physician relationship.¹¹ However, this is an argument not supported by historical interpretation. The physical exam was developed as an explicitly diagnostic procedure in the early 19th century, while the primacy of the doctor-physician relationship dates back millennia, long before the development of the modern physical exam. Furthermore, modern historiography has identified the development of the physical exam as part of a movement to minimize the experience of the patient in their own disease, and to situate the physician as the ultimate source of knowledge about a patient’s body rather than an attempt to strengthen a relationship.¹²

Ritual is indeed important, and the exam as currently practiced may indeed reinforce the physician-patient relationship. But we should also keep in mind what that relationship entails. Having full access to a patient’s unclothed body and having the ability to perform invasive procedures are far beyond regular social norms—these are powerful diagnostic tools, yes, but they also serve to reinforce an imbalance of power in the relationship. Medical rituals have also changed dramatically over time. Modern evidence suggests that pulse palpation alone, the form of the exam that was dominant for millennia, has profound physiological effects even on critically ill patients.¹³ Rather than a diagnostic exam that has potential downstream cost implications and consumes valuable time from an encounter, we suggest a return to a more traditional ritual of physical touch: sitting at the patient’s bedside, holding their hand, and speaking to them compassionately about their fears and hopes. This would be a far more valuable “routine” encounter to incorporate into the busy hospitalist’s day.

The authors of this point-counterpoint thank Chris Smith, MD, and the members of the BIDMC Internal Medicine Residency Clinician Educator Track for thoughtful discussion around these topics.

Every day, physicians engage in an elaborate performance with their patients—the routine complete physical exam. We argue that this purportedly time-tested ritual is at best a waste of time, and at worst potentially harmful.

The modern physical exam evolved throughout the 19th century as the first diagnostic tool in a medical field that was rapidly transforming from its traditional roots to a modern scientific discipline.¹ Despite the vast increase in diagnostic tools since then, the physical exam remains one of the most predictive. Several decades of investigation into the “evidence-based” physical exam have attempted to calculate the test characteristics of individual exam findings, confirming that the exam remains as useful a diagnostic tool today as it was for Laënnec or Osler.²

Performing a physical exam for the purposes of diagnosis and prognosis—not only on admission, but also on a daily basis to assess treatment response—remains a fundamental part of a hospitalist’s job. For example, a daily volume assessment, including cardiac auscultation for an S₃, evaluation of the jugular venous pulse, and measurement of edema, is essential in managing patients with decompensated heart failure. However, when we stray from these diagnostic purposes, we are no longer using the exam as intended.

The physical exam most frequently performed in the hospital today is the so-called routine daily exam. Generally, this involves passing a stethoscope fleetingly across the chest and abdomen, perhaps with some additional palpation of the abdomen. Cranial nerves II through XII may also occasionally be checked. This routine exam—and by extension, the templated physical exams that fill hospitalists’ documentation—not only lack an evidence base, but also are arguably harmful to patients. Such exams should not be part of a hospitalist’s daily practice.

The most concerning aspect of a routine daily exam is that examination of an asymptomatic patient—for example, auscultation of the lungs of a patient admitted with lower extremity cellulitis—is fundamentally a screening rather than a diagnostic test. While little work has been done in the inpatient setting, decades of studies on outpatient screening exams demonstrate that very few of them are effective.³ For example, a review of commonly used exam maneuvers in wellness visits concluded that “for the asymptomatic, nonpregnant adult of any age, no evidence supports the need for a complete physical exam as traditionally defined,” recommending against such popular maneuvers as lung and heart auscultation and peripheral pulse palpation.⁴ While the inpatient hospital medicine population has different characteristics that may warrant a routine exam, there is no evidence to support such practice.

It is often argued that the routine physical exam is “cheap” and “quick” and, therefore, should be performed regardless of evidence. While this is certainly true for many diagnostic physical exams, the literature suggests that there is no reason to think that a routine physical exam would be cost-effective.⁵ Even cost-effective screening physical exam tests, such as an outpatient nurse performing a 1-minute pulse palpation starting at age 55, have incremental costs measuring in the thousands of dollars.⁶ Furthermore, screening tests can have unexpected downstream effects that are both costly and associated with morbidity and mortality.⁷ For example, abdominal palpation of a “prominent” aorta can lead to imaging, where incidental findings can trigger procedures that may involve complications.

In addition to potentially adding more risk, the routine daily physical exam represents time that can be better allocated. Medical residents spend the vast majority of their day at the computer, while spending less than 10% of their time at the patient’s bedside.⁸ Anything that takes up that valuable time, including a “routine exam,” is time spent not talking to the patient, learning about their symptoms, their fears, and who they are as human beings.

It is also true that patients expect a physical exam to be performed, and that additional exam maneuvers, including potentially invasive exams, are associated with increasing patient satisfaction.⁹ However, these arguments miss much of the nuance of why patients have these expectations. Qualitative research suggests that much of a patient’s desire for unnecessary tests or exams is actually their concern about a lack of validation or empathy from the physician, as well as general skepticism about evidence-based medical decision-making.¹⁰ Perhaps spending more face time with patients discussing their issues, rather than idle time performing routine maneuvers, would lead to even greater patient satisfaction.

Finally, one of the most popular arguments in defense of a routine physical exam is that the exam is a “sacred ritual” essential to the patient-physician relationship.¹¹ However, this is an argument not supported by historical interpretation. The physical exam was developed as an explicitly diagnostic procedure in the early 19th century, while the primacy of the doctor-physician relationship dates back millennia, long before the development of the modern physical exam. Furthermore, modern historiography has identified the development of the physical exam as part of a movement to minimize the experience of the patient in their own disease, and to situate the physician as the ultimate source of knowledge about a patient’s body rather than an attempt to strengthen a relationship.¹²

Ritual is indeed important, and the exam as currently practiced may indeed reinforce the physician-patient relationship. But we should also keep in mind what that relationship entails. Having full access to a patient’s unclothed body and having the ability to perform invasive procedures are far beyond regular social norms—these are powerful diagnostic tools, yes, but they also serve to reinforce an imbalance of power in the relationship. Medical rituals have also changed dramatically over time. Modern evidence suggests that pulse palpation alone, the form of the exam that was dominant for millennia, has profound physiological effects even on critically ill patients.¹³ Rather than a diagnostic exam that has potential downstream cost implications and consumes valuable time from an encounter, we suggest a return to a more traditional ritual of physical touch: sitting at the patient’s bedside, holding their hand, and speaking to them compassionately about their fears and hopes. This would be a far more valuable “routine” encounter to incorporate into the busy hospitalist’s day.

The authors of this point-counterpoint thank Chris Smith, MD, and the members of the BIDMC Internal Medicine Residency Clinician Educator Track for thoughtful discussion around these topics.

Every day, physicians engage in an elaborate performance with their patients—the routine complete physical exam. We argue that this purportedly time-tested ritual is at best a waste of time, and at worst potentially harmful.

The modern physical exam evolved throughout the 19th century as the first diagnostic tool in a medical field that was rapidly transforming from its traditional roots to a modern scientific discipline.¹ Despite the vast increase in diagnostic tools since then, the physical exam remains one of the most predictive. Several decades of investigation into the “evidence-based” physical exam have attempted to calculate the test characteristics of individual exam findings, confirming that the exam remains as useful a diagnostic tool today as it was for Laënnec or Osler.²

Performing a physical exam for the purposes of diagnosis and prognosis—not only on admission, but also on a daily basis to assess treatment response—remains a fundamental part of a hospitalist’s job. For example, a daily volume assessment, including cardiac auscultation for an S₃, evaluation of the jugular venous pulse, and measurement of edema, is essential in managing patients with decompensated heart failure. However, when we stray from these diagnostic purposes, we are no longer using the exam as intended.

The physical exam most frequently performed in the hospital today is the so-called routine daily exam. Generally, this involves passing a stethoscope fleetingly across the chest and abdomen, perhaps with some additional palpation of the abdomen. Cranial nerves II through XII may also occasionally be checked. This routine exam—and by extension, the templated physical exams that fill hospitalists’ documentation—not only lack an evidence base, but also are arguably harmful to patients. Such exams should not be part of a hospitalist’s daily practice.

The most concerning aspect of a routine daily exam is that examination of an asymptomatic patient—for example, auscultation of the lungs of a patient admitted with lower extremity cellulitis—is fundamentally a screening rather than a diagnostic test. While little work has been done in the inpatient setting, decades of studies on outpatient screening exams demonstrate that very few of them are effective.³ For example, a review of commonly used exam maneuvers in wellness visits concluded that “for the asymptomatic, nonpregnant adult of any age, no evidence supports the need for a complete physical exam as traditionally defined,” recommending against such popular maneuvers as lung and heart auscultation and peripheral pulse palpation.⁴ While the inpatient hospital medicine population has different characteristics that may warrant a routine exam, there is no evidence to support such practice.

It is often argued that the routine physical exam is “cheap” and “quick” and, therefore, should be performed regardless of evidence. While this is certainly true for many diagnostic physical exams, the literature suggests that there is no reason to think that a routine physical exam would be cost-effective.⁵ Even cost-effective screening physical exam tests, such as an outpatient nurse performing a 1-minute pulse palpation starting at age 55, have incremental costs measuring in the thousands of dollars.⁶ Furthermore, screening tests can have unexpected downstream effects that are both costly and associated with morbidity and mortality.⁷ For example, abdominal palpation of a “prominent” aorta can lead to imaging, where incidental findings can trigger procedures that may involve complications.

In addition to potentially adding more risk, the routine daily physical exam represents time that can be better allocated. Medical residents spend the vast majority of their day at the computer, while spending less than 10% of their time at the patient’s bedside.⁸ Anything that takes up that valuable time, including a “routine exam,” is time spent not talking to the patient, learning about their symptoms, their fears, and who they are as human beings.

It is also true that patients expect a physical exam to be performed, and that additional exam maneuvers, including potentially invasive exams, are associated with increasing patient satisfaction.⁹ However, these arguments miss much of the nuance of why patients have these expectations. Qualitative research suggests that much of a patient’s desire for unnecessary tests or exams is actually their concern about a lack of validation or empathy from the physician, as well as general skepticism about evidence-based medical decision-making.¹⁰ Perhaps spending more face time with patients discussing their issues, rather than idle time performing routine maneuvers, would lead to even greater patient satisfaction.

Finally, one of the most popular arguments in defense of a routine physical exam is that the exam is a “sacred ritual” essential to the patient-physician relationship.¹¹ However, this is an argument not supported by historical interpretation. The physical exam was developed as an explicitly diagnostic procedure in the early 19th century, while the primacy of the doctor-physician relationship dates back millennia, long before the development of the modern physical exam. Furthermore, modern historiography has identified the development of the physical exam as part of a movement to minimize the experience of the patient in their own disease, and to situate the physician as the ultimate source of knowledge about a patient’s body rather than an attempt to strengthen a relationship.¹²

Ritual is indeed important, and the exam as currently practiced may indeed reinforce the physician-patient relationship. But we should also keep in mind what that relationship entails. Having full access to a patient’s unclothed body and having the ability to perform invasive procedures are far beyond regular social norms—these are powerful diagnostic tools, yes, but they also serve to reinforce an imbalance of power in the relationship. Medical rituals have also changed dramatically over time. Modern evidence suggests that pulse palpation alone, the form of the exam that was dominant for millennia, has profound physiological effects even on critically ill patients.¹³ Rather than a diagnostic exam that has potential downstream cost implications and consumes valuable time from an encounter, we suggest a return to a more traditional ritual of physical touch: sitting at the patient’s bedside, holding their hand, and speaking to them compassionately about their fears and hopes. This would be a far more valuable “routine” encounter to incorporate into the busy hospitalist’s day.

The authors of this point-counterpoint thank Chris Smith, MD, and the members of the BIDMC Internal Medicine Residency Clinician Educator Track for thoughtful discussion around these topics.

As the COVID-19 pandemic surged in March 2020 in the United States, it was clear that severe COVID-19 and rates of hospitalization were much higher in adults than in children.¹ Pediatric facilities grappled with how to leverage empty beds and other underutilized human, clinical, and material resources to offset the overflowing adult facilities.^2,3 Pediatricians agonized about how to identify adult patients for whom they could provide safe and effective care, not only as individual clinicians, but also with adequate support from their local pediatric facility and health system.

Maria* (*name changed) was a young adult whose experience with her local health system highlighted common and addressable issues that arose when pediatric facilities aimed to care for adult populations. Adult hospitals were already above capacity caring for acutely ill patients with COVID-19, and a local freestanding children’s hospital offered to offload young adult patients up to age 30 years. Maria, a 26-year-old, had just been transferred from an adult emergency department (ED) to the children’s hospital ED for management of postoperative pain after a recent appendectomy. There was concern for possible abscess formation, but no evidence of sepsis. During his oral presentation, a pediatric resident in the ED reported, “This patient has a history of drug abuse and should not be admitted to a children’s hospital. She has been demanding pain meds and I feel she would be better served at the adult hospital.” What was driving the discomfort from the pediatric resident? Was a history of substance use disorder the primary driver, or was it related to other uncertainties with pain management or risk of a surgical complication? What parameters were and should have been in place for adult admissions? Did this pediatric facility have the necessary resources to provide Maria safe and effective medical care should she develop sepsis or require further surgical intervention? The dissonance between the need to respect provider and staff comfort zones, implement new systemwide hospital operations during an emergency, and promote health justice by confronting implicit bias while providing quality care was jarringly evident.

At the intersection of these seemingly impossible questions, dually trained internal medicine and pediatrics (med-peds) physicians had a unique vantage point, as they were accustomed to bridging the divide between adult and pediatric medicine in their practices. While no standardized models of care or quality metrics existed for adults hospitalized in pediatric settings, med-peds groups across the United States had developed healthcare delivery models prior to the pandemic for young adults who had survived chronic childhood medical conditions and continue to receive care in pediatric health systems.^4-7 Two med-peds physicians (authors LR, AJ) developed the Pediatric Overflow Planning Contingency Response Network, known as POPCoRN, to facilitate rapid information sharing among pediatric facilities to help implement and deliver equitable care to adults during a crisis. POPCoRN initially blossomed in April 2020 through a combination of easy access to virtual meetings and intentional inclusive recruitment across pediatric and med-peds provider communities. POPCoRN network members from diverse training, geographic, and health system backgrounds (including those from community, rural, and nonacademic centers) joined the network from across the country for real-time collaboration to troubleshoot common obstacles in their home institutions.

As POPCoRN members shared their challenges and institutional learnings, common themes were identified, such as management of intubated patients in non–intensive care unit (ICU) spaces; gaps in staffing with redeployment of residents and hospitalists; and dissemination of education, such as Advanced Cardiac Life Support (ACLS) webinars to frontline staff. POPCoRN prioritized a coordinated response to disseminate this evolving knowledge in multiple ways, including development of online resources, continued virtual meetings, and ultimately writing “Lessons Learned From COVID-19: A Practical Guide for Pediatric Facility Preparedness and Repurposing” (Appendix). This POPCoRN guide is a compilation of obstacles and solutions from providers who worked in pediatric health systems during the COVID-19 pandemic that required adjustments in care delivery models. As POPCoRN cofounders and leaders, we highlighted key lessons learned that are described in more detail in the guide.

Many pediatric facilities reported perseveration over the adult age cutoff accepted to the pediatric facility, only to realize the initial arbitrary age cutoff usually did not encompass enough patients to benefit local adult health systems. Using only strict age cutoffs also created an unnecessary barrier to accepting otherwise appropriate adult patients (eg, adult patient with controlled hypertension and a soft tissue infection). The pediatric facilities also created exclusion criteria through an evaluation of their available resources, such as subspecialists, procedural capabilities, and structural resources. For example, if the pediatric facility had access to in-person or telemedicine specialists comfortable with prescribing pain medications in the context of prior substance use disorder, Maria would be appropriate for admission. In contrast, if Maria were over a prescribed weight limit for the pediatric facility’s computed tomography scanner, this could lead to a delay in necessary workup for postsurgical complications. If possible, inclusion criteria should be based on characteristics (eg, age, primary diagnosis, SARS-CoV-2 status, weight) of adults accessing the health system at the time of the local surge to optimize chances of unburdening adult facilities. By using objective data to determine inclusion and exclusion criteria based on pediatric facility capabilities and limitations, Maria’s admission would be less subject to implicit systemic bias in screening.

The pandemic response was rapidly evolving and unpredictable. Planning required all affected parties at the table to effectively identify problems and solutions. Clinical and nonclinical groups were critical to planning operational logistics to provide safe care for adults in pediatric facilities. Though Maria had previously suffered from substance abuse, she had been sober for 4 years—a fact that changed the resource mobilization required for her hospitalization and discharge planning. Supporting equitable and quality healthcare for all patients, especially during a crisis, required intentional diversity in stakeholders and a multidisciplinary approach that included clinicians, social work, financial services, and case management. POPCoRN members found that using existing frameworks (eg, Model for Improvement) to conduct local rapid stakeholder analyses allowed their teams to identify key members of the discussion and understand any resistance to planning.⁸ Diverse teams were then able to contextualize their individual roles and the collaboration required for action. As with many other rationed healthcare resources, iterative stakeholder analysis and inclusion were necessary for prioritizing equity as hospital care for adults was reallocated from adult to pediatric facilities.⁹

Across care settings and training levels, the power of timely, honest, and transparent communication with leadership echoed throughout the network and could not be overemphasized. The cadence and modes of communication, while established by facility leaders, was best determined by explicitly asking team members for their needs. Often, leaders attempted to avoid communicating abrupt protocol changes to spare their teams additional stress and excessive correspondence. However, POPCoRN members found this approach often increased the perception among staff of a lack of transparency, which exacerbated feelings of discomfort and stress. While other specific examples of communication strategies are included in the POPCoRN guide, network members consistently noted that virtual open forums with leadership at regular intervals allowed teams to ask questions, raise concerns, and share ideas. In addition to open forums, leaders’ written communications regarding local medicolegal limitations and malpractice protection related to adult care should be distributed to staff. In Maria’s case, would provider discomfort and anxiety have been ameliorated with a proactive open forum to discuss the care of adults at the pediatric facility? Would that forum have called attention to staff educational and preparation needs around taking care of adults with a history substance use disorder? If so, this may have added a downstream benefit of decreasing effects of implicit bias amplified by stress.¹⁰

“Just-in-time” resources included not only educational materials related to clinical management or new policies and procedures, but also adult medicine clinical expertise. The clinical team that admitted Maria, for example, would have benefited from an adult medicine–trained hospitalist to provide consultation for adult-specific issues, such as management of venous thromboembolism prophylaxis and appropriate doses of intravenous pain medications in the setting of prior substance use disorder. The type of pediatric facility and their resource availability determined the network of clinical support. Pediatric facilities within an adult institution often had direct access to adult medicine–trained hospitalists, adult subspecialists, and adult-specific procedures. A freestanding pediatric facility, geographically separated from any adult facility, required more frequent use of telehealth adult consultative services. Many facilities relied on med-peds colleagues as the backbone of various care delivery models, as these physicians, residents, fellows, and attendings could bridge both the adult patient’s clinical needs and issues related to health system navigation (eg, postdischarge follow-up).

