Anxiety over the U.S. economy intensified in June, with 87% of Americans reporting they are anxious or very anxious about inflation, up eight percentage points from the previous month, results from a new national report from the American Psychiatric Association show.

“The economy seems to have supplanted COVID as a major factor in Americans’ day-to-day anxiety,” APA President Rebecca W. Brendel, MD, JD, said in a news release.

“Knowing that so many Americans are concerned about finances is important because it can prepare clinicians to be ready to approach the subject, which is one that people are often reluctant or ashamed to raise on their own,” Dr. Brendel told this news organization.

What’s the best way to bring up the sensitive topic of money?

“In general, it’s best to start with open-ended questions to allow individuals in therapy to share what is on their minds, explore their concerns, and develop strategies to address these issues. Once a patient raises a concern, that is a good time to ask more about the issues they’ve raised and to explore other potential sources of anxiety or stress,” said Dr. Brendel.

The latest APA poll was conducted by Morning Consult, June 18-20, 2022, among a nationally representative sample of 2,210 adults.

In addition to an uptick in worry about inflation, the poll shows that more than half (51%) of adults are worried about a potential loss of income.

Hispanic adults (66%), mothers (65%), millennials (63%), and genZers (62%) are among the groups most likely to be concerned about income loss.

“Stress is not good for health, mental or physical. So, while it’s a reality that Americans are faced with finding ways of making ends meet, it’s more important than ever to make sure that we are all accessing the care that we need,” said Dr. Brendel.

“People should be aware that there may be low- or no-cost options such as community mental health centers or employer-sponsored resources to address mental health concerns,” she added.
 

Coping with traumatic events

The latest poll also shows that about one-third of adults are worried about gun violence (35% overall, 47% among genZers) or a natural disaster (29%) personally affecting them.

Climate change anxiety is also up slightly in June, compared with May (+4%).

The same goes for mid-term election-related anxiety (+3%) – particularly among Democrats (54% vs. 59%) compared with Republicans (48% vs. 48%).

The latest poll provides insight how Americans would cope after a traumatic event. More adults report they will turn to family and friends for support (60%) than practice self-care (42%), speak openly about their feelings (37%), or seek help from a professional (31%). Nearly one-third (30%) say they will move on from it and not dwell on their feelings.

GenZers are the least likely to say they will speak openly about their feelings (29%) and are less likely than millennials to say they would speak to a health professional (28% vs. 38%).

“While many people show resilience, it’s troubling that most Americans wouldn’t speak openly about their feelings after a traumatic event,” APA CEO and Medical Director Saul Levin, MD, said in the news release.

“In many ways, naming feelings is the most important step toward healing, and this reluctance to air our thoughts may indicate that mental health stigma is still a powerful force in our society,” Dr. Levin said.

After a traumatic current event, 41% of Americans say they consume more news and 30% say they take in more social media, but the majority say this does not impact their mental health, the poll shows.

Two in five adults (43%) say the news of a traumatic event makes them feel more informed, 32% say it makes them feel more anxious, and about one-quarter say it makes them feel overwhelmed (27%) or discouraged (24%).

Dr. Brendel noted that, after a traumatic event, “it’s expected that people may experience anxiety or other symptoms for brief periods of time. However, no two people experience things the same way. If symptoms don’t go away, are overwhelming, or get worse over time, for example, it’s critical to seek help right away.”

The June poll shows that 50% of Americans are anxious about the future of reproductive rights but the poll was conducted before the Dobbs ruling.

Anxiety around COVID-19 continues to ease, with about 47% of Americans saying they are concerned about the pandemic, down 2% among all Americans and 16% among Black Americans since May.

The APA’s Healthy Minds Monthly tracks timely mental health issues throughout the year. The APA also releases its annual Healthy Minds Poll each May in conjunction with Mental Health Awareness Month.

A version of this article first appeared on Medscape.com.

Anxiety over the U.S. economy intensified in June, with 87% of Americans reporting they are anxious or very anxious about inflation, up eight percentage points from the previous month, results from a new national report from the American Psychiatric Association show.

“The economy seems to have supplanted COVID as a major factor in Americans’ day-to-day anxiety,” APA President Rebecca W. Brendel, MD, JD, said in a news release.

“Knowing that so many Americans are concerned about finances is important because it can prepare clinicians to be ready to approach the subject, which is one that people are often reluctant or ashamed to raise on their own,” Dr. Brendel told this news organization.

What’s the best way to bring up the sensitive topic of money?