Addressing adult health emergencies efficiently and effectively necessitated identifying differences in pediatric versus adult code team responses. Pediatric facilities that already admitted patients older than 21 years used their experiences in redesigning rapid-response and code teams pre-COVID-19 to inform team and process variations. For example, how would the pediatric team have responded if Maria developed septic shock? Would they know first-line pressor support in adult patients? Collaboration with pharmacy and supply-management teams was required to ensure code carts were stocked with appropriate doses of adult medications used during ACLS. Operational leadership had to address patient flow factors such as: (1) which adults could receive critical care in the local pediatric ICU; and (2) how to emergently transfer an adult to another facility when necessary. In contrast, some community hospitals with both pediatric and adult departments had easier access to adult code teams and specialists. Although these processes were different among hospital systems and specific examples are included in the POPCoRN guide, the importance of collaboration and contingency planning remained the same.

Through intentionally fostering community and collaboration during the COVID-19 pandemic, the POPCoRN network has worked to help pediatric facilities build capacity to care for adult patients and overcome operational obstacles. POPCoRN has continued to bridge the silos in the healthcare system, particularly between pediatric and adult systems, in its larger vision to break down structural barriers leading to inadequate healthcare in vulnerable populations. The POPCoRN guide emphasizes the importance of addressing not only operational and logistical challenges, but centralizing health equity and provider wellness in all discussions and decisions. However, none of the implemented interventions discussed in the POPCoRN guide have been formally studied to evaluate outcomes for patients, healthcare staff and providers, or communities. This is an important limitation of this guide and is a necessary next step to effectively grow health-system capacity during crisis situations. Nonetheless, as a collaboratively written pragmatic tool, we hope the POPCoRN guide will serve as a reference for pediatric facilities, not only as COVID-19 continues to impact our communities, but also in the future if called upon to support adult patients during crisis.

Collaborators: All the collaborating authors listed below have contributed to the guide available in the appendix of the online version of this article, “Lessons Learned From COVID-19: A Practical Guide for Pediatric Facility Preparedness and Repurposing.” All the authors have provided consent to be listed.

Francisco Alvarez, MD, Stanford, California; Elizabeth Boggs, MD, MS, Aurora, Colorado; Rachel Boykan, MD, Stony Brook, New York; Alicia Caldwell, MD, Cincinnati, Ohio; Maryanne M. Chumpia, MD, MS, Torrance, California; Katharine N Clouser, MD, Hackensack, New Jersey; Alexandra L Coria, MD, Brooklyn, New York; Clare C Crosh, DO, Cincinnati, Ohio; Magna Dias, MD, Bridgeport, Connecticut; Laura N El-Hage, MD, Philadelphia, Pennsylvania; Jeff Foti, MD, Seattle, Washington; Mirna Giordano, MD, New York, New York; Sheena Gupta, MD, MBA, Evanston, Illinois; Laura Nell Hodo, MD, New York, New York; Ashley Jenkins, MD, MS, Rochester, New York; Anika Kumar, MD, Cleveland, Ohio; Merlin C Lowe, MD, Tuscon, Arizona; Brittany Middleton, MD, Pasadena, California; Sage Myers, MD, Philadelphia, Pennsylvania; Anik Patel, MD, Salt Lake City, Utah; Leah Ratner, MD, MS, Boston, Massachusetts; Shela Sridhar, MD, MPH, Boston, Massachusetts; Nathan Stehouwer, MD, Cleveland, Ohio; Julie Sylvester, DO, Mount Kisco, New York; Dava Szalda, MD, MSHP, Philadelphia, Pennsylvania; Heather Toth, MD, Milwaukee, Wisconsin; Krista Tuomela, MD, Milwaukee, Wisconsin; Ronald Williams, MD, Hershey, Pennsylvania.

As the COVID-19 pandemic surged in March 2020 in the United States, it was clear that severe COVID-19 and rates of hospitalization were much higher in adults than in children.¹ Pediatric facilities grappled with how to leverage empty beds and other underutilized human, clinical, and material resources to offset the overflowing adult facilities.^2,3 Pediatricians agonized about how to identify adult patients for whom they could provide safe and effective care, not only as individual clinicians, but also with adequate support from their local pediatric facility and health system.

Maria* (*name changed) was a young adult whose experience with her local health system highlighted common and addressable issues that arose when pediatric facilities aimed to care for adult populations. Adult hospitals were already above capacity caring for acutely ill patients with COVID-19, and a local freestanding children’s hospital offered to offload young adult patients up to age 30 years. Maria, a 26-year-old, had just been transferred from an adult emergency department (ED) to the children’s hospital ED for management of postoperative pain after a recent appendectomy. There was concern for possible abscess formation, but no evidence of sepsis. During his oral presentation, a pediatric resident in the ED reported, “This patient has a history of drug abuse and should not be admitted to a children’s hospital. She has been demanding pain meds and I feel she would be better served at the adult hospital.” What was driving the discomfort from the pediatric resident? Was a history of substance use disorder the primary driver, or was it related to other uncertainties with pain management or risk of a surgical complication? What parameters were and should have been in place for adult admissions? Did this pediatric facility have the necessary resources to provide Maria safe and effective medical care should she develop sepsis or require further surgical intervention? The dissonance between the need to respect provider and staff comfort zones, implement new systemwide hospital operations during an emergency, and promote health justice by confronting implicit bias while providing quality care was jarringly evident.

At the intersection of these seemingly impossible questions, dually trained internal medicine and pediatrics (med-peds) physicians had a unique vantage point, as they were accustomed to bridging the divide between adult and pediatric medicine in their practices. While no standardized models of care or quality metrics existed for adults hospitalized in pediatric settings, med-peds groups across the United States had developed healthcare delivery models prior to the pandemic for young adults who had survived chronic childhood medical conditions and continue to receive care in pediatric health systems.^4-7 Two med-peds physicians (authors LR, AJ) developed the Pediatric Overflow Planning Contingency Response Network, known as POPCoRN, to facilitate rapid information sharing among pediatric facilities to help implement and deliver equitable care to adults during a crisis. POPCoRN initially blossomed in April 2020 through a combination of easy access to virtual meetings and intentional inclusive recruitment across pediatric and med-peds provider communities. POPCoRN network members from diverse training, geographic, and health system backgrounds (including those from community, rural, and nonacademic centers) joined the network from across the country for real-time collaboration to troubleshoot common obstacles in their home institutions.

As POPCoRN members shared their challenges and institutional learnings, common themes were identified, such as management of intubated patients in non–intensive care unit (ICU) spaces; gaps in staffing with redeployment of residents and hospitalists; and dissemination of education, such as Advanced Cardiac Life Support (ACLS) webinars to frontline staff. POPCoRN prioritized a coordinated response to disseminate this evolving knowledge in multiple ways, including development of online resources, continued virtual meetings, and ultimately writing “Lessons Learned From COVID-19: A Practical Guide for Pediatric Facility Preparedness and Repurposing” (Appendix). This POPCoRN guide is a compilation of obstacles and solutions from providers who worked in pediatric health systems during the COVID-19 pandemic that required adjustments in care delivery models. As POPCoRN cofounders and leaders, we highlighted key lessons learned that are described in more detail in the guide.

Many pediatric facilities reported perseveration over the adult age cutoff accepted to the pediatric facility, only to realize the initial arbitrary age cutoff usually did not encompass enough patients to benefit local adult health systems. Using only strict age cutoffs also created an unnecessary barrier to accepting otherwise appropriate adult patients (eg, adult patient with controlled hypertension and a soft tissue infection). The pediatric facilities also created exclusion criteria through an evaluation of their available resources, such as subspecialists, procedural capabilities, and structural resources. For example, if the pediatric facility had access to in-person or telemedicine specialists comfortable with prescribing pain medications in the context of prior substance use disorder, Maria would be appropriate for admission. In contrast, if Maria were over a prescribed weight limit for the pediatric facility’s computed tomography scanner, this could lead to a delay in necessary workup for postsurgical complications. If possible, inclusion criteria should be based on characteristics (eg, age, primary diagnosis, SARS-CoV-2 status, weight) of adults accessing the health system at the time of the local surge to optimize chances of unburdening adult facilities. By using objective data to determine inclusion and exclusion criteria based on pediatric facility capabilities and limitations, Maria’s admission would be less subject to implicit systemic bias in screening.

The pandemic response was rapidly evolving and unpredictable. Planning required all affected parties at the table to effectively identify problems and solutions. Clinical and nonclinical groups were critical to planning operational logistics to provide safe care for adults in pediatric facilities. Though Maria had previously suffered from substance abuse, she had been sober for 4 years—a fact that changed the resource mobilization required for her hospitalization and discharge planning. Supporting equitable and quality healthcare for all patients, especially during a crisis, required intentional diversity in stakeholders and a multidisciplinary approach that included clinicians, social work, financial services, and case management. POPCoRN members found that using existing frameworks (eg, Model for Improvement) to conduct local rapid stakeholder analyses allowed their teams to identify key members of the discussion and understand any resistance to planning.⁸ Diverse teams were then able to contextualize their individual roles and the collaboration required for action. As with many other rationed healthcare resources, iterative stakeholder analysis and inclusion were necessary for prioritizing equity as hospital care for adults was reallocated from adult to pediatric facilities.⁹

Across care settings and training levels, the power of timely, honest, and transparent communication with leadership echoed throughout the network and could not be overemphasized. The cadence and modes of communication, while established by facility leaders, was best determined by explicitly asking team members for their needs. Often, leaders attempted to avoid communicating abrupt protocol changes to spare their teams additional stress and excessive correspondence. However, POPCoRN members found this approach often increased the perception among staff of a lack of transparency, which exacerbated feelings of discomfort and stress. While other specific examples of communication strategies are included in the POPCoRN guide, network members consistently noted that virtual open forums with leadership at regular intervals allowed teams to ask questions, raise concerns, and share ideas. In addition to open forums, leaders’ written communications regarding local medicolegal limitations and malpractice protection related to adult care should be distributed to staff. In Maria’s case, would provider discomfort and anxiety have been ameliorated with a proactive open forum to discuss the care of adults at the pediatric facility? Would that forum have called attention to staff educational and preparation needs around taking care of adults with a history substance use disorder? If so, this may have added a downstream benefit of decreasing effects of implicit bias amplified by stress.¹⁰

“Just-in-time” resources included not only educational materials related to clinical management or new policies and procedures, but also adult medicine clinical expertise. The clinical team that admitted Maria, for example, would have benefited from an adult medicine–trained hospitalist to provide consultation for adult-specific issues, such as management of venous thromboembolism prophylaxis and appropriate doses of intravenous pain medications in the setting of prior substance use disorder. The type of pediatric facility and their resource availability determined the network of clinical support. Pediatric facilities within an adult institution often had direct access to adult medicine–trained hospitalists, adult subspecialists, and adult-specific procedures. A freestanding pediatric facility, geographically separated from any adult facility, required more frequent use of telehealth adult consultative services. Many facilities relied on med-peds colleagues as the backbone of various care delivery models, as these physicians, residents, fellows, and attendings could bridge both the adult patient’s clinical needs and issues related to health system navigation (eg, postdischarge follow-up).

Addressing adult health emergencies efficiently and effectively necessitated identifying differences in pediatric versus adult code team responses. Pediatric facilities that already admitted patients older than 21 years used their experiences in redesigning rapid-response and code teams pre-COVID-19 to inform team and process variations. For example, how would the pediatric team have responded if Maria developed septic shock? Would they know first-line pressor support in adult patients? Collaboration with pharmacy and supply-management teams was required to ensure code carts were stocked with appropriate doses of adult medications used during ACLS. Operational leadership had to address patient flow factors such as: (1) which adults could receive critical care in the local pediatric ICU; and (2) how to emergently transfer an adult to another facility when necessary. In contrast, some community hospitals with both pediatric and adult departments had easier access to adult code teams and specialists. Although these processes were different among hospital systems and specific examples are included in the POPCoRN guide, the importance of collaboration and contingency planning remained the same.

Through intentionally fostering community and collaboration during the COVID-19 pandemic, the POPCoRN network has worked to help pediatric facilities build capacity to care for adult patients and overcome operational obstacles. POPCoRN has continued to bridge the silos in the healthcare system, particularly between pediatric and adult systems, in its larger vision to break down structural barriers leading to inadequate healthcare in vulnerable populations. The POPCoRN guide emphasizes the importance of addressing not only operational and logistical challenges, but centralizing health equity and provider wellness in all discussions and decisions. However, none of the implemented interventions discussed in the POPCoRN guide have been formally studied to evaluate outcomes for patients, healthcare staff and providers, or communities. This is an important limitation of this guide and is a necessary next step to effectively grow health-system capacity during crisis situations. Nonetheless, as a collaboratively written pragmatic tool, we hope the POPCoRN guide will serve as a reference for pediatric facilities, not only as COVID-19 continues to impact our communities, but also in the future if called upon to support adult patients during crisis.

Collaborators: All the collaborating authors listed below have contributed to the guide available in the appendix of the online version of this article, “Lessons Learned From COVID-19: A Practical Guide for Pediatric Facility Preparedness and Repurposing.” All the authors have provided consent to be listed.

Francisco Alvarez, MD, Stanford, California; Elizabeth Boggs, MD, MS, Aurora, Colorado; Rachel Boykan, MD, Stony Brook, New York; Alicia Caldwell, MD, Cincinnati, Ohio; Maryanne M. Chumpia, MD, MS, Torrance, California; Katharine N Clouser, MD, Hackensack, New Jersey; Alexandra L Coria, MD, Brooklyn, New York; Clare C Crosh, DO, Cincinnati, Ohio; Magna Dias, MD, Bridgeport, Connecticut; Laura N El-Hage, MD, Philadelphia, Pennsylvania; Jeff Foti, MD, Seattle, Washington; Mirna Giordano, MD, New York, New York; Sheena Gupta, MD, MBA, Evanston, Illinois; Laura Nell Hodo, MD, New York, New York; Ashley Jenkins, MD, MS, Rochester, New York; Anika Kumar, MD, Cleveland, Ohio; Merlin C Lowe, MD, Tuscon, Arizona; Brittany Middleton, MD, Pasadena, California; Sage Myers, MD, Philadelphia, Pennsylvania; Anik Patel, MD, Salt Lake City, Utah; Leah Ratner, MD, MS, Boston, Massachusetts; Shela Sridhar, MD, MPH, Boston, Massachusetts; Nathan Stehouwer, MD, Cleveland, Ohio; Julie Sylvester, DO, Mount Kisco, New York; Dava Szalda, MD, MSHP, Philadelphia, Pennsylvania; Heather Toth, MD, Milwaukee, Wisconsin; Krista Tuomela, MD, Milwaukee, Wisconsin; Ronald Williams, MD, Hershey, Pennsylvania.

As the COVID-19 pandemic surged in March 2020 in the United States, it was clear that severe COVID-19 and rates of hospitalization were much higher in adults than in children.¹ Pediatric facilities grappled with how to leverage empty beds and other underutilized human, clinical, and material resources to offset the overflowing adult facilities.^2,3 Pediatricians agonized about how to identify adult patients for whom they could provide safe and effective care, not only as individual clinicians, but also with adequate support from their local pediatric facility and health system.

Maria* (*name changed) was a young adult whose experience with her local health system highlighted common and addressable issues that arose when pediatric facilities aimed to care for adult populations. Adult hospitals were already above capacity caring for acutely ill patients with COVID-19, and a local freestanding children’s hospital offered to offload young adult patients up to age 30 years. Maria, a 26-year-old, had just been transferred from an adult emergency department (ED) to the children’s hospital ED for management of postoperative pain after a recent appendectomy. There was concern for possible abscess formation, but no evidence of sepsis. During his oral presentation, a pediatric resident in the ED reported, “This patient has a history of drug abuse and should not be admitted to a children’s hospital. She has been demanding pain meds and I feel she would be better served at the adult hospital.” What was driving the discomfort from the pediatric resident? Was a history of substance use disorder the primary driver, or was it related to other uncertainties with pain management or risk of a surgical complication? What parameters were and should have been in place for adult admissions? Did this pediatric facility have the necessary resources to provide Maria safe and effective medical care should she develop sepsis or require further surgical intervention? The dissonance between the need to respect provider and staff comfort zones, implement new systemwide hospital operations during an emergency, and promote health justice by confronting implicit bias while providing quality care was jarringly evident.