“In general, it’s best to start with open-ended questions to allow individuals in therapy to share what is on their minds, explore their concerns, and develop strategies to address these issues. Once a patient raises a concern, that is a good time to ask more about the issues they’ve raised and to explore other potential sources of anxiety or stress,” said Dr. Brendel.

The latest APA poll was conducted by Morning Consult, June 18-20, 2022, among a nationally representative sample of 2,210 adults.

In addition to an uptick in worry about inflation, the poll shows that more than half (51%) of adults are worried about a potential loss of income.

Hispanic adults (66%), mothers (65%), millennials (63%), and genZers (62%) are among the groups most likely to be concerned about income loss.

“Stress is not good for health, mental or physical. So, while it’s a reality that Americans are faced with finding ways of making ends meet, it’s more important than ever to make sure that we are all accessing the care that we need,” said Dr. Brendel.

“People should be aware that there may be low- or no-cost options such as community mental health centers or employer-sponsored resources to address mental health concerns,” she added.
 

Coping with traumatic events

The latest poll also shows that about one-third of adults are worried about gun violence (35% overall, 47% among genZers) or a natural disaster (29%) personally affecting them.

Climate change anxiety is also up slightly in June, compared with May (+4%).

The same goes for mid-term election-related anxiety (+3%) – particularly among Democrats (54% vs. 59%) compared with Republicans (48% vs. 48%).

The latest poll provides insight how Americans would cope after a traumatic event. More adults report they will turn to family and friends for support (60%) than practice self-care (42%), speak openly about their feelings (37%), or seek help from a professional (31%). Nearly one-third (30%) say they will move on from it and not dwell on their feelings.

GenZers are the least likely to say they will speak openly about their feelings (29%) and are less likely than millennials to say they would speak to a health professional (28% vs. 38%).

“While many people show resilience, it’s troubling that most Americans wouldn’t speak openly about their feelings after a traumatic event,” APA CEO and Medical Director Saul Levin, MD, said in the news release.

“In many ways, naming feelings is the most important step toward healing, and this reluctance to air our thoughts may indicate that mental health stigma is still a powerful force in our society,” Dr. Levin said.

After a traumatic current event, 41% of Americans say they consume more news and 30% say they take in more social media, but the majority say this does not impact their mental health, the poll shows.

Two in five adults (43%) say the news of a traumatic event makes them feel more informed, 32% say it makes them feel more anxious, and about one-quarter say it makes them feel overwhelmed (27%) or discouraged (24%).

Dr. Brendel noted that, after a traumatic event, “it’s expected that people may experience anxiety or other symptoms for brief periods of time. However, no two people experience things the same way. If symptoms don’t go away, are overwhelming, or get worse over time, for example, it’s critical to seek help right away.”

The June poll shows that 50% of Americans are anxious about the future of reproductive rights but the poll was conducted before the Dobbs ruling.

Anxiety around COVID-19 continues to ease, with about 47% of Americans saying they are concerned about the pandemic, down 2% among all Americans and 16% among Black Americans since May.

The APA’s Healthy Minds Monthly tracks timely mental health issues throughout the year. The APA also releases its annual Healthy Minds Poll each May in conjunction with Mental Health Awareness Month.

A version of this article first appeared on Medscape.com.

Anxiety over the U.S. economy intensified in June, with 87% of Americans reporting they are anxious or very anxious about inflation, up eight percentage points from the previous month, results from a new national report from the American Psychiatric Association show.

“The economy seems to have supplanted COVID as a major factor in Americans’ day-to-day anxiety,” APA President Rebecca W. Brendel, MD, JD, said in a news release.

“Knowing that so many Americans are concerned about finances is important because it can prepare clinicians to be ready to approach the subject, which is one that people are often reluctant or ashamed to raise on their own,” Dr. Brendel told this news organization.

What’s the best way to bring up the sensitive topic of money?

“In general, it’s best to start with open-ended questions to allow individuals in therapy to share what is on their minds, explore their concerns, and develop strategies to address these issues. Once a patient raises a concern, that is a good time to ask more about the issues they’ve raised and to explore other potential sources of anxiety or stress,” said Dr. Brendel.

The latest APA poll was conducted by Morning Consult, June 18-20, 2022, among a nationally representative sample of 2,210 adults.

In addition to an uptick in worry about inflation, the poll shows that more than half (51%) of adults are worried about a potential loss of income.

Hispanic adults (66%), mothers (65%), millennials (63%), and genZers (62%) are among the groups most likely to be concerned about income loss.