At the intersection of these seemingly impossible questions, dually trained internal medicine and pediatrics (med-peds) physicians had a unique vantage point, as they were accustomed to bridging the divide between adult and pediatric medicine in their practices. While no standardized models of care or quality metrics existed for adults hospitalized in pediatric settings, med-peds groups across the United States had developed healthcare delivery models prior to the pandemic for young adults who had survived chronic childhood medical conditions and continue to receive care in pediatric health systems.^4-7 Two med-peds physicians (authors LR, AJ) developed the Pediatric Overflow Planning Contingency Response Network, known as POPCoRN, to facilitate rapid information sharing among pediatric facilities to help implement and deliver equitable care to adults during a crisis. POPCoRN initially blossomed in April 2020 through a combination of easy access to virtual meetings and intentional inclusive recruitment across pediatric and med-peds provider communities. POPCoRN network members from diverse training, geographic, and health system backgrounds (including those from community, rural, and nonacademic centers) joined the network from across the country for real-time collaboration to troubleshoot common obstacles in their home institutions.

As POPCoRN members shared their challenges and institutional learnings, common themes were identified, such as management of intubated patients in non–intensive care unit (ICU) spaces; gaps in staffing with redeployment of residents and hospitalists; and dissemination of education, such as Advanced Cardiac Life Support (ACLS) webinars to frontline staff. POPCoRN prioritized a coordinated response to disseminate this evolving knowledge in multiple ways, including development of online resources, continued virtual meetings, and ultimately writing “Lessons Learned From COVID-19: A Practical Guide for Pediatric Facility Preparedness and Repurposing” (Appendix). This POPCoRN guide is a compilation of obstacles and solutions from providers who worked in pediatric health systems during the COVID-19 pandemic that required adjustments in care delivery models. As POPCoRN cofounders and leaders, we highlighted key lessons learned that are described in more detail in the guide.

Many pediatric facilities reported perseveration over the adult age cutoff accepted to the pediatric facility, only to realize the initial arbitrary age cutoff usually did not encompass enough patients to benefit local adult health systems. Using only strict age cutoffs also created an unnecessary barrier to accepting otherwise appropriate adult patients (eg, adult patient with controlled hypertension and a soft tissue infection). The pediatric facilities also created exclusion criteria through an evaluation of their available resources, such as subspecialists, procedural capabilities, and structural resources. For example, if the pediatric facility had access to in-person or telemedicine specialists comfortable with prescribing pain medications in the context of prior substance use disorder, Maria would be appropriate for admission. In contrast, if Maria were over a prescribed weight limit for the pediatric facility’s computed tomography scanner, this could lead to a delay in necessary workup for postsurgical complications. If possible, inclusion criteria should be based on characteristics (eg, age, primary diagnosis, SARS-CoV-2 status, weight) of adults accessing the health system at the time of the local surge to optimize chances of unburdening adult facilities. By using objective data to determine inclusion and exclusion criteria based on pediatric facility capabilities and limitations, Maria’s admission would be less subject to implicit systemic bias in screening.

The pandemic response was rapidly evolving and unpredictable. Planning required all affected parties at the table to effectively identify problems and solutions. Clinical and nonclinical groups were critical to planning operational logistics to provide safe care for adults in pediatric facilities. Though Maria had previously suffered from substance abuse, she had been sober for 4 years—a fact that changed the resource mobilization required for her hospitalization and discharge planning. Supporting equitable and quality healthcare for all patients, especially during a crisis, required intentional diversity in stakeholders and a multidisciplinary approach that included clinicians, social work, financial services, and case management. POPCoRN members found that using existing frameworks (eg, Model for Improvement) to conduct local rapid stakeholder analyses allowed their teams to identify key members of the discussion and understand any resistance to planning.⁸ Diverse teams were then able to contextualize their individual roles and the collaboration required for action. As with many other rationed healthcare resources, iterative stakeholder analysis and inclusion were necessary for prioritizing equity as hospital care for adults was reallocated from adult to pediatric facilities.⁹

Across care settings and training levels, the power of timely, honest, and transparent communication with leadership echoed throughout the network and could not be overemphasized. The cadence and modes of communication, while established by facility leaders, was best determined by explicitly asking team members for their needs. Often, leaders attempted to avoid communicating abrupt protocol changes to spare their teams additional stress and excessive correspondence. However, POPCoRN members found this approach often increased the perception among staff of a lack of transparency, which exacerbated feelings of discomfort and stress. While other specific examples of communication strategies are included in the POPCoRN guide, network members consistently noted that virtual open forums with leadership at regular intervals allowed teams to ask questions, raise concerns, and share ideas. In addition to open forums, leaders’ written communications regarding local medicolegal limitations and malpractice protection related to adult care should be distributed to staff. In Maria’s case, would provider discomfort and anxiety have been ameliorated with a proactive open forum to discuss the care of adults at the pediatric facility? Would that forum have called attention to staff educational and preparation needs around taking care of adults with a history substance use disorder? If so, this may have added a downstream benefit of decreasing effects of implicit bias amplified by stress.¹⁰

“Just-in-time” resources included not only educational materials related to clinical management or new policies and procedures, but also adult medicine clinical expertise. The clinical team that admitted Maria, for example, would have benefited from an adult medicine–trained hospitalist to provide consultation for adult-specific issues, such as management of venous thromboembolism prophylaxis and appropriate doses of intravenous pain medications in the setting of prior substance use disorder. The type of pediatric facility and their resource availability determined the network of clinical support. Pediatric facilities within an adult institution often had direct access to adult medicine–trained hospitalists, adult subspecialists, and adult-specific procedures. A freestanding pediatric facility, geographically separated from any adult facility, required more frequent use of telehealth adult consultative services. Many facilities relied on med-peds colleagues as the backbone of various care delivery models, as these physicians, residents, fellows, and attendings could bridge both the adult patient’s clinical needs and issues related to health system navigation (eg, postdischarge follow-up).

Addressing adult health emergencies efficiently and effectively necessitated identifying differences in pediatric versus adult code team responses. Pediatric facilities that already admitted patients older than 21 years used their experiences in redesigning rapid-response and code teams pre-COVID-19 to inform team and process variations. For example, how would the pediatric team have responded if Maria developed septic shock? Would they know first-line pressor support in adult patients? Collaboration with pharmacy and supply-management teams was required to ensure code carts were stocked with appropriate doses of adult medications used during ACLS. Operational leadership had to address patient flow factors such as: (1) which adults could receive critical care in the local pediatric ICU; and (2) how to emergently transfer an adult to another facility when necessary. In contrast, some community hospitals with both pediatric and adult departments had easier access to adult code teams and specialists. Although these processes were different among hospital systems and specific examples are included in the POPCoRN guide, the importance of collaboration and contingency planning remained the same.

Through intentionally fostering community and collaboration during the COVID-19 pandemic, the POPCoRN network has worked to help pediatric facilities build capacity to care for adult patients and overcome operational obstacles. POPCoRN has continued to bridge the silos in the healthcare system, particularly between pediatric and adult systems, in its larger vision to break down structural barriers leading to inadequate healthcare in vulnerable populations. The POPCoRN guide emphasizes the importance of addressing not only operational and logistical challenges, but centralizing health equity and provider wellness in all discussions and decisions. However, none of the implemented interventions discussed in the POPCoRN guide have been formally studied to evaluate outcomes for patients, healthcare staff and providers, or communities. This is an important limitation of this guide and is a necessary next step to effectively grow health-system capacity during crisis situations. Nonetheless, as a collaboratively written pragmatic tool, we hope the POPCoRN guide will serve as a reference for pediatric facilities, not only as COVID-19 continues to impact our communities, but also in the future if called upon to support adult patients during crisis.

Collaborators: All the collaborating authors listed below have contributed to the guide available in the appendix of the online version of this article, “Lessons Learned From COVID-19: A Practical Guide for Pediatric Facility Preparedness and Repurposing.” All the authors have provided consent to be listed.

Francisco Alvarez, MD, Stanford, California; Elizabeth Boggs, MD, MS, Aurora, Colorado; Rachel Boykan, MD, Stony Brook, New York; Alicia Caldwell, MD, Cincinnati, Ohio; Maryanne M. Chumpia, MD, MS, Torrance, California; Katharine N Clouser, MD, Hackensack, New Jersey; Alexandra L Coria, MD, Brooklyn, New York; Clare C Crosh, DO, Cincinnati, Ohio; Magna Dias, MD, Bridgeport, Connecticut; Laura N El-Hage, MD, Philadelphia, Pennsylvania; Jeff Foti, MD, Seattle, Washington; Mirna Giordano, MD, New York, New York; Sheena Gupta, MD, MBA, Evanston, Illinois; Laura Nell Hodo, MD, New York, New York; Ashley Jenkins, MD, MS, Rochester, New York; Anika Kumar, MD, Cleveland, Ohio; Merlin C Lowe, MD, Tuscon, Arizona; Brittany Middleton, MD, Pasadena, California; Sage Myers, MD, Philadelphia, Pennsylvania; Anik Patel, MD, Salt Lake City, Utah; Leah Ratner, MD, MS, Boston, Massachusetts; Shela Sridhar, MD, MPH, Boston, Massachusetts; Nathan Stehouwer, MD, Cleveland, Ohio; Julie Sylvester, DO, Mount Kisco, New York; Dava Szalda, MD, MSHP, Philadelphia, Pennsylvania; Heather Toth, MD, Milwaukee, Wisconsin; Krista Tuomela, MD, Milwaukee, Wisconsin; Ronald Williams, MD, Hershey, Pennsylvania.

During hospitalization, patients spend 87% to 100% of their time in bed.¹ This prolonged immobilization is a key contributor to the development of hospital-associated disability (HAD), defined as a new loss of ability to complete one or more activities of daily living (ADLs) without assistance after hospital discharge. HAD can lead to readmissions, institutionalization, and death and occurs in approximately one-third of all hospitalized patients.^2,3 The most effective way to prevent HAD is by mobilizing patients early and throughout their hospitalization.⁴ Typically, physical therapists are the primary team members responsible for mobilizing patients, but they are a constrained resource in most inpatient settings.

The Activity Measure-Post Acute Care Inpatient Mobility Short Form (AM-PAC IMSF) is a validated tool for measuring physical function.⁵ The AM-PAC score has been used to predict discharge destination within 48 hours of admission⁶ and as a guide to allocate inpatient therapy referrals on a medical and a neurosurgical service.^7,8 To date, however, no studies have used AM-PAC scores to evaluate overutilization of physical therapy consults on direct care hospital medicine services. In this study, we aimed to assess the potential overutilization of physical therapy consults on direct care hospital medicine services using validated AM-PAC score cutoffs.

Study Design and Setting

We analyzed a retrospective cohort of admissions from September 30, 2018, through September 29, 2019, on all direct care hospital medicine services at the University of Chicago Medical Center (UC), Illinois. These services included general medicine, oncology, transplant (renal, lung, and liver), cardiology, and cirrhotic populations at the medical-surgical and telemetry level of care. All patients were hospitalized for longer than 48 hours. Patients who left against medical advice; died; were discharged to hospice, another hospital, or an inpatient psychiatric facility; or received no physical therapy referral during admission were excluded. For the remaining patients, we obtained age, sex, admission and discharge dates, admission and discharge AM-PAC scores, and discharge disposition.

Mobility Measure

At UC, the inpatient mobility protocol requires nursing staff to assess and document AM-PAC mobility scores for each patient at the time of admission and every nursing shift thereafter. They utilize the original version of the AM-PAC “6-Clicks” Basic Mobility score, which includes three questions assessing difficulty with mobility and three questions assessing help needed with mobility activities. It has high interrater reliability, with an intraclass correlation coefficient of 0.85.⁹

Outcomes and Predictors

The primary outcome was “potential overutilization.” Secondary outcomes were discharge disposition and change in mobility. Our predictors included admission AM-PAC score, age, and sex. Based on previous studies that validated an AM-PAC score of 42.9 (raw score, 17) as a cutoff for predicting discharge to home,⁶ we defined physical therapy consults as “potentially inappropriate” in patients with admission AM-PAC scores >43.63 (raw score, 18) who were discharged to home. Likewise, in the UC mobility protocol, nursing staff independently mobilize patients with AM-PAC scores >18, another rationale to use this cutoff for defining physical therapy consult inappropriateness. “Discharge to home” was defined as going home with no additional needs or services, going home with outpatient physical therapy, or going home with home health physical therapy services, since none of these require inpatient physical therapy assessment for the order to be placed. Discharge to long-term acute care, skilled nursing facility, subacute rehabilitation facility, or acute rehabilitation facility were considered “discharge to post–acute care.” Loss of mobility was calculated as: discharge AM-PAC − admission AM-PAC, termed delta AM-PAC.

Statistical Analysis

Descriptive statistics were used to summarize age (mean and SD) and age categorized as <65 years or ≥65 years, sex (male or female), admission AM-PAC score (mean and SD) and categorization (≤43.63 or >43.63), discharge AM-PAC score (mean and SD), and discharge destination (home vs post–acute care). Chi-square analysis was used to test for associations between admission AM-PAC score and delta AM-PAC. Two-sample t-test was used to test for difference in mean delta AM-PAC between admission AM-PAC groups. Multivariable logistic regression was used to test for independent associations between age, sex, and admission AM-PAC score and odds of being discharged to home, controlling for length of stay. P values of <.05 were considered statistically significant for all tests. Analyses were performed using Stata statistical software, release 16 (StataCorp LLC).

During the 1-year study period, 3592 admissions with physical therapy consults occurred on the direct care hospital medicine services (58% of all admissions). Mean age was 66.3 years (SD, 15.4 years), and 48% of patients were female. The mean admission AM-PAC score was 43.9 (SD, 11.1), and the mean discharge AM-PAC score was 46.8 (SD, 10.8). In our sample, 38% of physical therapy consults were for patients with an AM-PAC score >43.63 who were discharged to home and were therefore deemed “potential overutilization.” Of those, 40% were for patients who were 65 years or younger (18% of all physical therapy consults) (Table 1).

A higher proportion of patients with AM-PAC scores >43.63 were discharged to home compared with those with AM-PAC scores ≤43.63 (89% vs 55%; χ² [1, N = 3099], 396.5; P < .001). More patients younger than 65 years were discharged to home compared with those 65 years and older (79% vs 63%; χ² [1, N = 3099], 113.6; P < .001). Additionally, for all patients younger than 65 years, those with AM-PAC score >43.63 were discharged to home more frequently than those with AM-PAC ≤43.63 (92% vs 66%, χ² [1, N = 1,354], 134.4; P < .001). For 11% (n = 147) of the high-mobility group, the patient was not discharged home but was sent to post–acute care. Reviewing these patient charts showed the reasons for discharge to post–acute care were predominantly personal or social needs (eg, homelessness, need for 24-hour supervision with no family support, patient request) or medical needs (eg, intravenous antibiotics or new tubes, lines, drains, or medications requiring extra nursing support or management). Only 16% of patients in this group (n = 23) experienced deconditioning necessitating physical therapy consult during hospitalization, per their record.

Compared with patients with admission AM-PAC score >43.63, patients with admission AM-PAC ≤43.63 had significantly different changes in mobility as measured by mean delta AM-PAC score (delta AM-PAC, –0.41 for AM-PAC >43.63 vs +5.69 for AM-PAC ≤43.63; t (3097) = –20.3; P < .001) (Table 1).

In multivariate logistic regression, AM-PAC >43.63 (OR, 5.38; 95% CI, 4.36-2.89; P < .001) and age younger than 65 years (OR, 2.40; 95% CI, 1.99-2.90; P < .001) were associated with increased odds of discharge to home (Table 2).

In this study, we found that physical therapists may be unnecessarily consulted on direct care hospitalist services as much as 38% of the time based on AM-PAC score. We also demonstrated that patients admitted with high mobility by AM-PAC score are more than five times as likely to be discharged to home. When admitted with high AM-PAC scores, patients had virtually no change in mobility during hospitalization, whereas patients with low AM-PAC scores gained mobility during hospitalization, underscoring the benefit of physical therapy referrals for this group.

Given resource scarcity and cost, achieving optimal physical therapy utilization is an important goal for healthcare systems.¹⁰ Appropriate allocation of physical therapy has the potential to improve outcomes from the patient to the payor level. While it may be necessary to consult physical therapy for reasons other than mobility later in the hospitalization, identifying patients who will benefit from skilled physical therapy at the time of admission can help prevent disability and institutionalization and shorten length of stay.^5,6 Likewise, decreasing physical therapy referrals for low-risk patients can increase the amount of time spent rehabilitating at-risk patients.

There are limitations of our study worth considering. First, our analyses did not consider whether physical therapy contributed to patients’ ability to return home after discharge. However, in our hospital, patients with AM-PAC >43.63 who cannot safely ambulate independently do progressive mobility with nursing staff. Our physical therapy leadership has also observed that the vast majority of highly mobile patients who are referred for physical therapy ultimately receive no treatment. Second, we did not consider discharge diagnosis, but our patient populations present with a wide variety of conditions, and it is impossible to predict their discharge diagnosis. By not including discharge diagnosis, we assess how AM-PAC performs on admission regardless of the medical condition for which someone is treated. Our hospital treats a high proportion of African American and a low proportion of White, Hispanic, and Asian American patients, limiting the generalizability of our findings. Although the AM-PAC “6-Clicks” score has been shown to have high interrater reliability among physical therapists, our AM-PAC scores are assessed and documented by our nursing staff, which might decrease accuracy. However, one single-center study noted an intraclass correlation coefficient of 0.96 between nurses and physical therapists for the AM-PAC “6-Clicks.”¹¹Despite these limitations, this study underscores the need to be more judicious in the decision to refer a patient for inpatient physical therapy, especially at the time of admission, and demonstrates the utility of using standardized mobility assessment to help in that decision-making process.