“Stress is not good for health, mental or physical. So, while it’s a reality that Americans are faced with finding ways of making ends meet, it’s more important than ever to make sure that we are all accessing the care that we need,” said Dr. Brendel.

“People should be aware that there may be low- or no-cost options such as community mental health centers or employer-sponsored resources to address mental health concerns,” she added.
 

Coping with traumatic events

The latest poll also shows that about one-third of adults are worried about gun violence (35% overall, 47% among genZers) or a natural disaster (29%) personally affecting them.

Climate change anxiety is also up slightly in June, compared with May (+4%).

The same goes for mid-term election-related anxiety (+3%) – particularly among Democrats (54% vs. 59%) compared with Republicans (48% vs. 48%).

The latest poll provides insight how Americans would cope after a traumatic event. More adults report they will turn to family and friends for support (60%) than practice self-care (42%), speak openly about their feelings (37%), or seek help from a professional (31%). Nearly one-third (30%) say they will move on from it and not dwell on their feelings.

GenZers are the least likely to say they will speak openly about their feelings (29%) and are less likely than millennials to say they would speak to a health professional (28% vs. 38%).

“While many people show resilience, it’s troubling that most Americans wouldn’t speak openly about their feelings after a traumatic event,” APA CEO and Medical Director Saul Levin, MD, said in the news release.

“In many ways, naming feelings is the most important step toward healing, and this reluctance to air our thoughts may indicate that mental health stigma is still a powerful force in our society,” Dr. Levin said.

After a traumatic current event, 41% of Americans say they consume more news and 30% say they take in more social media, but the majority say this does not impact their mental health, the poll shows.

Two in five adults (43%) say the news of a traumatic event makes them feel more informed, 32% say it makes them feel more anxious, and about one-quarter say it makes them feel overwhelmed (27%) or discouraged (24%).

Dr. Brendel noted that, after a traumatic event, “it’s expected that people may experience anxiety or other symptoms for brief periods of time. However, no two people experience things the same way. If symptoms don’t go away, are overwhelming, or get worse over time, for example, it’s critical to seek help right away.”

The June poll shows that 50% of Americans are anxious about the future of reproductive rights but the poll was conducted before the Dobbs ruling.

Anxiety around COVID-19 continues to ease, with about 47% of Americans saying they are concerned about the pandemic, down 2% among all Americans and 16% among Black Americans since May.

The APA’s Healthy Minds Monthly tracks timely mental health issues throughout the year. The APA also releases its annual Healthy Minds Poll each May in conjunction with Mental Health Awareness Month.

A version of this article first appeared on Medscape.com.

When I began in private practice several eons ago, I employed only registered nurses (RNs) and licensed practical nurses (LPNs) in my office – as did, I think, most other physicians.

That is still the preferred way to go from an efficiency perspective, as well as the ability to delegate such tasks as blood collection and administering intramuscular injections. Unfortunately, the current state of medical practice – driven by payment reform, regulatory changes, technology costs, inflation, and other factors – has forced most independent practitioners to pivot from RNs and LPNs to medical assistants in a majority of situations.

Given this reality, it makes sense to understand how the use of medical assistants has changed private medical practice, and how the most effective MAs manage their roles and maximize their efficiency in the office.

A recent article by two physicians at the University of Michigan, Ann Arbor, is one of the few published papers to address this issue. It presents the results of a cross-sectional study examining the MA’s experience and key factors that enhance or reduce efficiencies.

The authors sent an email survey to 86 MAs working in six clinics within the department of family medicine at the University of Michigan Medical Center, and received responses from 75 of them, including 61 who completed the entire survey. They then singled out 18 individuals deemed “most efficient” by their peers and conducted face-to-face interviews with them.

The surveys and interviews looked at how MAs identified personal strategies for efficiency, dealt with barriers to implementing those strategies, and navigated interoffice relationships, as well as how all of this affected overall job satisfaction.

All 61 respondents who completed the full survey agreed that the MA role was “very important to keep the clinic functioning” and nearly all said that working in health care was “a calling” for them. About half agreed that their work was very stressful, and about the same percentage reported that there was inadequate MA staffing at their clinic. Others complained of limited pay and promotion opportunities.

The surveyed MAs described important work values that increased their efficiency. These included good communication, strong teamwork, and workload sharing, as well as individual strategies such as multitasking, limiting patient conversations, and completing tasks in a consistent way to improve accuracy.

Other strategies identified as contributing to an efficient operation included preclinic huddles, reviews of patient records before the patient’s arrival, and completing routine office duties before the start of office hours.

Respondents were then asked to identify barriers to clinic efficiency, and most of them involved physicians who barked orders at them, did not complete paperwork or sign orders in a timely manner, and agreed to see late-arriving patients. Some MAs suggested that physicians refrain from “talking down” to them, and teach rather than criticize. They also faulted decisions affecting patient flow made by other staffers without soliciting the MAs’ input.

Despite these barriers, the authors found that most of the surveyed MAs agreed that their work was valued by doctors. “Proper training of managers to provide ... support and ensure equitable workloads may be one strategy to ensure that staff members feel the workplace is fair and collegial,” they said.

“Many described the working relationships with physicians as critical to their satisfaction at work and indicated that strong partnerships motivated them to do their best to make the physician’s day easier,” they added.

At the same time, the authors noted that most survey subjects reported that their jobs were “stressful,” and believed that their stress went underrecognized by physicians. They argued that “it’s important for physicians to be cognizant of these patterns and clinic culture, as reducing a hierarchy-based environment will be appreciated by MAs.”

Since this study involved only MAs in a family practice setting, further studies will be needed to determine whether these results translate to specialty offices – and whether the unique issues inherent in various specialty environments elicit different efficiency contributors and barriers.

Overall, though, “staff job satisfaction is linked to improved quality of care, so treating staff well contributes to high-value care for patients,” the authors wrote. “Disseminating practices that staff members themselves have identified as effective, and being attentive to how staff members are treated, may increase individual efficiency while improving staff retention and satisfaction.”

Dr. Eastern practices dermatology and dermatologic surgery in Belleville, N.J. He is the author of numerous articles and textbook chapters, and is a longtime monthly columnist for Dermatology News. Write to him at [email protected].

When I began in private practice several eons ago, I employed only registered nurses (RNs) and licensed practical nurses (LPNs) in my office – as did, I think, most other physicians.

That is still the preferred way to go from an efficiency perspective, as well as the ability to delegate such tasks as blood collection and administering intramuscular injections. Unfortunately, the current state of medical practice – driven by payment reform, regulatory changes, technology costs, inflation, and other factors – has forced most independent practitioners to pivot from RNs and LPNs to medical assistants in a majority of situations.

Given this reality, it makes sense to understand how the use of medical assistants has changed private medical practice, and how the most effective MAs manage their roles and maximize their efficiency in the office.

A recent article by two physicians at the University of Michigan, Ann Arbor, is one of the few published papers to address this issue. It presents the results of a cross-sectional study examining the MA’s experience and key factors that enhance or reduce efficiencies.

The authors sent an email survey to 86 MAs working in six clinics within the department of family medicine at the University of Michigan Medical Center, and received responses from 75 of them, including 61 who completed the entire survey. They then singled out 18 individuals deemed “most efficient” by their peers and conducted face-to-face interviews with them.

The surveys and interviews looked at how MAs identified personal strategies for efficiency, dealt with barriers to implementing those strategies, and navigated interoffice relationships, as well as how all of this affected overall job satisfaction.

All 61 respondents who completed the full survey agreed that the MA role was “very important to keep the clinic functioning” and nearly all said that working in health care was “a calling” for them. About half agreed that their work was very stressful, and about the same percentage reported that there was inadequate MA staffing at their clinic. Others complained of limited pay and promotion opportunities.

The surveyed MAs described important work values that increased their efficiency. These included good communication, strong teamwork, and workload sharing, as well as individual strategies such as multitasking, limiting patient conversations, and completing tasks in a consistent way to improve accuracy.

Other strategies identified as contributing to an efficient operation included preclinic huddles, reviews of patient records before the patient’s arrival, and completing routine office duties before the start of office hours.

Respondents were then asked to identify barriers to clinic efficiency, and most of them involved physicians who barked orders at them, did not complete paperwork or sign orders in a timely manner, and agreed to see late-arriving patients. Some MAs suggested that physicians refrain from “talking down” to them, and teach rather than criticize. They also faulted decisions affecting patient flow made by other staffers without soliciting the MAs’ input.

Despite these barriers, the authors found that most of the surveyed MAs agreed that their work was valued by doctors. “Proper training of managers to provide ... support and ensure equitable workloads may be one strategy to ensure that staff members feel the workplace is fair and collegial,” they said.

“Many described the working relationships with physicians as critical to their satisfaction at work and indicated that strong partnerships motivated them to do their best to make the physician’s day easier,” they added.

At the same time, the authors noted that most survey subjects reported that their jobs were “stressful,” and believed that their stress went underrecognized by physicians. They argued that “it’s important for physicians to be cognizant of these patterns and clinic culture, as reducing a hierarchy-based environment will be appreciated by MAs.”