During hospitalization, patients spend 87% to 100% of their time in bed.¹ This prolonged immobilization is a key contributor to the development of hospital-associated disability (HAD), defined as a new loss of ability to complete one or more activities of daily living (ADLs) without assistance after hospital discharge. HAD can lead to readmissions, institutionalization, and death and occurs in approximately one-third of all hospitalized patients.^2,3 The most effective way to prevent HAD is by mobilizing patients early and throughout their hospitalization.⁴ Typically, physical therapists are the primary team members responsible for mobilizing patients, but they are a constrained resource in most inpatient settings.

The Activity Measure-Post Acute Care Inpatient Mobility Short Form (AM-PAC IMSF) is a validated tool for measuring physical function.⁵ The AM-PAC score has been used to predict discharge destination within 48 hours of admission⁶ and as a guide to allocate inpatient therapy referrals on a medical and a neurosurgical service.^7,8 To date, however, no studies have used AM-PAC scores to evaluate overutilization of physical therapy consults on direct care hospital medicine services. In this study, we aimed to assess the potential overutilization of physical therapy consults on direct care hospital medicine services using validated AM-PAC score cutoffs.

Study Design and Setting

We analyzed a retrospective cohort of admissions from September 30, 2018, through September 29, 2019, on all direct care hospital medicine services at the University of Chicago Medical Center (UC), Illinois. These services included general medicine, oncology, transplant (renal, lung, and liver), cardiology, and cirrhotic populations at the medical-surgical and telemetry level of care. All patients were hospitalized for longer than 48 hours. Patients who left against medical advice; died; were discharged to hospice, another hospital, or an inpatient psychiatric facility; or received no physical therapy referral during admission were excluded. For the remaining patients, we obtained age, sex, admission and discharge dates, admission and discharge AM-PAC scores, and discharge disposition.

Mobility Measure

At UC, the inpatient mobility protocol requires nursing staff to assess and document AM-PAC mobility scores for each patient at the time of admission and every nursing shift thereafter. They utilize the original version of the AM-PAC “6-Clicks” Basic Mobility score, which includes three questions assessing difficulty with mobility and three questions assessing help needed with mobility activities. It has high interrater reliability, with an intraclass correlation coefficient of 0.85.⁹

Outcomes and Predictors

The primary outcome was “potential overutilization.” Secondary outcomes were discharge disposition and change in mobility. Our predictors included admission AM-PAC score, age, and sex. Based on previous studies that validated an AM-PAC score of 42.9 (raw score, 17) as a cutoff for predicting discharge to home,⁶ we defined physical therapy consults as “potentially inappropriate” in patients with admission AM-PAC scores >43.63 (raw score, 18) who were discharged to home. Likewise, in the UC mobility protocol, nursing staff independently mobilize patients with AM-PAC scores >18, another rationale to use this cutoff for defining physical therapy consult inappropriateness. “Discharge to home” was defined as going home with no additional needs or services, going home with outpatient physical therapy, or going home with home health physical therapy services, since none of these require inpatient physical therapy assessment for the order to be placed. Discharge to long-term acute care, skilled nursing facility, subacute rehabilitation facility, or acute rehabilitation facility were considered “discharge to post–acute care.” Loss of mobility was calculated as: discharge AM-PAC − admission AM-PAC, termed delta AM-PAC.

Statistical Analysis

Descriptive statistics were used to summarize age (mean and SD) and age categorized as <65 years or ≥65 years, sex (male or female), admission AM-PAC score (mean and SD) and categorization (≤43.63 or >43.63), discharge AM-PAC score (mean and SD), and discharge destination (home vs post–acute care). Chi-square analysis was used to test for associations between admission AM-PAC score and delta AM-PAC. Two-sample t-test was used to test for difference in mean delta AM-PAC between admission AM-PAC groups. Multivariable logistic regression was used to test for independent associations between age, sex, and admission AM-PAC score and odds of being discharged to home, controlling for length of stay. P values of <.05 were considered statistically significant for all tests. Analyses were performed using Stata statistical software, release 16 (StataCorp LLC).

During the 1-year study period, 3592 admissions with physical therapy consults occurred on the direct care hospital medicine services (58% of all admissions). Mean age was 66.3 years (SD, 15.4 years), and 48% of patients were female. The mean admission AM-PAC score was 43.9 (SD, 11.1), and the mean discharge AM-PAC score was 46.8 (SD, 10.8). In our sample, 38% of physical therapy consults were for patients with an AM-PAC score >43.63 who were discharged to home and were therefore deemed “potential overutilization.” Of those, 40% were for patients who were 65 years or younger (18% of all physical therapy consults) (Table 1).

A higher proportion of patients with AM-PAC scores >43.63 were discharged to home compared with those with AM-PAC scores ≤43.63 (89% vs 55%; χ² [1, N = 3099], 396.5; P < .001). More patients younger than 65 years were discharged to home compared with those 65 years and older (79% vs 63%; χ² [1, N = 3099], 113.6; P < .001). Additionally, for all patients younger than 65 years, those with AM-PAC score >43.63 were discharged to home more frequently than those with AM-PAC ≤43.63 (92% vs 66%, χ² [1, N = 1,354], 134.4; P < .001). For 11% (n = 147) of the high-mobility group, the patient was not discharged home but was sent to post–acute care. Reviewing these patient charts showed the reasons for discharge to post–acute care were predominantly personal or social needs (eg, homelessness, need for 24-hour supervision with no family support, patient request) or medical needs (eg, intravenous antibiotics or new tubes, lines, drains, or medications requiring extra nursing support or management). Only 16% of patients in this group (n = 23) experienced deconditioning necessitating physical therapy consult during hospitalization, per their record.

Compared with patients with admission AM-PAC score >43.63, patients with admission AM-PAC ≤43.63 had significantly different changes in mobility as measured by mean delta AM-PAC score (delta AM-PAC, –0.41 for AM-PAC >43.63 vs +5.69 for AM-PAC ≤43.63; t (3097) = –20.3; P < .001) (Table 1).

In multivariate logistic regression, AM-PAC >43.63 (OR, 5.38; 95% CI, 4.36-2.89; P < .001) and age younger than 65 years (OR, 2.40; 95% CI, 1.99-2.90; P < .001) were associated with increased odds of discharge to home (Table 2).

In this study, we found that physical therapists may be unnecessarily consulted on direct care hospitalist services as much as 38% of the time based on AM-PAC score. We also demonstrated that patients admitted with high mobility by AM-PAC score are more than five times as likely to be discharged to home. When admitted with high AM-PAC scores, patients had virtually no change in mobility during hospitalization, whereas patients with low AM-PAC scores gained mobility during hospitalization, underscoring the benefit of physical therapy referrals for this group.

Given resource scarcity and cost, achieving optimal physical therapy utilization is an important goal for healthcare systems.¹⁰ Appropriate allocation of physical therapy has the potential to improve outcomes from the patient to the payor level. While it may be necessary to consult physical therapy for reasons other than mobility later in the hospitalization, identifying patients who will benefit from skilled physical therapy at the time of admission can help prevent disability and institutionalization and shorten length of stay.^5,6 Likewise, decreasing physical therapy referrals for low-risk patients can increase the amount of time spent rehabilitating at-risk patients.

There are limitations of our study worth considering. First, our analyses did not consider whether physical therapy contributed to patients’ ability to return home after discharge. However, in our hospital, patients with AM-PAC >43.63 who cannot safely ambulate independently do progressive mobility with nursing staff. Our physical therapy leadership has also observed that the vast majority of highly mobile patients who are referred for physical therapy ultimately receive no treatment. Second, we did not consider discharge diagnosis, but our patient populations present with a wide variety of conditions, and it is impossible to predict their discharge diagnosis. By not including discharge diagnosis, we assess how AM-PAC performs on admission regardless of the medical condition for which someone is treated. Our hospital treats a high proportion of African American and a low proportion of White, Hispanic, and Asian American patients, limiting the generalizability of our findings. Although the AM-PAC “6-Clicks” score has been shown to have high interrater reliability among physical therapists, our AM-PAC scores are assessed and documented by our nursing staff, which might decrease accuracy. However, one single-center study noted an intraclass correlation coefficient of 0.96 between nurses and physical therapists for the AM-PAC “6-Clicks.”¹¹Despite these limitations, this study underscores the need to be more judicious in the decision to refer a patient for inpatient physical therapy, especially at the time of admission, and demonstrates the utility of using standardized mobility assessment to help in that decision-making process.

During hospitalization, patients spend 87% to 100% of their time in bed.¹ This prolonged immobilization is a key contributor to the development of hospital-associated disability (HAD), defined as a new loss of ability to complete one or more activities of daily living (ADLs) without assistance after hospital discharge. HAD can lead to readmissions, institutionalization, and death and occurs in approximately one-third of all hospitalized patients.^2,3 The most effective way to prevent HAD is by mobilizing patients early and throughout their hospitalization.⁴ Typically, physical therapists are the primary team members responsible for mobilizing patients, but they are a constrained resource in most inpatient settings.

The Activity Measure-Post Acute Care Inpatient Mobility Short Form (AM-PAC IMSF) is a validated tool for measuring physical function.⁵ The AM-PAC score has been used to predict discharge destination within 48 hours of admission⁶ and as a guide to allocate inpatient therapy referrals on a medical and a neurosurgical service.^7,8 To date, however, no studies have used AM-PAC scores to evaluate overutilization of physical therapy consults on direct care hospital medicine services. In this study, we aimed to assess the potential overutilization of physical therapy consults on direct care hospital medicine services using validated AM-PAC score cutoffs.

Study Design and Setting

We analyzed a retrospective cohort of admissions from September 30, 2018, through September 29, 2019, on all direct care hospital medicine services at the University of Chicago Medical Center (UC), Illinois. These services included general medicine, oncology, transplant (renal, lung, and liver), cardiology, and cirrhotic populations at the medical-surgical and telemetry level of care. All patients were hospitalized for longer than 48 hours. Patients who left against medical advice; died; were discharged to hospice, another hospital, or an inpatient psychiatric facility; or received no physical therapy referral during admission were excluded. For the remaining patients, we obtained age, sex, admission and discharge dates, admission and discharge AM-PAC scores, and discharge disposition.

Mobility Measure

At UC, the inpatient mobility protocol requires nursing staff to assess and document AM-PAC mobility scores for each patient at the time of admission and every nursing shift thereafter. They utilize the original version of the AM-PAC “6-Clicks” Basic Mobility score, which includes three questions assessing difficulty with mobility and three questions assessing help needed with mobility activities. It has high interrater reliability, with an intraclass correlation coefficient of 0.85.⁹

Outcomes and Predictors

The primary outcome was “potential overutilization.” Secondary outcomes were discharge disposition and change in mobility. Our predictors included admission AM-PAC score, age, and sex. Based on previous studies that validated an AM-PAC score of 42.9 (raw score, 17) as a cutoff for predicting discharge to home,⁶ we defined physical therapy consults as “potentially inappropriate” in patients with admission AM-PAC scores >43.63 (raw score, 18) who were discharged to home. Likewise, in the UC mobility protocol, nursing staff independently mobilize patients with AM-PAC scores >18, another rationale to use this cutoff for defining physical therapy consult inappropriateness. “Discharge to home” was defined as going home with no additional needs or services, going home with outpatient physical therapy, or going home with home health physical therapy services, since none of these require inpatient physical therapy assessment for the order to be placed. Discharge to long-term acute care, skilled nursing facility, subacute rehabilitation facility, or acute rehabilitation facility were considered “discharge to post–acute care.” Loss of mobility was calculated as: discharge AM-PAC − admission AM-PAC, termed delta AM-PAC.

Statistical Analysis

Descriptive statistics were used to summarize age (mean and SD) and age categorized as <65 years or ≥65 years, sex (male or female), admission AM-PAC score (mean and SD) and categorization (≤43.63 or >43.63), discharge AM-PAC score (mean and SD), and discharge destination (home vs post–acute care). Chi-square analysis was used to test for associations between admission AM-PAC score and delta AM-PAC. Two-sample t-test was used to test for difference in mean delta AM-PAC between admission AM-PAC groups. Multivariable logistic regression was used to test for independent associations between age, sex, and admission AM-PAC score and odds of being discharged to home, controlling for length of stay. P values of <.05 were considered statistically significant for all tests. Analyses were performed using Stata statistical software, release 16 (StataCorp LLC).

During the 1-year study period, 3592 admissions with physical therapy consults occurred on the direct care hospital medicine services (58% of all admissions). Mean age was 66.3 years (SD, 15.4 years), and 48% of patients were female. The mean admission AM-PAC score was 43.9 (SD, 11.1), and the mean discharge AM-PAC score was 46.8 (SD, 10.8). In our sample, 38% of physical therapy consults were for patients with an AM-PAC score >43.63 who were discharged to home and were therefore deemed “potential overutilization.” Of those, 40% were for patients who were 65 years or younger (18% of all physical therapy consults) (Table 1).

A higher proportion of patients with AM-PAC scores >43.63 were discharged to home compared with those with AM-PAC scores ≤43.63 (89% vs 55%; χ² [1, N = 3099], 396.5; P < .001). More patients younger than 65 years were discharged to home compared with those 65 years and older (79% vs 63%; χ² [1, N = 3099], 113.6; P < .001). Additionally, for all patients younger than 65 years, those with AM-PAC score >43.63 were discharged to home more frequently than those with AM-PAC ≤43.63 (92% vs 66%, χ² [1, N = 1,354], 134.4; P < .001). For 11% (n = 147) of the high-mobility group, the patient was not discharged home but was sent to post–acute care. Reviewing these patient charts showed the reasons for discharge to post–acute care were predominantly personal or social needs (eg, homelessness, need for 24-hour supervision with no family support, patient request) or medical needs (eg, intravenous antibiotics or new tubes, lines, drains, or medications requiring extra nursing support or management). Only 16% of patients in this group (n = 23) experienced deconditioning necessitating physical therapy consult during hospitalization, per their record.

Compared with patients with admission AM-PAC score >43.63, patients with admission AM-PAC ≤43.63 had significantly different changes in mobility as measured by mean delta AM-PAC score (delta AM-PAC, –0.41 for AM-PAC >43.63 vs +5.69 for AM-PAC ≤43.63; t (3097) = –20.3; P < .001) (Table 1).

In multivariate logistic regression, AM-PAC >43.63 (OR, 5.38; 95% CI, 4.36-2.89; P < .001) and age younger than 65 years (OR, 2.40; 95% CI, 1.99-2.90; P < .001) were associated with increased odds of discharge to home (Table 2).

In this study, we found that physical therapists may be unnecessarily consulted on direct care hospitalist services as much as 38% of the time based on AM-PAC score. We also demonstrated that patients admitted with high mobility by AM-PAC score are more than five times as likely to be discharged to home. When admitted with high AM-PAC scores, patients had virtually no change in mobility during hospitalization, whereas patients with low AM-PAC scores gained mobility during hospitalization, underscoring the benefit of physical therapy referrals for this group.

Given resource scarcity and cost, achieving optimal physical therapy utilization is an important goal for healthcare systems.¹⁰ Appropriate allocation of physical therapy has the potential to improve outcomes from the patient to the payor level. While it may be necessary to consult physical therapy for reasons other than mobility later in the hospitalization, identifying patients who will benefit from skilled physical therapy at the time of admission can help prevent disability and institutionalization and shorten length of stay.^5,6 Likewise, decreasing physical therapy referrals for low-risk patients can increase the amount of time spent rehabilitating at-risk patients.

There are limitations of our study worth considering. First, our analyses did not consider whether physical therapy contributed to patients’ ability to return home after discharge. However, in our hospital, patients with AM-PAC >43.63 who cannot safely ambulate independently do progressive mobility with nursing staff. Our physical therapy leadership has also observed that the vast majority of highly mobile patients who are referred for physical therapy ultimately receive no treatment. Second, we did not consider discharge diagnosis, but our patient populations present with a wide variety of conditions, and it is impossible to predict their discharge diagnosis. By not including discharge diagnosis, we assess how AM-PAC performs on admission regardless of the medical condition for which someone is treated. Our hospital treats a high proportion of African American and a low proportion of White, Hispanic, and Asian American patients, limiting the generalizability of our findings. Although the AM-PAC “6-Clicks” score has been shown to have high interrater reliability among physical therapists, our AM-PAC scores are assessed and documented by our nursing staff, which might decrease accuracy. However, one single-center study noted an intraclass correlation coefficient of 0.96 between nurses and physical therapists for the AM-PAC “6-Clicks.”¹¹Despite these limitations, this study underscores the need to be more judicious in the decision to refer a patient for inpatient physical therapy, especially at the time of admission, and demonstrates the utility of using standardized mobility assessment to help in that decision-making process.

The COVID-19 pandemic dramatically altered how healthcare providers care for hospitalized patients. Many hospitals provided physical-distancing guidance to minimize viral transmission and preserve personal protective equipment. This guidance informed clinician behavior on rounds and in workspaces.¹ One study reported that clinicians maintained distance from patients by grouping medical interventions, utilizing telemedicine for rounding and consultations, and implementing respiratory isolation units (RIUs) to cohort patients with COVID-19.²

Although physical distancing is recommended during inpatient care, no study to date has used objective measures to quantify the degree to which clinical practice was influenced. We aimed to objectively quantify changes in 24-hour patient room–entries before and during the COVID-19 pandemic using data from existing heat sensors to assess differences in physical distancing in RIUs and general medicine units.

Study Design

A single-institution study was conducted at the University of Chicago Medicine, Illinois. Room entries were compared between a general medicine unit that transitioned to an RIU (unit A/RIU) and four general medicine units (unit B) using 24-hour patient room–entry data. Unit A was commissioned as an RIU to care exclusively for patients with confirmed COVID-19 on March 25, 2020, and decommissioned on June 23, 2020. Unit B cared for patients under investigation (PUIs) for COVID-19 and patients admitted for other reasons. PUIs were transferred to the RIU if positive for COVID-19. Hospital visitor restrictions were implemented on March 14, 2020, and lifted on June 29, 2020. The University of Chicago Institutional Review Board granted this project an exempt determination.