Since this study involved only MAs in a family practice setting, further studies will be needed to determine whether these results translate to specialty offices – and whether the unique issues inherent in various specialty environments elicit different efficiency contributors and barriers.

Overall, though, “staff job satisfaction is linked to improved quality of care, so treating staff well contributes to high-value care for patients,” the authors wrote. “Disseminating practices that staff members themselves have identified as effective, and being attentive to how staff members are treated, may increase individual efficiency while improving staff retention and satisfaction.”

Dr. Eastern practices dermatology and dermatologic surgery in Belleville, N.J. He is the author of numerous articles and textbook chapters, and is a longtime monthly columnist for Dermatology News. Write to him at [email protected].

When I began in private practice several eons ago, I employed only registered nurses (RNs) and licensed practical nurses (LPNs) in my office – as did, I think, most other physicians.

That is still the preferred way to go from an efficiency perspective, as well as the ability to delegate such tasks as blood collection and administering intramuscular injections. Unfortunately, the current state of medical practice – driven by payment reform, regulatory changes, technology costs, inflation, and other factors – has forced most independent practitioners to pivot from RNs and LPNs to medical assistants in a majority of situations.

Given this reality, it makes sense to understand how the use of medical assistants has changed private medical practice, and how the most effective MAs manage their roles and maximize their efficiency in the office.

A recent article by two physicians at the University of Michigan, Ann Arbor, is one of the few published papers to address this issue. It presents the results of a cross-sectional study examining the MA’s experience and key factors that enhance or reduce efficiencies.

The authors sent an email survey to 86 MAs working in six clinics within the department of family medicine at the University of Michigan Medical Center, and received responses from 75 of them, including 61 who completed the entire survey. They then singled out 18 individuals deemed “most efficient” by their peers and conducted face-to-face interviews with them.

The surveys and interviews looked at how MAs identified personal strategies for efficiency, dealt with barriers to implementing those strategies, and navigated interoffice relationships, as well as how all of this affected overall job satisfaction.

All 61 respondents who completed the full survey agreed that the MA role was “very important to keep the clinic functioning” and nearly all said that working in health care was “a calling” for them. About half agreed that their work was very stressful, and about the same percentage reported that there was inadequate MA staffing at their clinic. Others complained of limited pay and promotion opportunities.

The surveyed MAs described important work values that increased their efficiency. These included good communication, strong teamwork, and workload sharing, as well as individual strategies such as multitasking, limiting patient conversations, and completing tasks in a consistent way to improve accuracy.

Other strategies identified as contributing to an efficient operation included preclinic huddles, reviews of patient records before the patient’s arrival, and completing routine office duties before the start of office hours.

Respondents were then asked to identify barriers to clinic efficiency, and most of them involved physicians who barked orders at them, did not complete paperwork or sign orders in a timely manner, and agreed to see late-arriving patients. Some MAs suggested that physicians refrain from “talking down” to them, and teach rather than criticize. They also faulted decisions affecting patient flow made by other staffers without soliciting the MAs’ input.

Despite these barriers, the authors found that most of the surveyed MAs agreed that their work was valued by doctors. “Proper training of managers to provide ... support and ensure equitable workloads may be one strategy to ensure that staff members feel the workplace is fair and collegial,” they said.

“Many described the working relationships with physicians as critical to their satisfaction at work and indicated that strong partnerships motivated them to do their best to make the physician’s day easier,” they added.

At the same time, the authors noted that most survey subjects reported that their jobs were “stressful,” and believed that their stress went underrecognized by physicians. They argued that “it’s important for physicians to be cognizant of these patterns and clinic culture, as reducing a hierarchy-based environment will be appreciated by MAs.”

Since this study involved only MAs in a family practice setting, further studies will be needed to determine whether these results translate to specialty offices – and whether the unique issues inherent in various specialty environments elicit different efficiency contributors and barriers.

Overall, though, “staff job satisfaction is linked to improved quality of care, so treating staff well contributes to high-value care for patients,” the authors wrote. “Disseminating practices that staff members themselves have identified as effective, and being attentive to how staff members are treated, may increase individual efficiency while improving staff retention and satisfaction.”

Dr. Eastern practices dermatology and dermatologic surgery in Belleville, N.J. He is the author of numerous articles and textbook chapters, and is a longtime monthly columnist for Dermatology News. Write to him at [email protected].

Walking 10,000 steps per day may reduce the risk of death for those who have trouble regulating their blood sugar, according to the findings from a study of almost 1,700 American adults with prediabetes or diabetes.