Data Collection

From January 1, 2020, to August 10, 2020, room-entry data were collected using the PURELL SMARTLINK hand-hygiene system (GOJO Industries, Inc.). This hand-hygiene compliance system tracks unit-level sanitizer dispenses and total room entries and exits via body heat sensors. Similar to our prior studies, this study extracted heat-sensor data to monitor room entries.^3,4

Data Analysis

Objective 24-hour room-entry data were analyzed for all units. Rooms with less than two daily entries were assumed to be unoccupied and excluded from the analysis. Hospital-wide physical-distancing guidance published on March 10, 2020, was used to delineate “prepandemic” and “pandemic” periods. Each department adopted these recommendations (eg, physical distancing, conducting prerounds virtually, limiting the number of people seeing patients, using iPads for virtual patient visits) as appropriate.

Interrupted time series analyses were used to examine room-entry changes before and during the pandemic. The segmented function in R 4.0.2 (R Core Team) was used to create a model and estimate final fitting parameters, uncertainties, and data breakpoints using a bootstrap restarting algorithm.⁵ The Davies test was used to determine statistical significance of breakpoints, which was defined as P < .05.

We examined data from January 1, 2020, to August 10, 2020, from 3283 patients who collectively experienced 655,615 room entries. Unit A/RIU cared for 395 patients during the prepandemic period and 542 patients during the pandemic period. Compared with patients from the prepandemic period, patients during the pandemic period were more likely to be Black (73.7% vs 77.9%) and less likely to be White (17.0% vs 8.7%) (P = .002); were less likely to have respiratory (19.8% vs 8.0%, P < .001) or gastrointestinal (12.4% vs 6.5%, P = .008) primary diagnoses; and had a higher mean case-mix index (1.74 vs 2.38, P < .001) (Table).

Unit B had 718 patients during the prepandemic period and 1628 patients during the pandemic period. Compared with patients from the prepandemic period, patients during the pandemic period were less likely to be female (58.8% vs 53.4%, P = .018); less likely to be Black (77.7% vs 74.7%) and Hispanic (5.7% vs 3.4%) (P < .001); more likely to have circulatory (10.1% vs 14.8%, P = .009) primary diagnoses; less likely to have respiratory (14.4% vs 6.8%, P < .001) primary diagnoses; and had a higher mean case-mix index (1.58 vs 1.82, P = .014) (Table).

During the prepandemic period, unit A/RIU averaged 27 occupied rooms per day. These rooms averaged 85.0 entries per room per day, with no statistically significant change over time. During the pandemic period, this unit averaged 24 daily occupied rooms, and these rooms averaged 44.4 entries per room per day. At the start of the pandemic, daily entries per room decreased by 51.9 (95% CI, 51.1-52.7). This equated to a 60.6% reduction from baseline (95% CI, 59.6%-61.5%; P < .001), with the lowest average occurring after RIU conversion on March 25, 2020 (letter F in Figure, A). Entries remained constant through the end of statewide stay-at-home orders (letter G in Figure, A) until RIU decommission on June 23, 2020 (letter H in Figure, A). Entries then increased by an average of 0.150 entries per room per day (95% CI, 0.097-0.202; P < .001), reaching 52.5 daily entries on August 10, 2020. This equated to 61.3% of prepandemic levels (95% CI, 61.3%-61.6%; P < .001) (Figure, A).

During the prepandemic period, Unit B averaged 63 daily occupied rooms, and these rooms averaged 76.9 entries per room per day, with no statistically significant change over time. During the pandemic period, these units averaged 64 daily occupied rooms, and these rooms averaged 72.4 entries per room per day. Briefly, at the start of the pandemic, daily entries per room decreased by 11.8 (95% CI, 11.6-12), equating to a 14.7% reduction from baseline (95% CI, 14.4%-14.9%; P < .001). Entries then increased by an average of 0.052 entries per room per day (95% CI, –0.01 to 0.115; P = .051), stabilizing in early August 2020 at an average of 74.1 daily entries. This equated to 92.2% of prepandemic levels (95% CI, 92%-92.3%; P < .001) (Figure, B).

Unit A/RIU experienced significantly greater average daily room entries during the prepandemic period (P < .001) and significantly fewer average daily room entries during the pandemic period (P < .001) than unit B. Although unit A and unit B cared for similar patient populations prior to the pandemic, unit B was located in a different building from the resident work room. This likely resulted in batched visits to patients, leading to fewer total room entries per day.

This is the first study to measure 24-hour patient room– entries as an objective proxy for physical distancing during the pandemic. Unit A/RIU saw an initial 60.6% decrease in room entries. In contrast, unit B, which cared for PUIs, saw a brief 14.7% decrease in room entries before returning to baseline. In all units, room entries increased over time, although this increase was greater in unit B.

Despite the institutional recommendation of physical distancing, only unit A/RIU saw a large and sustained decrease in room entries. The presence of patients with COVID-19 within this unit likely reminded clinicians of the ongoing need to physically distance. Clinicians may have been fearful of contracting COVID-19 and therefore more stringently followed physical-distancing guidance.

Changes in unit A/RIU room entries tracked with RIU conversion and decommission timeline (letters F and H in Figure, A, respectively) rather than statewide stay-at-home orders (letters E and G in Figure, A). Caring for patients with COVID-19 within the unit might have influenced clinician physical distancing more than state policy. Correspondingly, as the number of hospitalized patients with COVID-19 decreased, room entries trended toward baseline. The difficulty of sustaining behavioral changes has been demonstrated in healthcare settings, including at our own institution.^6-8 This gradual extinction in physical distancing could be due to several factors, such as fewer patients with COVID-19 or staff fatigue. Physical distancing may have been more extreme and suboptimal for care at the beginning of the pandemic owing to uncertainty or fear.

This work has implications for how to monitor physical distancing in healthcare facilities. Our study shows that behaviors can change rapidly, but sustaining change is difficult. This suggests the need for regular reinforcement of physical distancing with all staff. Additionally, cohorting patients on RIUs may result in greater physical distancing. It also highlights that PUIs serve as less of a cue to promote physical distancing, possibly due to increased confidence in and availability of COVID-19 tests and/or precautions fatigue.⁹ Objective room-entry monitoring systems, such as the one used in this study, can provide hospital leaders with crucial real-time feedback to monitor physical distancing practices and determine when and where re-education may be needed.

This study was conducted at a single urban, academic medical center, limiting its generalizability. Many other hospital policies implemented at the beginning of the pandemic may have influenced our results. We are unable to examine the type of clinician entering each room and for how long as well as entries in workrooms and breakrooms. Clinicians were not given real-time or retrospective feedback on room entries during the pandemic period. These data would be important to understand staff responses to physical distancing. Finally, while clinicians responded differently as the pandemic progressed and depending on which unit they were in, the ideal degree of physical distancing remains unknown. Although minimizing patient contact limits nosocomial viral spread, too little contact can also cause harm.

At the onset of the COVID-19 pandemic, 24-hour patient room–entries fell significantly in all units before increasing. This decrease was more pronounced in unit A/RIU. As the pandemic continues, hospitals could consider utilizing novel room-entry monitoring systems to guide physical-distancing implementation and staff education.

The authors thank Vera Chu for her support with data requests.

The COVID-19 pandemic dramatically altered how healthcare providers care for hospitalized patients. Many hospitals provided physical-distancing guidance to minimize viral transmission and preserve personal protective equipment. This guidance informed clinician behavior on rounds and in workspaces.¹ One study reported that clinicians maintained distance from patients by grouping medical interventions, utilizing telemedicine for rounding and consultations, and implementing respiratory isolation units (RIUs) to cohort patients with COVID-19.²

Although physical distancing is recommended during inpatient care, no study to date has used objective measures to quantify the degree to which clinical practice was influenced. We aimed to objectively quantify changes in 24-hour patient room–entries before and during the COVID-19 pandemic using data from existing heat sensors to assess differences in physical distancing in RIUs and general medicine units.

Study Design

A single-institution study was conducted at the University of Chicago Medicine, Illinois. Room entries were compared between a general medicine unit that transitioned to an RIU (unit A/RIU) and four general medicine units (unit B) using 24-hour patient room–entry data. Unit A was commissioned as an RIU to care exclusively for patients with confirmed COVID-19 on March 25, 2020, and decommissioned on June 23, 2020. Unit B cared for patients under investigation (PUIs) for COVID-19 and patients admitted for other reasons. PUIs were transferred to the RIU if positive for COVID-19. Hospital visitor restrictions were implemented on March 14, 2020, and lifted on June 29, 2020. The University of Chicago Institutional Review Board granted this project an exempt determination.

Data Collection

From January 1, 2020, to August 10, 2020, room-entry data were collected using the PURELL SMARTLINK hand-hygiene system (GOJO Industries, Inc.). This hand-hygiene compliance system tracks unit-level sanitizer dispenses and total room entries and exits via body heat sensors. Similar to our prior studies, this study extracted heat-sensor data to monitor room entries.^3,4

Data Analysis

Objective 24-hour room-entry data were analyzed for all units. Rooms with less than two daily entries were assumed to be unoccupied and excluded from the analysis. Hospital-wide physical-distancing guidance published on March 10, 2020, was used to delineate “prepandemic” and “pandemic” periods. Each department adopted these recommendations (eg, physical distancing, conducting prerounds virtually, limiting the number of people seeing patients, using iPads for virtual patient visits) as appropriate.

Interrupted time series analyses were used to examine room-entry changes before and during the pandemic. The segmented function in R 4.0.2 (R Core Team) was used to create a model and estimate final fitting parameters, uncertainties, and data breakpoints using a bootstrap restarting algorithm.⁵ The Davies test was used to determine statistical significance of breakpoints, which was defined as P < .05.

We examined data from January 1, 2020, to August 10, 2020, from 3283 patients who collectively experienced 655,615 room entries. Unit A/RIU cared for 395 patients during the prepandemic period and 542 patients during the pandemic period. Compared with patients from the prepandemic period, patients during the pandemic period were more likely to be Black (73.7% vs 77.9%) and less likely to be White (17.0% vs 8.7%) (P = .002); were less likely to have respiratory (19.8% vs 8.0%, P < .001) or gastrointestinal (12.4% vs 6.5%, P = .008) primary diagnoses; and had a higher mean case-mix index (1.74 vs 2.38, P < .001) (Table).

Unit B had 718 patients during the prepandemic period and 1628 patients during the pandemic period. Compared with patients from the prepandemic period, patients during the pandemic period were less likely to be female (58.8% vs 53.4%, P = .018); less likely to be Black (77.7% vs 74.7%) and Hispanic (5.7% vs 3.4%) (P < .001); more likely to have circulatory (10.1% vs 14.8%, P = .009) primary diagnoses; less likely to have respiratory (14.4% vs 6.8%, P < .001) primary diagnoses; and had a higher mean case-mix index (1.58 vs 1.82, P = .014) (Table).

During the prepandemic period, unit A/RIU averaged 27 occupied rooms per day. These rooms averaged 85.0 entries per room per day, with no statistically significant change over time. During the pandemic period, this unit averaged 24 daily occupied rooms, and these rooms averaged 44.4 entries per room per day. At the start of the pandemic, daily entries per room decreased by 51.9 (95% CI, 51.1-52.7). This equated to a 60.6% reduction from baseline (95% CI, 59.6%-61.5%; P < .001), with the lowest average occurring after RIU conversion on March 25, 2020 (letter F in Figure, A). Entries remained constant through the end of statewide stay-at-home orders (letter G in Figure, A) until RIU decommission on June 23, 2020 (letter H in Figure, A). Entries then increased by an average of 0.150 entries per room per day (95% CI, 0.097-0.202; P < .001), reaching 52.5 daily entries on August 10, 2020. This equated to 61.3% of prepandemic levels (95% CI, 61.3%-61.6%; P < .001) (Figure, A).

During the prepandemic period, Unit B averaged 63 daily occupied rooms, and these rooms averaged 76.9 entries per room per day, with no statistically significant change over time. During the pandemic period, these units averaged 64 daily occupied rooms, and these rooms averaged 72.4 entries per room per day. Briefly, at the start of the pandemic, daily entries per room decreased by 11.8 (95% CI, 11.6-12), equating to a 14.7% reduction from baseline (95% CI, 14.4%-14.9%; P < .001). Entries then increased by an average of 0.052 entries per room per day (95% CI, –0.01 to 0.115; P = .051), stabilizing in early August 2020 at an average of 74.1 daily entries. This equated to 92.2% of prepandemic levels (95% CI, 92%-92.3%; P < .001) (Figure, B).

Unit A/RIU experienced significantly greater average daily room entries during the prepandemic period (P < .001) and significantly fewer average daily room entries during the pandemic period (P < .001) than unit B. Although unit A and unit B cared for similar patient populations prior to the pandemic, unit B was located in a different building from the resident work room. This likely resulted in batched visits to patients, leading to fewer total room entries per day.

This is the first study to measure 24-hour patient room– entries as an objective proxy for physical distancing during the pandemic. Unit A/RIU saw an initial 60.6% decrease in room entries. In contrast, unit B, which cared for PUIs, saw a brief 14.7% decrease in room entries before returning to baseline. In all units, room entries increased over time, although this increase was greater in unit B.

Despite the institutional recommendation of physical distancing, only unit A/RIU saw a large and sustained decrease in room entries. The presence of patients with COVID-19 within this unit likely reminded clinicians of the ongoing need to physically distance. Clinicians may have been fearful of contracting COVID-19 and therefore more stringently followed physical-distancing guidance.

Changes in unit A/RIU room entries tracked with RIU conversion and decommission timeline (letters F and H in Figure, A, respectively) rather than statewide stay-at-home orders (letters E and G in Figure, A). Caring for patients with COVID-19 within the unit might have influenced clinician physical distancing more than state policy. Correspondingly, as the number of hospitalized patients with COVID-19 decreased, room entries trended toward baseline. The difficulty of sustaining behavioral changes has been demonstrated in healthcare settings, including at our own institution.^6-8 This gradual extinction in physical distancing could be due to several factors, such as fewer patients with COVID-19 or staff fatigue. Physical distancing may have been more extreme and suboptimal for care at the beginning of the pandemic owing to uncertainty or fear.

This work has implications for how to monitor physical distancing in healthcare facilities. Our study shows that behaviors can change rapidly, but sustaining change is difficult. This suggests the need for regular reinforcement of physical distancing with all staff. Additionally, cohorting patients on RIUs may result in greater physical distancing. It also highlights that PUIs serve as less of a cue to promote physical distancing, possibly due to increased confidence in and availability of COVID-19 tests and/or precautions fatigue.⁹ Objective room-entry monitoring systems, such as the one used in this study, can provide hospital leaders with crucial real-time feedback to monitor physical distancing practices and determine when and where re-education may be needed.

This study was conducted at a single urban, academic medical center, limiting its generalizability. Many other hospital policies implemented at the beginning of the pandemic may have influenced our results. We are unable to examine the type of clinician entering each room and for how long as well as entries in workrooms and breakrooms. Clinicians were not given real-time or retrospective feedback on room entries during the pandemic period. These data would be important to understand staff responses to physical distancing. Finally, while clinicians responded differently as the pandemic progressed and depending on which unit they were in, the ideal degree of physical distancing remains unknown. Although minimizing patient contact limits nosocomial viral spread, too little contact can also cause harm.

At the onset of the COVID-19 pandemic, 24-hour patient room–entries fell significantly in all units before increasing. This decrease was more pronounced in unit A/RIU. As the pandemic continues, hospitals could consider utilizing novel room-entry monitoring systems to guide physical-distancing implementation and staff education.

The authors thank Vera Chu for her support with data requests.

The COVID-19 pandemic dramatically altered how healthcare providers care for hospitalized patients. Many hospitals provided physical-distancing guidance to minimize viral transmission and preserve personal protective equipment. This guidance informed clinician behavior on rounds and in workspaces.¹ One study reported that clinicians maintained distance from patients by grouping medical interventions, utilizing telemedicine for rounding and consultations, and implementing respiratory isolation units (RIUs) to cohort patients with COVID-19.²

Although physical distancing is recommended during inpatient care, no study to date has used objective measures to quantify the degree to which clinical practice was influenced. We aimed to objectively quantify changes in 24-hour patient room–entries before and during the COVID-19 pandemic using data from existing heat sensors to assess differences in physical distancing in RIUs and general medicine units.

Study Design

A single-institution study was conducted at the University of Chicago Medicine, Illinois. Room entries were compared between a general medicine unit that transitioned to an RIU (unit A/RIU) and four general medicine units (unit B) using 24-hour patient room–entry data. Unit A was commissioned as an RIU to care exclusively for patients with confirmed COVID-19 on March 25, 2020, and decommissioned on June 23, 2020. Unit B cared for patients under investigation (PUIs) for COVID-19 and patients admitted for other reasons. PUIs were transferred to the RIU if positive for COVID-19. Hospital visitor restrictions were implemented on March 14, 2020, and lifted on June 29, 2020. The University of Chicago Institutional Review Board granted this project an exempt determination.

Data Collection

From January 1, 2020, to August 10, 2020, room-entry data were collected using the PURELL SMARTLINK hand-hygiene system (GOJO Industries, Inc.). This hand-hygiene compliance system tracks unit-level sanitizer dispenses and total room entries and exits via body heat sensors. Similar to our prior studies, this study extracted heat-sensor data to monitor room entries.^3,4

Data Analysis

Objective 24-hour room-entry data were analyzed for all units. Rooms with less than two daily entries were assumed to be unoccupied and excluded from the analysis. Hospital-wide physical-distancing guidance published on March 10, 2020, was used to delineate “prepandemic” and “pandemic” periods. Each department adopted these recommendations (eg, physical distancing, conducting prerounds virtually, limiting the number of people seeing patients, using iPads for virtual patient visits) as appropriate.

Interrupted time series analyses were used to examine room-entry changes before and during the pandemic. The segmented function in R 4.0.2 (R Core Team) was used to create a model and estimate final fitting parameters, uncertainties, and data breakpoints using a bootstrap restarting algorithm.⁵ The Davies test was used to determine statistical significance of breakpoints, which was defined as P < .05.

We examined data from January 1, 2020, to August 10, 2020, from 3283 patients who collectively experienced 655,615 room entries. Unit A/RIU cared for 395 patients during the prepandemic period and 542 patients during the pandemic period. Compared with patients from the prepandemic period, patients during the pandemic period were more likely to be Black (73.7% vs 77.9%) and less likely to be White (17.0% vs 8.7%) (P = .002); were less likely to have respiratory (19.8% vs 8.0%, P < .001) or gastrointestinal (12.4% vs 6.5%, P = .008) primary diagnoses; and had a higher mean case-mix index (1.74 vs 2.38, P < .001) (Table).

Unit B had 718 patients during the prepandemic period and 1628 patients during the pandemic period. Compared with patients from the prepandemic period, patients during the pandemic period were less likely to be female (58.8% vs 53.4%, P = .018); less likely to be Black (77.7% vs 74.7%) and Hispanic (5.7% vs 3.4%) (P < .001); more likely to have circulatory (10.1% vs 14.8%, P = .009) primary diagnoses; less likely to have respiratory (14.4% vs 6.8%, P < .001) primary diagnoses; and had a higher mean case-mix index (1.58 vs 1.82, P = .014) (Table).

During the prepandemic period, unit A/RIU averaged 27 occupied rooms per day. These rooms averaged 85.0 entries per room per day, with no statistically significant change over time. During the pandemic period, this unit averaged 24 daily occupied rooms, and these rooms averaged 44.4 entries per room per day. At the start of the pandemic, daily entries per room decreased by 51.9 (95% CI, 51.1-52.7). This equated to a 60.6% reduction from baseline (95% CI, 59.6%-61.5%; P < .001), with the lowest average occurring after RIU conversion on March 25, 2020 (letter F in Figure, A). Entries remained constant through the end of statewide stay-at-home orders (letter G in Figure, A) until RIU decommission on June 23, 2020 (letter H in Figure, A). Entries then increased by an average of 0.150 entries per room per day (95% CI, 0.097-0.202; P < .001), reaching 52.5 daily entries on August 10, 2020. This equated to 61.3% of prepandemic levels (95% CI, 61.3%-61.6%; P < .001) (Figure, A).

During the prepandemic period, Unit B averaged 63 daily occupied rooms, and these rooms averaged 76.9 entries per room per day, with no statistically significant change over time. During the pandemic period, these units averaged 64 daily occupied rooms, and these rooms averaged 72.4 entries per room per day. Briefly, at the start of the pandemic, daily entries per room decreased by 11.8 (95% CI, 11.6-12), equating to a 14.7% reduction from baseline (95% CI, 14.4%-14.9%; P < .001). Entries then increased by an average of 0.052 entries per room per day (95% CI, –0.01 to 0.115; P = .051), stabilizing in early August 2020 at an average of 74.1 daily entries. This equated to 92.2% of prepandemic levels (95% CI, 92%-92.3%; P < .001) (Figure, B).

Unit A/RIU experienced significantly greater average daily room entries during the prepandemic period (P < .001) and significantly fewer average daily room entries during the pandemic period (P < .001) than unit B. Although unit A and unit B cared for similar patient populations prior to the pandemic, unit B was located in a different building from the resident work room. This likely resulted in batched visits to patients, leading to fewer total room entries per day.

This is the first study to measure 24-hour patient room– entries as an objective proxy for physical distancing during the pandemic. Unit A/RIU saw an initial 60.6% decrease in room entries. In contrast, unit B, which cared for PUIs, saw a brief 14.7% decrease in room entries before returning to baseline. In all units, room entries increased over time, although this increase was greater in unit B.

Despite the institutional recommendation of physical distancing, only unit A/RIU saw a large and sustained decrease in room entries. The presence of patients with COVID-19 within this unit likely reminded clinicians of the ongoing need to physically distance. Clinicians may have been fearful of contracting COVID-19 and therefore more stringently followed physical-distancing guidance.

Changes in unit A/RIU room entries tracked with RIU conversion and decommission timeline (letters F and H in Figure, A, respectively) rather than statewide stay-at-home orders (letters E and G in Figure, A). Caring for patients with COVID-19 within the unit might have influenced clinician physical distancing more than state policy. Correspondingly, as the number of hospitalized patients with COVID-19 decreased, room entries trended toward baseline. The difficulty of sustaining behavioral changes has been demonstrated in healthcare settings, including at our own institution.^6-8 This gradual extinction in physical distancing could be due to several factors, such as fewer patients with COVID-19 or staff fatigue. Physical distancing may have been more extreme and suboptimal for care at the beginning of the pandemic owing to uncertainty or fear.

This work has implications for how to monitor physical distancing in healthcare facilities. Our study shows that behaviors can change rapidly, but sustaining change is difficult. This suggests the need for regular reinforcement of physical distancing with all staff. Additionally, cohorting patients on RIUs may result in greater physical distancing. It also highlights that PUIs serve as less of a cue to promote physical distancing, possibly due to increased confidence in and availability of COVID-19 tests and/or precautions fatigue.⁹ Objective room-entry monitoring systems, such as the one used in this study, can provide hospital leaders with crucial real-time feedback to monitor physical distancing practices and determine when and where re-education may be needed.

This study was conducted at a single urban, academic medical center, limiting its generalizability. Many other hospital policies implemented at the beginning of the pandemic may have influenced our results. We are unable to examine the type of clinician entering each room and for how long as well as entries in workrooms and breakrooms. Clinicians were not given real-time or retrospective feedback on room entries during the pandemic period. These data would be important to understand staff responses to physical distancing. Finally, while clinicians responded differently as the pandemic progressed and depending on which unit they were in, the ideal degree of physical distancing remains unknown. Although minimizing patient contact limits nosocomial viral spread, too little contact can also cause harm.

At the onset of the COVID-19 pandemic, 24-hour patient room–entries fell significantly in all units before increasing. This decrease was more pronounced in unit A/RIU. As the pandemic continues, hospitals could consider utilizing novel room-entry monitoring systems to guide physical-distancing implementation and staff education.

The authors thank Vera Chu for her support with data requests.

Type 1 diabetes mellitus (T1D) is a common chronic condition of childhood. Its incidence has risen steadily over the past two decades.¹ Treatment requires complex daily tasks, including blood glucose monitoring and insulin administration. The potential long-term complications of T1D—kidney failure, retinopathy, cardiovascular disease, and death—are grave but can be attenuated with effective glycemic management.² Still, less than 25% of youths aged 13 to 19 years achieve target hemoglobin A1c (HbA1c) levels.³

Diabetic ketoacidosis (DKA) is an acute, life-threatening complication of T1D associated with suboptimal glycemic management. In the United States, DKA hospitalizations increased during 2009-2014.⁴ One study found the average standardized cost of a DKA-related pediatric hospitalization exceeded $7000,⁵ an amount not including missed school days, parent/caregiver workdays, and social and family disruptions. Costs extend to psychological well-being, with patients reporting that they “keep thinking [DKA] may happen again.”⁶

The burden of T1D-related morbidity is not equitably distributed. Recent evidence suggests that racial/ethnic minorities and those with public insurance experience disproportionately high rates of DKA.^7,8 Neighborhood socioeconomic measures, like poverty, add an important dimension illustrative of ecological conditions in which patients reside. These measures have been studied in relation to hospitalization rates for chronic conditions like asthma and heart failure.^9,10 For example, one recent study found links between area-level poverty in the United States and the likelihood of readmissions for pediatric patients following an admission for DKA.¹¹ We sought to build on this finding by investigating whether area-level poverty, which we measured as the census tract poverty rate, patient race, and insurance status were associated with the likelihood of initial DKA hospitalization and severity of DKA presentation.

Design

We conducted a single-center, retrospective cohort study that examined data on youth with T1D extracted from the Cincinnati Children’s Hospital Medical Center (CCHMC) electronic medical record (EMR). CCHMC is an urban, tertiary-care, freestanding pediatric hospital, with near-complete market penetration in Hamilton County, Ohio, particularly for subspecialty care. The racial and ethnic breakdown of Hamilton County’s general population is 68% White, 27% Black/African American, 3% Asian, and 4% Hispanic or Latino (adds to more than 100% due to rounding).¹² The CCHMC Institutional Review Board approved this study.

Study Population

We identified eligible patients through a clinical registry within the EMR, which includes all T1D patients seen at CCHMC within the preceding 24 months. Patients are added to this registry upon initial T1D diagnosis by a research nurse who evaluates autoantibody status and clinical presentation. All patients ≤18 years old who had T1D, were active on the registry, and had an address within Hamilton County as of December 31, 2017, were included. Those with type 2 diabetes mellitus or secondary causes of diabetes were excluded. We captured patients on the registry as of December 31, 2017, and then examined whether they had been hospitalized from January 1, 2011, to December 31, 2017.

Covariates

We extracted demographic and clinical data for patients within the T1D registry as of December 31, 2017, and, for those hospitalized, at the time of their admission. Specifically, we extracted date of birth (to calculate age at encounter), sex, and ethnicity. Age was treated as a continuous variable. We defined ethnicity as Latinx (inclusive of Hispanic) or non-Latinx. We calculated duration of T1D diagnosis for patients within the registry by taking the difference of the diabetes onset date and the end of the study period. We extracted whether registry patients had an insulin pump or a continuous glucose monitor (CGM). Hospitalized patients were identified as new-onset or established T1D by comparing their hospitalization with the T1D diagnosis onset date.

For the entire T1D registry, we also assessed a key clinical characteristic: a patient’s median HbA1c. We enumerated the median HbA1c of all patients’ medians within the calendar year 2017, consistent with the methodology of a large international T1D registry.¹³ Among the subset of patients who were hospitalized, we obtained the HbA1c value during or prior to hospitalization.

Exposures

We captured exposure variables from EMR documentation of patients’ address, race, and insurance status. Given the small regional population of Latinx patients, we did not include ethnicity as an exposure. We used the address and insurance status at the beginning of the study period for established T1D patients, or at the time of diagnosis for new-onset patients, for analyses related to the likelihood of DKA admission. We used the address at the time of hospitalization for analyses of DKA presentation severity. We geocoded home addresses using ArcGIS software (Version 10.5.1; Esri), successfully geocoding >95% to the street level and assigning patients to a corresponding census tract—a census-defined geography consisting of approximately 4000 people.¹⁴ Census tracts, when drawn after the decennial (every 10 years) census, are designed to be sociodemographically homogenous¹⁵ and have long been used in medical and public health research.¹⁶ We connected geocoded addresses to the census tract poverty rate, defined as the percentage of individuals within a tract living below the federal poverty level. For our analyses, we primarily treated census tract poverty as a continuous variable ranging from 0% to 100%. We visualized the distribution of census tract poverty across Hamilton County using a choropleth map (Figure, part A). We also grouped the registry population into quartiles, ordered by census tract poverty levels, to ease the graphical display for Figure, part B. We defined race as White, Black (inclusive of African American), or other. We categorized insurance as Medicaid/public or Non-Medicaid/private.

Outcomes

The primary outcome was a hospitalization for DKA during the study period. We identified eligible events by admission diagnosis of DKA (International Classification of Disease, 10th Revision, Clinical Modification, E13.10 or E10.10). Secondarily, we assessed admission severity for each patient’s first DKA hospitalization using initial pH from a venous blood gas and initial bicarbonate from the basic metabolic panel. These measures were chosen in accordance with the American Diabetes Association’s published classification of DKA severity, which uses pH, bicarbonate, and mental status.¹⁷ We defined the time for correction as the minutes from admission time until the anion gap was ≤12 or the bicarbonate was ≥18. We calculated total inpatient bed-days and pediatric intensive care unit (PICU) bed-days, and identified whether a patient was readmitted for DKA during the study period.

Statistical Analyses

We used descriptive statistics to characterize demographic and clinical information for patients within the T1D registry and for the hospitalized subset. We examined DKA severity data for all admitted patients by diagnosis status (new-onset T1D vs established). Associations between categorical outcomes were assessed using the Chi-square or Fisher exact test, and continuous outcomes were evaluated using the Mann-Whitney Rank Sum test.

We individually examined the effects of census tract poverty, race, and insurance on the odds of hospitalization through logistic regression. After confirming overlap and variability in census tract poverty rates by race and admission status (Appendix Table 1), we included these exposures in a multivariable model to capture their independent effects. Because the impact of census tract poverty, race, and insurance on hospitalization odds could vary depending on T1D diagnosis status, we conducted a sensitivity analysis by repeating the multivariable model with only established T1D patients.

Given that median HbA1c levels differed for T1D registry patients based on census tract poverty, race, and insurance, we used linear regression in a post-hoc analysis to evaluate associations between those exposures and median HbA1c. Age, gender, ethnicity, and duration of diabetes were not included in the models given that these variables did not differ by outcome variables. Diabetes technology—CGM and insulin pumps—was not included as a confounder because we hypothesized it mediated the effect of area-level poverty. For analyses, SigmaPlot (Version 14.0; SPSS Inc.) and R statistical software were used (The R Project for Statistical Computing).

A total of 439 patients were in the T1D registry and lived in Hamilton County as of December 31, 2017. Their demographic characteristics are listed in Table 1. Their median age was 14 years; 48% were female. The median poverty rate of the census tracts in which these youth lived was 11%. A total of 24.6% of patients identified as Black, 73.1% as White, and 2.3% as Hispanic; 35.8% were publicly insured. Patients had a median duration of T1D of 61 months (5 years, 1 month), with a median HbA1c (of all patients’ medians) of 8.7%. A total of 58.1% had a CGM, and 56% used an insulin pump.

Hospitalization Characteristics of Those in the T1D Registry

Approximately one-third of registry patients (n = 152) experienced ≥1 hospitalization for DKA during the study period (inclusive of new-onset and established diagnosis). Age, gender, ethnicity, and duration of T1D diagnosis were similar between those who experienced a hospitalization and those who did not. The median census tract poverty rate was higher among those who were hospitalized compared to those who were not (15% vs 8%, P < .01). Among hospitalized patients, 41% identified as Black; among all who were not hospitalized, 16% identified as Black (P < .01). Among all hospitalized patients, 56.6% were covered by public insurance, while 24.7% of non-hospitalized patients were covered by public insurance (P < .01). Hospitalized patients were more likely to have a higher median HbA1c (9.5% vs 8.4%, P < .01) and were significantly less likely to have a CGM or insulin pump (both P < .01). These associations were magnified among the roughly 10% of the T1D registry population (n = 42) who experienced ≥2 hospitalizations.

Figure, part A is a choropleth map depicting the distribution of poverty by census tract across Hamilton County. The map is complemented by part B of the Figure, which displays the T1D registry population split into poverty quartiles and each quartile’s DKA admission rate per T1D registry population. As census tract poverty increases, so too does the rate of DKA hospitalizations. There is an almost threefold difference between the hospitalization rate of the highest poverty quartile and that of the lowest.

Likelihood of Hospitalization

We examined the exposures of interest—census tract poverty, race, and insurance—and odds of hospitalization at a patient level using logistic regression models (Table 2). In the multivariable model that included these exposures, we found that for every 10% increase in the census tract poverty rate, the odds of being hospitalized for DKA increased by 22% (95% CI, 1.03-1.47). Race was not significantly associated with odds of DKA hospitalization in the multivariable model. Public insurance was associated with a 2.71-times higher odds of hospitalization compared to those with private insurance (95% CI, 1.62-4.55). Our sensitivity analysis of only those with established T1D diagnoses showed similar results to that of the entire T1D registry population.

Severity of DKA Presentation

A total of 152 registry patients were admitted; 89 patients had new-onset T1D, and 63 had an established T1D diagnosis. We examined presenting factors (initial pH, initial bicarbonate, HbA1c) and clinical characteristics (time to correction, inpatient bed-days, PICU bed-days) by census tract poverty, race, and insurance for all hospitalized patients. There were no significant differences in presenting factors or outcomes by census tract poverty or insurance. However, we did note certain differences by race. Indeed, though initial pH, bicarbonate, and time to correction did not meaningfully differ by race, Black patients had a longer length of stay than their White peers (Appendix Table 2; 3.06 vs 2.16 days, P < .01). We further examined this by diagnosis status, as newly diagnosed patients often stay longer for diabetes education.¹⁸ Black patients had a longer length of stay than White patients, regardless of whether they had new-onset (3.85 vs 2.99 days, P < .01) or established T1D (1.83 vs 0.97 days, P < .01). Additionally, we found that the HbA1c was significantly higher in Black versus White patients (12.4% vs 10.8%, P = .01). HbA1c was significantly different between established T1D patients identifying as Black and those identifying as White (11.7% vs 9.5%, P < .01); however, HbA1c did not differ by race for newly diagnosed patients (12.5% vs 13.0%, P = .63).

Median HbA1c of All T1D Registry Patients

Given the suggestion of HbA1c variability by exposure variables, we performed a post-hoc linear regression analysis. Table 3 displays the linear regression models assessing associations of census tract poverty, race, and insurance on median HbA1c for all T1D registry patients. In the multivariable model, census tract poverty was no longer associated with higher HbA1c levels (0.13, 95% CI, –0.01 to 0.26). However, Black patients continued to have significantly higher HbA1c levels than White patients (1.09% higher; 95% CI, 0.59-1.59). Also, those with public insurance had significantly higher HbA1c levels than those with private insurance (0.93% higher; 95% CI, 0.51-1.35).