Researchers from the University of Seville, Spain, evaluated U.S. adults with prediabetes and diabetes using data from the Centers for Disease Control and Prevention’s National Health and Nutrition Examination Survey, collected between 2005 and 2006.

The findings were published this month in Diabetes Care.

Ariel Skelley/Getty Images

Of the total, 1,194 adults had prediabetes, and 493 had diabetes. People with diabetes in the study were diagnosed by a doctor or had a fasting blood glucose level higher than 126 mg/dL. People with prediabetes in the study were also diagnosed by a doctor or had a fasting glucose level from 100 to 125 mg/dL.

Over half (56%) of prediabetic adults were male (average age 55 years), and they took an average of 8,500 steps per day. Half (51%) of the diabetic adults were also male (average age 61 years), and they took fewer steps per day – about 6,300.

The people in the study wore an accelerometer on their waist to count their steps for 7 consecutive days. The researchers adjusted for age, sex, ethnicity, smoking, alcohol use, diet, and use of diabetes medications.

Over 9 years, 200 people with prediabetes and 138 with diabetes died. Based on those who survived after follow-up, walking nearly 10,000 steps per day was best for reducing the risk of death from any cause for people with prediabetes and diabetes.

But about 20% of people in the study were removed from the analysis because they had invalid accelerometry data. Adults who are healthy enough to walk 10,000 steps may have different rates of death from those who aren’t, according to the study authors, who called for more research to compare these two groups.

If 10,000 steps seem like a daunting task, talking to a doctor about finding a routine that works for your physical ability could be helpful, the study authors suggest.

A version of this article first appeared on Medscape.com.

Walking 10,000 steps per day may reduce the risk of death for those who have trouble regulating their blood sugar, according to the findings from a study of almost 1,700 American adults with prediabetes or diabetes.

Researchers from the University of Seville, Spain, evaluated U.S. adults with prediabetes and diabetes using data from the Centers for Disease Control and Prevention’s National Health and Nutrition Examination Survey, collected between 2005 and 2006.

The findings were published this month in Diabetes Care.

Ariel Skelley/Getty Images

Of the total, 1,194 adults had prediabetes, and 493 had diabetes. People with diabetes in the study were diagnosed by a doctor or had a fasting blood glucose level higher than 126 mg/dL. People with prediabetes in the study were also diagnosed by a doctor or had a fasting glucose level from 100 to 125 mg/dL.

Over half (56%) of prediabetic adults were male (average age 55 years), and they took an average of 8,500 steps per day. Half (51%) of the diabetic adults were also male (average age 61 years), and they took fewer steps per day – about 6,300.

The people in the study wore an accelerometer on their waist to count their steps for 7 consecutive days. The researchers adjusted for age, sex, ethnicity, smoking, alcohol use, diet, and use of diabetes medications.

Over 9 years, 200 people with prediabetes and 138 with diabetes died. Based on those who survived after follow-up, walking nearly 10,000 steps per day was best for reducing the risk of death from any cause for people with prediabetes and diabetes.

But about 20% of people in the study were removed from the analysis because they had invalid accelerometry data. Adults who are healthy enough to walk 10,000 steps may have different rates of death from those who aren’t, according to the study authors, who called for more research to compare these two groups.

If 10,000 steps seem like a daunting task, talking to a doctor about finding a routine that works for your physical ability could be helpful, the study authors suggest.

A version of this article first appeared on Medscape.com.

Walking 10,000 steps per day may reduce the risk of death for those who have trouble regulating their blood sugar, according to the findings from a study of almost 1,700 American adults with prediabetes or diabetes.

Researchers from the University of Seville, Spain, evaluated U.S. adults with prediabetes and diabetes using data from the Centers for Disease Control and Prevention’s National Health and Nutrition Examination Survey, collected between 2005 and 2006.

The findings were published this month in Diabetes Care.

Ariel Skelley/Getty Images

Of the total, 1,194 adults had prediabetes, and 493 had diabetes. People with diabetes in the study were diagnosed by a doctor or had a fasting blood glucose level higher than 126 mg/dL. People with prediabetes in the study were also diagnosed by a doctor or had a fasting glucose level from 100 to 125 mg/dL.

Over half (56%) of prediabetic adults were male (average age 55 years), and they took an average of 8,500 steps per day. Half (51%) of the diabetic adults were also male (average age 61 years), and they took fewer steps per day – about 6,300.

The people in the study wore an accelerometer on their waist to count their steps for 7 consecutive days. The researchers adjusted for age, sex, ethnicity, smoking, alcohol use, diet, and use of diabetes medications.

Over 9 years, 200 people with prediabetes and 138 with diabetes died. Based on those who survived after follow-up, walking nearly 10,000 steps per day was best for reducing the risk of death from any cause for people with prediabetes and diabetes.

But about 20% of people in the study were removed from the analysis because they had invalid accelerometry data. Adults who are healthy enough to walk 10,000 steps may have different rates of death from those who aren’t, according to the study authors, who called for more research to compare these two groups.

If 10,000 steps seem like a daunting task, talking to a doctor about finding a routine that works for your physical ability could be helpful, the study authors suggest.

A version of this article first appeared on Medscape.com.

MUSCATINE, IOWA – Bailee Tordai, who was 33 weeks into her pregnancy, barely made it to the prenatal checkup. Her clunky old Jeep couldn’t complete the 2-mile trip from her house to the University of Iowa’s outreach clinic in her southeastern Iowa hometown. It was a hot June day, and a wiring problem made the Jeep conk out in the street.

A passerby helped Ms.Tordai, 22, push her stricken vehicle off the road and into a parking lot. Then she called her stepdad for a ride to the clinic.

Jaclyn Roman, a nurse-midwife, walked into the exam room. “I heard your car broke down.”

“Yup. You want to buy it? Five bucks!” Ms. Tordai joked.

Her lack of reliable transportation won’t be a laughing matter in August, when her baby is due. She will need to arrange for someone to drive her about 40 miles northwest to the University of Iowa Hospitals and Clinics in Iowa City. She can’t give birth at Muscatine’s hospital because it shuttered its birthing unit in 2020.

Ms. Roman is part of an unusual effort to minimize the harm caused by such closures. She’s one of 11 certified nurse-midwives from the University of Iowa who travel regularly to Muscatine and Washington, another southeastern Iowa town where the local hospital closed its birthing unit. The university’s pilot project, which is supported by a federal grant, doesn’t aim to reopen shuttered birthing units. Instead, the midwife team helps ensure area women receive related services. Last year, it served more than 500 patients in Muscatine and Washington.

Muscatine is one of hundreds of rural areas in the United States where hospitals have dropped birthing services during the past 2 decades, often because they lack obstetricians and other specialized staff members.

Hospital industry leaders say birthing units also tend to lose money, largely because of low payments from Medicaid, the public health insurance program that covers more than 40% of births in the United States and an even greater share in many rural areas.

The loss of labor-and-delivery services hits especially hard for women who lack resources and time to travel for care.

Muscatine, which is on the Mississippi River, has more than 23,000 residents, making it a relatively large town by Iowa standards. But its hospital is one of 41 Iowa facilities that have closed their birthing units since 2000, according to the Iowa Department of Public Health. Most were in rural areas. Just one has reopened, and only 56 Iowa hospitals now have birthing units.

The nurse-midwife team’s work includes crucial prenatal checkups. Most pregnant people are supposed to have a dozen or more such appointments before giving birth. Health care providers use the checkups to track how a pregnancy is progressing and to watch for signs of high blood pressure and other problems that can lead to premature births, stillbirths, or even maternal deaths. The midwives also advise women on how to keep themselves and their babies healthy after birth.

Karen Jefferson, DM, director of midwifery practice for the American College of Nurse-Midwives, said the University of Iowa team’s approach is an innovative way to address needs in rural areas that have lost hospital birthing units. “How wonderful would it be to see a provider in your town, instead of driving 40 miles for your prenatal visits – especially toward the end of pregnancy, when you’re going every week,” said Dr. Jefferson, who lives in rural New York.

Midwives can provide many other types of care for women and for babies. In theory, they could even open rural birthing centers outside of hospitals, Dr. Jefferson said. But they would need to overcome concerns about financing and about the availability of surgeons to do emergency cesarean sections, which she said are rarely needed in low-risk births.

The University of Iowa midwives focus on low-risk pregnancies, referring patients with significant health issues to physician specialists in Iowa City. Often, those specialists can visit with the patients and the midwives via video conference in the small-town clinics.

The loss of a hospital obstetrics unit can make finding local maternity care harder for rural families.

Ms. Tordai can attest that if patients must travel far for prenatal appointments, they’re less likely to get to them all. If she had to go to Iowa City for each of hers, repeatedly taking 3 hours off from her job managing a pizza restaurant would be tough, she said. On that June day her Jeep broke down, she would have canceled her appointment.