Living in high poverty areas, identifying as Black, and having public insurance were each associated with a higher likelihood of hospitalization for DKA in an unadjusted model. When we examined these exposures together in a multivariable model, census tract poverty and insurance remained associated with a higher likelihood of hospitalization for DKA. Race—with a wide confidence interval—was no longer associated with a higher likelihood of hospitalization. The increased likelihood of hospitalization initially seen for Black patients may be because they are more likely to reside in high poverty areas and be on public insurance—inequities associated with structural racism.

These results mirror those of other studies that have found area-level poverty to be associated with higher rates of intensive care admissions and hospitalizations for conditions like asthma.^10,19 Our findings are also similar to those of a previous study that found that area-level deprivation was associated with a greater likelihood of DKA readmissions in pediatric and adult patients, as well as an adult study that showed increased neighborhood deprivation was associated with higher odds of DKA hospitalization.^11,20 Further examination of the mechanisms underlying DKA hospitalization is clearly warranted and would be informed by a deeper awareness of the social determinants of health—such as community context and the neighborhood/built environment²¹—and their links to racial inequities.

Neighborhood socioeconomic status, here measured as census tract poverty, is likely linked to hospitalization for DKA through contextual factors like food insecurity, limited access to supportive care (eg, presence of a school nurse who can co-manage T1D with the patient/family and medical team), and adverse environmental exposures.^22-24 To improve the health of patients, prevent hospitalizations, and achieve health equity, we suggest that both individual factors and these contextual factors need to be evaluated and addressed through community-connected interventions.^25,26 For example, interventions that deploy community health workers to navigate the complexities of health systems have been effective in reducing HbA1c levels in adults with type 2 diabetes mellitus²⁷ and are beginning to be studied in pediatrics.²⁸

Black patients had a higher median HbA1c compared to their White counterparts in the T1D registry. One study has previously shown that hemoglobin glycation biologically differed by around 0.5% in Black patients compared to White patients.²⁹ Despite this purported biological difference, a greater difference in HbA1c—nearly 2 points—was seen in the univariate model, illustrating that contextual factors such as socioeconomic status and treatment patterns, like the use of diabetes technology, exceed this biological difference.^30-32 Our multivariable results support the importance of these contextual factors, as the HbA1c difference decreased from 1.9 to 1.09 after census tract poverty and insurance were included.

Such differences by race also emerge from experiences of discrimination and bias. Thus, to improve outcomes and equity, we must understand and address racism and its many ill effects. Different levels of racism, such as structural or personally mediated racism, interact with social determinants of health in various ways.^33,34 For example, redlining—racially discriminatory grading of neighborhoods’ creditworthiness in the 1930s—is an example of structural racism that continues to affect the health of patients today.^35,36 Black families are more likely to live in areas with higher poverty, even after controlling for household income level.³⁷ The interaction between race and poverty is complex given these historical underpinnings of structural racism, as evidenced by race no longer being associated with an increased likelihood of DKA hospitalization in our multivariable analysis. The longer length of stay for Black patients despite similar presenting severity also exemplifies this complexity. For example, longer stays could result from explicit or implicit biases that lead providers to keep patients longer due to preconceived beliefs that the caregiver does not understand diabetes management.³⁸ However, it also could be that the caregiver takes longer to complete diabetes education due to systemic barriers, such as work-related obligations during business hours when diabetes educators are present, or transportation and childcare needs.³⁹ These barriers are often tied to socioeconomic status, making it more difficult for those in poverty to complete required education. We cannot conclude from our study the reasons for the difference in length of stay, just that differences exist. Moving forward, we suggest that design of interventions aimed at achieving health equity consider the complex interplay between health, healthcare systems, racism, and neighborhood context.

There were limitations to our study. First, given that this was an observational study, unmeasured confounders like patient or parental education and personal income could affect the strength of associations seen in our models. Second, although the CCHMC’s Diabetes Center treats nearly all children with T1D in Hamilton County, some patients, or hospitalizations at other centers, could have been missed. Third, the demographics of Hamilton County may not generalize to other diabetes centers. Fourth, although more granular data—such as census block groups or blocks—are available from the US Census Bureau, we chose census tracts as our unit of analysis to reduce sampling error that emerges from the use of such smaller geographies. Last, we measured census tract poverty rather than other contextual markers of one’s neighborhood, like vacant housing or the proportion of those receiving public benefits. However, these measures are often highly correlated with census tract poverty.⁴⁰

Type 1 diabetes mellitus (T1D) is a common chronic condition of childhood. Its incidence has risen steadily over the past two decades.¹ Treatment requires complex daily tasks, including blood glucose monitoring and insulin administration. The potential long-term complications of T1D—kidney failure, retinopathy, cardiovascular disease, and death—are grave but can be attenuated with effective glycemic management.² Still, less than 25% of youths aged 13 to 19 years achieve target hemoglobin A1c (HbA1c) levels.³

Diabetic ketoacidosis (DKA) is an acute, life-threatening complication of T1D associated with suboptimal glycemic management. In the United States, DKA hospitalizations increased during 2009-2014.⁴ One study found the average standardized cost of a DKA-related pediatric hospitalization exceeded $7000,⁵ an amount not including missed school days, parent/caregiver workdays, and social and family disruptions. Costs extend to psychological well-being, with patients reporting that they “keep thinking [DKA] may happen again.”⁶

The burden of T1D-related morbidity is not equitably distributed. Recent evidence suggests that racial/ethnic minorities and those with public insurance experience disproportionately high rates of DKA.^7,8 Neighborhood socioeconomic measures, like poverty, add an important dimension illustrative of ecological conditions in which patients reside. These measures have been studied in relation to hospitalization rates for chronic conditions like asthma and heart failure.^9,10 For example, one recent study found links between area-level poverty in the United States and the likelihood of readmissions for pediatric patients following an admission for DKA.¹¹ We sought to build on this finding by investigating whether area-level poverty, which we measured as the census tract poverty rate, patient race, and insurance status were associated with the likelihood of initial DKA hospitalization and severity of DKA presentation.

Design

We conducted a single-center, retrospective cohort study that examined data on youth with T1D extracted from the Cincinnati Children’s Hospital Medical Center (CCHMC) electronic medical record (EMR). CCHMC is an urban, tertiary-care, freestanding pediatric hospital, with near-complete market penetration in Hamilton County, Ohio, particularly for subspecialty care. The racial and ethnic breakdown of Hamilton County’s general population is 68% White, 27% Black/African American, 3% Asian, and 4% Hispanic or Latino (adds to more than 100% due to rounding).¹² The CCHMC Institutional Review Board approved this study.

Study Population

We identified eligible patients through a clinical registry within the EMR, which includes all T1D patients seen at CCHMC within the preceding 24 months. Patients are added to this registry upon initial T1D diagnosis by a research nurse who evaluates autoantibody status and clinical presentation. All patients ≤18 years old who had T1D, were active on the registry, and had an address within Hamilton County as of December 31, 2017, were included. Those with type 2 diabetes mellitus or secondary causes of diabetes were excluded. We captured patients on the registry as of December 31, 2017, and then examined whether they had been hospitalized from January 1, 2011, to December 31, 2017.

Covariates

We extracted demographic and clinical data for patients within the T1D registry as of December 31, 2017, and, for those hospitalized, at the time of their admission. Specifically, we extracted date of birth (to calculate age at encounter), sex, and ethnicity. Age was treated as a continuous variable. We defined ethnicity as Latinx (inclusive of Hispanic) or non-Latinx. We calculated duration of T1D diagnosis for patients within the registry by taking the difference of the diabetes onset date and the end of the study period. We extracted whether registry patients had an insulin pump or a continuous glucose monitor (CGM). Hospitalized patients were identified as new-onset or established T1D by comparing their hospitalization with the T1D diagnosis onset date.

For the entire T1D registry, we also assessed a key clinical characteristic: a patient’s median HbA1c. We enumerated the median HbA1c of all patients’ medians within the calendar year 2017, consistent with the methodology of a large international T1D registry.¹³ Among the subset of patients who were hospitalized, we obtained the HbA1c value during or prior to hospitalization.

Exposures

We captured exposure variables from EMR documentation of patients’ address, race, and insurance status. Given the small regional population of Latinx patients, we did not include ethnicity as an exposure. We used the address and insurance status at the beginning of the study period for established T1D patients, or at the time of diagnosis for new-onset patients, for analyses related to the likelihood of DKA admission. We used the address at the time of hospitalization for analyses of DKA presentation severity. We geocoded home addresses using ArcGIS software (Version 10.5.1; Esri), successfully geocoding >95% to the street level and assigning patients to a corresponding census tract—a census-defined geography consisting of approximately 4000 people.¹⁴ Census tracts, when drawn after the decennial (every 10 years) census, are designed to be sociodemographically homogenous¹⁵ and have long been used in medical and public health research.¹⁶ We connected geocoded addresses to the census tract poverty rate, defined as the percentage of individuals within a tract living below the federal poverty level. For our analyses, we primarily treated census tract poverty as a continuous variable ranging from 0% to 100%. We visualized the distribution of census tract poverty across Hamilton County using a choropleth map (Figure, part A). We also grouped the registry population into quartiles, ordered by census tract poverty levels, to ease the graphical display for Figure, part B. We defined race as White, Black (inclusive of African American), or other. We categorized insurance as Medicaid/public or Non-Medicaid/private.

Outcomes

The primary outcome was a hospitalization for DKA during the study period. We identified eligible events by admission diagnosis of DKA (International Classification of Disease, 10th Revision, Clinical Modification, E13.10 or E10.10). Secondarily, we assessed admission severity for each patient’s first DKA hospitalization using initial pH from a venous blood gas and initial bicarbonate from the basic metabolic panel. These measures were chosen in accordance with the American Diabetes Association’s published classification of DKA severity, which uses pH, bicarbonate, and mental status.¹⁷ We defined the time for correction as the minutes from admission time until the anion gap was ≤12 or the bicarbonate was ≥18. We calculated total inpatient bed-days and pediatric intensive care unit (PICU) bed-days, and identified whether a patient was readmitted for DKA during the study period.

Statistical Analyses

We used descriptive statistics to characterize demographic and clinical information for patients within the T1D registry and for the hospitalized subset. We examined DKA severity data for all admitted patients by diagnosis status (new-onset T1D vs established). Associations between categorical outcomes were assessed using the Chi-square or Fisher exact test, and continuous outcomes were evaluated using the Mann-Whitney Rank Sum test.

We individually examined the effects of census tract poverty, race, and insurance on the odds of hospitalization through logistic regression. After confirming overlap and variability in census tract poverty rates by race and admission status (Appendix Table 1), we included these exposures in a multivariable model to capture their independent effects. Because the impact of census tract poverty, race, and insurance on hospitalization odds could vary depending on T1D diagnosis status, we conducted a sensitivity analysis by repeating the multivariable model with only established T1D patients.

Given that median HbA1c levels differed for T1D registry patients based on census tract poverty, race, and insurance, we used linear regression in a post-hoc analysis to evaluate associations between those exposures and median HbA1c. Age, gender, ethnicity, and duration of diabetes were not included in the models given that these variables did not differ by outcome variables. Diabetes technology—CGM and insulin pumps—was not included as a confounder because we hypothesized it mediated the effect of area-level poverty. For analyses, SigmaPlot (Version 14.0; SPSS Inc.) and R statistical software were used (The R Project for Statistical Computing).

A total of 439 patients were in the T1D registry and lived in Hamilton County as of December 31, 2017. Their demographic characteristics are listed in Table 1. Their median age was 14 years; 48% were female. The median poverty rate of the census tracts in which these youth lived was 11%. A total of 24.6% of patients identified as Black, 73.1% as White, and 2.3% as Hispanic; 35.8% were publicly insured. Patients had a median duration of T1D of 61 months (5 years, 1 month), with a median HbA1c (of all patients’ medians) of 8.7%. A total of 58.1% had a CGM, and 56% used an insulin pump.

Hospitalization Characteristics of Those in the T1D Registry

Approximately one-third of registry patients (n = 152) experienced ≥1 hospitalization for DKA during the study period (inclusive of new-onset and established diagnosis). Age, gender, ethnicity, and duration of T1D diagnosis were similar between those who experienced a hospitalization and those who did not. The median census tract poverty rate was higher among those who were hospitalized compared to those who were not (15% vs 8%, P < .01). Among hospitalized patients, 41% identified as Black; among all who were not hospitalized, 16% identified as Black (P < .01). Among all hospitalized patients, 56.6% were covered by public insurance, while 24.7% of non-hospitalized patients were covered by public insurance (P < .01). Hospitalized patients were more likely to have a higher median HbA1c (9.5% vs 8.4%, P < .01) and were significantly less likely to have a CGM or insulin pump (both P < .01). These associations were magnified among the roughly 10% of the T1D registry population (n = 42) who experienced ≥2 hospitalizations.

Figure, part A is a choropleth map depicting the distribution of poverty by census tract across Hamilton County. The map is complemented by part B of the Figure, which displays the T1D registry population split into poverty quartiles and each quartile’s DKA admission rate per T1D registry population. As census tract poverty increases, so too does the rate of DKA hospitalizations. There is an almost threefold difference between the hospitalization rate of the highest poverty quartile and that of the lowest.

Likelihood of Hospitalization

We examined the exposures of interest—census tract poverty, race, and insurance—and odds of hospitalization at a patient level using logistic regression models (Table 2). In the multivariable model that included these exposures, we found that for every 10% increase in the census tract poverty rate, the odds of being hospitalized for DKA increased by 22% (95% CI, 1.03-1.47). Race was not significantly associated with odds of DKA hospitalization in the multivariable model. Public insurance was associated with a 2.71-times higher odds of hospitalization compared to those with private insurance (95% CI, 1.62-4.55). Our sensitivity analysis of only those with established T1D diagnoses showed similar results to that of the entire T1D registry population.

Severity of DKA Presentation

A total of 152 registry patients were admitted; 89 patients had new-onset T1D, and 63 had an established T1D diagnosis. We examined presenting factors (initial pH, initial bicarbonate, HbA1c) and clinical characteristics (time to correction, inpatient bed-days, PICU bed-days) by census tract poverty, race, and insurance for all hospitalized patients. There were no significant differences in presenting factors or outcomes by census tract poverty or insurance. However, we did note certain differences by race. Indeed, though initial pH, bicarbonate, and time to correction did not meaningfully differ by race, Black patients had a longer length of stay than their White peers (Appendix Table 2; 3.06 vs 2.16 days, P < .01). We further examined this by diagnosis status, as newly diagnosed patients often stay longer for diabetes education.¹⁸ Black patients had a longer length of stay than White patients, regardless of whether they had new-onset (3.85 vs 2.99 days, P < .01) or established T1D (1.83 vs 0.97 days, P < .01). Additionally, we found that the HbA1c was significantly higher in Black versus White patients (12.4% vs 10.8%, P = .01). HbA1c was significantly different between established T1D patients identifying as Black and those identifying as White (11.7% vs 9.5%, P < .01); however, HbA1c did not differ by race for newly diagnosed patients (12.5% vs 13.0%, P = .63).

Median HbA1c of All T1D Registry Patients

Given the suggestion of HbA1c variability by exposure variables, we performed a post-hoc linear regression analysis. Table 3 displays the linear regression models assessing associations of census tract poverty, race, and insurance on median HbA1c for all T1D registry patients. In the multivariable model, census tract poverty was no longer associated with higher HbA1c levels (0.13, 95% CI, –0.01 to 0.26). However, Black patients continued to have significantly higher HbA1c levels than White patients (1.09% higher; 95% CI, 0.59-1.59). Also, those with public insurance had significantly higher HbA1c levels than those with private insurance (0.93% higher; 95% CI, 0.51-1.35).

Living in high poverty areas, identifying as Black, and having public insurance were each associated with a higher likelihood of hospitalization for DKA in an unadjusted model. When we examined these exposures together in a multivariable model, census tract poverty and insurance remained associated with a higher likelihood of hospitalization for DKA. Race—with a wide confidence interval—was no longer associated with a higher likelihood of hospitalization. The increased likelihood of hospitalization initially seen for Black patients may be because they are more likely to reside in high poverty areas and be on public insurance—inequities associated with structural racism.

These results mirror those of other studies that have found area-level poverty to be associated with higher rates of intensive care admissions and hospitalizations for conditions like asthma.^10,19 Our findings are also similar to those of a previous study that found that area-level deprivation was associated with a greater likelihood of DKA readmissions in pediatric and adult patients, as well as an adult study that showed increased neighborhood deprivation was associated with higher odds of DKA hospitalization.^11,20 Further examination of the mechanisms underlying DKA hospitalization is clearly warranted and would be informed by a deeper awareness of the social determinants of health—such as community context and the neighborhood/built environment²¹—and their links to racial inequities.

Neighborhood socioeconomic status, here measured as census tract poverty, is likely linked to hospitalization for DKA through contextual factors like food insecurity, limited access to supportive care (eg, presence of a school nurse who can co-manage T1D with the patient/family and medical team), and adverse environmental exposures.^22-24 To improve the health of patients, prevent hospitalizations, and achieve health equity, we suggest that both individual factors and these contextual factors need to be evaluated and addressed through community-connected interventions.^25,26 For example, interventions that deploy community health workers to navigate the complexities of health systems have been effective in reducing HbA1c levels in adults with type 2 diabetes mellitus²⁷ and are beginning to be studied in pediatrics.²⁸

Black patients had a higher median HbA1c compared to their White counterparts in the T1D registry. One study has previously shown that hemoglobin glycation biologically differed by around 0.5% in Black patients compared to White patients.²⁹ Despite this purported biological difference, a greater difference in HbA1c—nearly 2 points—was seen in the univariate model, illustrating that contextual factors such as socioeconomic status and treatment patterns, like the use of diabetes technology, exceed this biological difference.^30-32 Our multivariable results support the importance of these contextual factors, as the HbA1c difference decreased from 1.9 to 1.09 after census tract poverty and insurance were included.

Such differences by race also emerge from experiences of discrimination and bias. Thus, to improve outcomes and equity, we must understand and address racism and its many ill effects. Different levels of racism, such as structural or personally mediated racism, interact with social determinants of health in various ways.^33,34 For example, redlining—racially discriminatory grading of neighborhoods’ creditworthiness in the 1930s—is an example of structural racism that continues to affect the health of patients today.^35,36 Black families are more likely to live in areas with higher poverty, even after controlling for household income level.³⁷ The interaction between race and poverty is complex given these historical underpinnings of structural racism, as evidenced by race no longer being associated with an increased likelihood of DKA hospitalization in our multivariable analysis. The longer length of stay for Black patients despite similar presenting severity also exemplifies this complexity. For example, longer stays could result from explicit or implicit biases that lead providers to keep patients longer due to preconceived beliefs that the caregiver does not understand diabetes management.³⁸ However, it also could be that the caregiver takes longer to complete diabetes education due to systemic barriers, such as work-related obligations during business hours when diabetes educators are present, or transportation and childcare needs.³⁹ These barriers are often tied to socioeconomic status, making it more difficult for those in poverty to complete required education. We cannot conclude from our study the reasons for the difference in length of stay, just that differences exist. Moving forward, we suggest that design of interventions aimed at achieving health equity consider the complex interplay between health, healthcare systems, racism, and neighborhood context.

There were limitations to our study. First, given that this was an observational study, unmeasured confounders like patient or parental education and personal income could affect the strength of associations seen in our models. Second, although the CCHMC’s Diabetes Center treats nearly all children with T1D in Hamilton County, some patients, or hospitalizations at other centers, could have been missed. Third, the demographics of Hamilton County may not generalize to other diabetes centers. Fourth, although more granular data—such as census block groups or blocks—are available from the US Census Bureau, we chose census tracts as our unit of analysis to reduce sampling error that emerges from the use of such smaller geographies. Last, we measured census tract poverty rather than other contextual markers of one’s neighborhood, like vacant housing or the proportion of those receiving public benefits. However, these measures are often highly correlated with census tract poverty.⁴⁰

Type 1 diabetes mellitus (T1D) is a common chronic condition of childhood. Its incidence has risen steadily over the past two decades.¹ Treatment requires complex daily tasks, including blood glucose monitoring and insulin administration. The potential long-term complications of T1D—kidney failure, retinopathy, cardiovascular disease, and death—are grave but can be attenuated with effective glycemic management.² Still, less than 25% of youths aged 13 to 19 years achieve target hemoglobin A1c (HbA1c) levels.³

Diabetic ketoacidosis (DKA) is an acute, life-threatening complication of T1D associated with suboptimal glycemic management. In the United States, DKA hospitalizations increased during 2009-2014.⁴ One study found the average standardized cost of a DKA-related pediatric hospitalization exceeded $7000,⁵ an amount not including missed school days, parent/caregiver workdays, and social and family disruptions. Costs extend to psychological well-being, with patients reporting that they “keep thinking [DKA] may happen again.”⁶

The burden of T1D-related morbidity is not equitably distributed. Recent evidence suggests that racial/ethnic minorities and those with public insurance experience disproportionately high rates of DKA.^7,8 Neighborhood socioeconomic measures, like poverty, add an important dimension illustrative of ecological conditions in which patients reside. These measures have been studied in relation to hospitalization rates for chronic conditions like asthma and heart failure.^9,10 For example, one recent study found links between area-level poverty in the United States and the likelihood of readmissions for pediatric patients following an admission for DKA.¹¹ We sought to build on this finding by investigating whether area-level poverty, which we measured as the census tract poverty rate, patient race, and insurance status were associated with the likelihood of initial DKA hospitalization and severity of DKA presentation.

Design

We conducted a single-center, retrospective cohort study that examined data on youth with T1D extracted from the Cincinnati Children’s Hospital Medical Center (CCHMC) electronic medical record (EMR). CCHMC is an urban, tertiary-care, freestanding pediatric hospital, with near-complete market penetration in Hamilton County, Ohio, particularly for subspecialty care. The racial and ethnic breakdown of Hamilton County’s general population is 68% White, 27% Black/African American, 3% Asian, and 4% Hispanic or Latino (adds to more than 100% due to rounding).¹² The CCHMC Institutional Review Board approved this study.

Study Population

We identified eligible patients through a clinical registry within the EMR, which includes all T1D patients seen at CCHMC within the preceding 24 months. Patients are added to this registry upon initial T1D diagnosis by a research nurse who evaluates autoantibody status and clinical presentation. All patients ≤18 years old who had T1D, were active on the registry, and had an address within Hamilton County as of December 31, 2017, were included. Those with type 2 diabetes mellitus or secondary causes of diabetes were excluded. We captured patients on the registry as of December 31, 2017, and then examined whether they had been hospitalized from January 1, 2011, to December 31, 2017.

Covariates

We extracted demographic and clinical data for patients within the T1D registry as of December 31, 2017, and, for those hospitalized, at the time of their admission. Specifically, we extracted date of birth (to calculate age at encounter), sex, and ethnicity. Age was treated as a continuous variable. We defined ethnicity as Latinx (inclusive of Hispanic) or non-Latinx. We calculated duration of T1D diagnosis for patients within the registry by taking the difference of the diabetes onset date and the end of the study period. We extracted whether registry patients had an insulin pump or a continuous glucose monitor (CGM). Hospitalized patients were identified as new-onset or established T1D by comparing their hospitalization with the T1D diagnosis onset date.

For the entire T1D registry, we also assessed a key clinical characteristic: a patient’s median HbA1c. We enumerated the median HbA1c of all patients’ medians within the calendar year 2017, consistent with the methodology of a large international T1D registry.¹³ Among the subset of patients who were hospitalized, we obtained the HbA1c value during or prior to hospitalization.

Exposures

We captured exposure variables from EMR documentation of patients’ address, race, and insurance status. Given the small regional population of Latinx patients, we did not include ethnicity as an exposure. We used the address and insurance status at the beginning of the study period for established T1D patients, or at the time of diagnosis for new-onset patients, for analyses related to the likelihood of DKA admission. We used the address at the time of hospitalization for analyses of DKA presentation severity. We geocoded home addresses using ArcGIS software (Version 10.5.1; Esri), successfully geocoding >95% to the street level and assigning patients to a corresponding census tract—a census-defined geography consisting of approximately 4000 people.¹⁴ Census tracts, when drawn after the decennial (every 10 years) census, are designed to be sociodemographically homogenous¹⁵ and have long been used in medical and public health research.¹⁶ We connected geocoded addresses to the census tract poverty rate, defined as the percentage of individuals within a tract living below the federal poverty level. For our analyses, we primarily treated census tract poverty as a continuous variable ranging from 0% to 100%. We visualized the distribution of census tract poverty across Hamilton County using a choropleth map (Figure, part A). We also grouped the registry population into quartiles, ordered by census tract poverty levels, to ease the graphical display for Figure, part B. We defined race as White, Black (inclusive of African American), or other. We categorized insurance as Medicaid/public or Non-Medicaid/private.

Outcomes

The primary outcome was a hospitalization for DKA during the study period. We identified eligible events by admission diagnosis of DKA (International Classification of Disease, 10th Revision, Clinical Modification, E13.10 or E10.10). Secondarily, we assessed admission severity for each patient’s first DKA hospitalization using initial pH from a venous blood gas and initial bicarbonate from the basic metabolic panel. These measures were chosen in accordance with the American Diabetes Association’s published classification of DKA severity, which uses pH, bicarbonate, and mental status.¹⁷ We defined the time for correction as the minutes from admission time until the anion gap was ≤12 or the bicarbonate was ≥18. We calculated total inpatient bed-days and pediatric intensive care unit (PICU) bed-days, and identified whether a patient was readmitted for DKA during the study period.

Statistical Analyses

We used descriptive statistics to characterize demographic and clinical information for patients within the T1D registry and for the hospitalized subset. We examined DKA severity data for all admitted patients by diagnosis status (new-onset T1D vs established). Associations between categorical outcomes were assessed using the Chi-square or Fisher exact test, and continuous outcomes were evaluated using the Mann-Whitney Rank Sum test.

We individually examined the effects of census tract poverty, race, and insurance on the odds of hospitalization through logistic regression. After confirming overlap and variability in census tract poverty rates by race and admission status (Appendix Table 1), we included these exposures in a multivariable model to capture their independent effects. Because the impact of census tract poverty, race, and insurance on hospitalization odds could vary depending on T1D diagnosis status, we conducted a sensitivity analysis by repeating the multivariable model with only established T1D patients.

Given that median HbA1c levels differed for T1D registry patients based on census tract poverty, race, and insurance, we used linear regression in a post-hoc analysis to evaluate associations between those exposures and median HbA1c. Age, gender, ethnicity, and duration of diabetes were not included in the models given that these variables did not differ by outcome variables. Diabetes technology—CGM and insulin pumps—was not included as a confounder because we hypothesized it mediated the effect of area-level poverty. For analyses, SigmaPlot (Version 14.0; SPSS Inc.) and R statistical software were used (The R Project for Statistical Computing).

A total of 439 patients were in the T1D registry and lived in Hamilton County as of December 31, 2017. Their demographic characteristics are listed in Table 1. Their median age was 14 years; 48% were female. The median poverty rate of the census tracts in which these youth lived was 11%. A total of 24.6% of patients identified as Black, 73.1% as White, and 2.3% as Hispanic; 35.8% were publicly insured. Patients had a median duration of T1D of 61 months (5 years, 1 month), with a median HbA1c (of all patients’ medians) of 8.7%. A total of 58.1% had a CGM, and 56% used an insulin pump.

Hospitalization Characteristics of Those in the T1D Registry

Approximately one-third of registry patients (n = 152) experienced ≥1 hospitalization for DKA during the study period (inclusive of new-onset and established diagnosis). Age, gender, ethnicity, and duration of T1D diagnosis were similar between those who experienced a hospitalization and those who did not. The median census tract poverty rate was higher among those who were hospitalized compared to those who were not (15% vs 8%, P < .01). Among hospitalized patients, 41% identified as Black; among all who were not hospitalized, 16% identified as Black (P < .01). Among all hospitalized patients, 56.6% were covered by public insurance, while 24.7% of non-hospitalized patients were covered by public insurance (P < .01). Hospitalized patients were more likely to have a higher median HbA1c (9.5% vs 8.4%, P < .01) and were significantly less likely to have a CGM or insulin pump (both P < .01). These associations were magnified among the roughly 10% of the T1D registry population (n = 42) who experienced ≥2 hospitalizations.

Figure, part A is a choropleth map depicting the distribution of poverty by census tract across Hamilton County. The map is complemented by part B of the Figure, which displays the T1D registry population split into poverty quartiles and each quartile’s DKA admission rate per T1D registry population. As census tract poverty increases, so too does the rate of DKA hospitalizations. There is an almost threefold difference between the hospitalization rate of the highest poverty quartile and that of the lowest.

Likelihood of Hospitalization

We examined the exposures of interest—census tract poverty, race, and insurance—and odds of hospitalization at a patient level using logistic regression models (Table 2). In the multivariable model that included these exposures, we found that for every 10% increase in the census tract poverty rate, the odds of being hospitalized for DKA increased by 22% (95% CI, 1.03-1.47). Race was not significantly associated with odds of DKA hospitalization in the multivariable model. Public insurance was associated with a 2.71-times higher odds of hospitalization compared to those with private insurance (95% CI, 1.62-4.55). Our sensitivity analysis of only those with established T1D diagnoses showed similar results to that of the entire T1D registry population.

Severity of DKA Presentation

A total of 152 registry patients were admitted; 89 patients had new-onset T1D, and 63 had an established T1D diagnosis. We examined presenting factors (initial pH, initial bicarbonate, HbA1c) and clinical characteristics (time to correction, inpatient bed-days, PICU bed-days) by census tract poverty, race, and insurance for all hospitalized patients. There were no significant differences in presenting factors or outcomes by census tract poverty or insurance. However, we did note certain differences by race. Indeed, though initial pH, bicarbonate, and time to correction did not meaningfully differ by race, Black patients had a longer length of stay than their White peers (Appendix Table 2; 3.06 vs 2.16 days, P < .01). We further examined this by diagnosis status, as newly diagnosed patients often stay longer for diabetes education.¹⁸ Black patients had a longer length of stay than White patients, regardless of whether they had new-onset (3.85 vs 2.99 days, P < .01) or established T1D (1.83 vs 0.97 days, P < .01). Additionally, we found that the HbA1c was significantly higher in Black versus White patients (12.4% vs 10.8%, P = .01). HbA1c was significantly different between established T1D patients identifying as Black and those identifying as White (11.7% vs 9.5%, P < .01); however, HbA1c did not differ by race for newly diagnosed patients (12.5% vs 13.0%, P = .63).

Median HbA1c of All T1D Registry Patients

Given the suggestion of HbA1c variability by exposure variables, we performed a post-hoc linear regression analysis. Table 3 displays the linear regression models assessing associations of census tract poverty, race, and insurance on median HbA1c for all T1D registry patients. In the multivariable model, census tract poverty was no longer associated with higher HbA1c levels (0.13, 95% CI, –0.01 to 0.26). However, Black patients continued to have significantly higher HbA1c levels than White patients (1.09% higher; 95% CI, 0.59-1.59). Also, those with public insurance had significantly higher HbA1c levels than those with private insurance (0.93% higher; 95% CI, 0.51-1.35).

Living in high poverty areas, identifying as Black, and having public insurance were each associated with a higher likelihood of hospitalization for DKA in an unadjusted model. When we examined these exposures together in a multivariable model, census tract poverty and insurance remained associated with a higher likelihood of hospitalization for DKA. Race—with a wide confidence interval—was no longer associated with a higher likelihood of hospitalization. The increased likelihood of hospitalization initially seen for Black patients may be because they are more likely to reside in high poverty areas and be on public insurance—inequities associated with structural racism.

These results mirror those of other studies that have found area-level poverty to be associated with higher rates of intensive care admissions and hospitalizations for conditions like asthma.^10,19 Our findings are also similar to those of a previous study that found that area-level deprivation was associated with a greater likelihood of DKA readmissions in pediatric and adult patients, as well as an adult study that showed increased neighborhood deprivation was associated with higher odds of DKA hospitalization.^11,20 Further examination of the mechanisms underlying DKA hospitalization is clearly warranted and would be informed by a deeper awareness of the social determinants of health—such as community context and the neighborhood/built environment²¹—and their links to racial inequities.

Neighborhood socioeconomic status, here measured as census tract poverty, is likely linked to hospitalization for DKA through contextual factors like food insecurity, limited access to supportive care (eg, presence of a school nurse who can co-manage T1D with the patient/family and medical team), and adverse environmental exposures.^22-24 To improve the health of patients, prevent hospitalizations, and achieve health equity, we suggest that both individual factors and these contextual factors need to be evaluated and addressed through community-connected interventions.^25,26 For example, interventions that deploy community health workers to navigate the complexities of health systems have been effective in reducing HbA1c levels in adults with type 2 diabetes mellitus²⁷ and are beginning to be studied in pediatrics.²⁸

Black patients had a higher median HbA1c compared to their White counterparts in the T1D registry. One study has previously shown that hemoglobin glycation biologically differed by around 0.5% in Black patients compared to White patients.²⁹ Despite this purported biological difference, a greater difference in HbA1c—nearly 2 points—was seen in the univariate model, illustrating that contextual factors such as socioeconomic status and treatment patterns, like the use of diabetes technology, exceed this biological difference.^30-32 Our multivariable results support the importance of these contextual factors, as the HbA1c difference decreased from 1.9 to 1.09 after census tract poverty and insurance were included.

Such differences by race also emerge from experiences of discrimination and bias. Thus, to improve outcomes and equity, we must understand and address racism and its many ill effects. Different levels of racism, such as structural or personally mediated racism, interact with social determinants of health in various ways.^33,34 For example, redlining—racially discriminatory grading of neighborhoods’ creditworthiness in the 1930s—is an example of structural racism that continues to affect the health of patients today.^35,36 Black families are more likely to live in areas with higher poverty, even after controlling for household income level.³⁷ The interaction between race and poverty is complex given these historical underpinnings of structural racism, as evidenced by race no longer being associated with an increased likelihood of DKA hospitalization in our multivariable analysis. The longer length of stay for Black patients despite similar presenting severity also exemplifies this complexity. For example, longer stays could result from explicit or implicit biases that lead providers to keep patients longer due to preconceived beliefs that the caregiver does not understand diabetes management.³⁸ However, it also could be that the caregiver takes longer to complete diabetes education due to systemic barriers, such as work-related obligations during business hours when diabetes educators are present, or transportation and childcare needs.³⁹ These barriers are often tied to socioeconomic status, making it more difficult for those in poverty to complete required education. We cannot conclude from our study the reasons for the difference in length of stay, just that differences exist. Moving forward, we suggest that design of interventions aimed at achieving health equity consider the complex interplay between health, healthcare systems, racism, and neighborhood context.

There were limitations to our study. First, given that this was an observational study, unmeasured confounders like patient or parental education and personal income could affect the strength of associations seen in our models. Second, although the CCHMC’s Diabetes Center treats nearly all children with T1D in Hamilton County, some patients, or hospitalizations at other centers, could have been missed. Third, the demographics of Hamilton County may not generalize to other diabetes centers. Fourth, although more granular data—such as census block groups or blocks—are available from the US Census Bureau, we chose census tracts as our unit of analysis to reduce sampling error that emerges from the use of such smaller geographies. Last, we measured census tract poverty rather than other contextual markers of one’s neighborhood, like vacant housing or the proportion of those receiving public benefits. However, these measures are often highly correlated with census tract poverty.⁴⁰