Instead, she wound up on an exam table at the Muscatine clinic listening to her baby’s heartbeat on a monitor and watching as Ms. Roman measured her belly.

“Nice job being perfect,” the midwife told her during the checkup.

Ms. Roman asked Ms. Tordai to describe her baby’s movements. “Constant,” she replied with a smile.

Ms. Roman asked whether she planned to breastfeed. Ms. Tordai said she didn’t have much luck with her first daughter, Aspen, now 4.

“Have you thought about a breastfeeding class?” the midwife asked.

“I don’t have time for that,” Ms. Tordai replied. Ms. Roman continued to coax her, noting where a breastfeeding class is available online.

Near the end of the appointment, Ms. Tordai asked Ms. Roman whether she could schedule an induced birth at the University of Iowa hospital. The midwife told her that, in general, letting labor begin on its own is better than artificially starting it.

But there was the matter of unreliable transportation. Ms. Tordai explained that scheduling the birth would help her arrange to have her mother drive her to the hospital in Iowa City. Ms. Roman agreed that transportation is a legitimate reason and arranged for an induced labor on Aug. 10.

The University of Iowa midwife team started offering services in 2020 in a clinic about 2 miles from Trinity Muscatine hospital. The hospital is owned by UnityPoint Health, a large nonprofit hospital system that blamed a lack of available obstetricians for the closure of the Muscatine birthing unit. At the time, UnityPoint leaders said they hoped to reopen the unit if they could recruit new obstetricians to the area.

Kristy Phillipson, a UnityPoint Health spokesperson, told KHN in June that the company has continued to try to recruit physicians, including for the Muscatine hospital. Although it has not reopened the birthing unit, the company regularly sends an obstetrician and other staff members to provide prenatal care and related services, she said.

Most pregnant patients from the area who choose UnityPoint for their care wind up giving birth at the system’s hospital in Bettendorf, a 45-minute drive to the east.

The University of Iowa midwife team has no plans to open its own birthing centers, but it hopes to expand its rural clinic service to other underserved towns. To do so, the university would need to hire more nurse-midwives, which could be a challenge. According to the Iowa Board of Nursing, 120 licensed nurse-midwives live in the state of 3 million people.

The University of Iowa plans to address that by starting the state’s first nurse-midwife training program in 2023. The master’s degree program, which will emphasize rural service, will train registered nurses to become nurse-midwives. It eventually could graduate eight people per year, said Amber Goodrich, a University of Iowa midwife helping lead the effort.

Those graduates could fill gaps throughout rural areas, where even more hospitals may shutter their birthing units in the coming years.

“This crisis is going nowhere fast,” Ms. Goodrich said.

KHN (Kaiser Health News) is a national newsroom that produces in-depth journalism about health issues. Together with Policy Analysis and Polling, KHN is one of the three major operating programs at KFF (Kaiser Family Foundation). KFF is an endowed nonprofit organization providing information on health issues to the nation.

MUSCATINE, IOWA – Bailee Tordai, who was 33 weeks into her pregnancy, barely made it to the prenatal checkup. Her clunky old Jeep couldn’t complete the 2-mile trip from her house to the University of Iowa’s outreach clinic in her southeastern Iowa hometown. It was a hot June day, and a wiring problem made the Jeep conk out in the street.

A passerby helped Ms.Tordai, 22, push her stricken vehicle off the road and into a parking lot. Then she called her stepdad for a ride to the clinic.

Jaclyn Roman, a nurse-midwife, walked into the exam room. “I heard your car broke down.”

“Yup. You want to buy it? Five bucks!” Ms. Tordai joked.

Her lack of reliable transportation won’t be a laughing matter in August, when her baby is due. She will need to arrange for someone to drive her about 40 miles northwest to the University of Iowa Hospitals and Clinics in Iowa City. She can’t give birth at Muscatine’s hospital because it shuttered its birthing unit in 2020.

Ms. Roman is part of an unusual effort to minimize the harm caused by such closures. She’s one of 11 certified nurse-midwives from the University of Iowa who travel regularly to Muscatine and Washington, another southeastern Iowa town where the local hospital closed its birthing unit. The university’s pilot project, which is supported by a federal grant, doesn’t aim to reopen shuttered birthing units. Instead, the midwife team helps ensure area women receive related services. Last year, it served more than 500 patients in Muscatine and Washington.

Muscatine is one of hundreds of rural areas in the United States where hospitals have dropped birthing services during the past 2 decades, often because they lack obstetricians and other specialized staff members.

Hospital industry leaders say birthing units also tend to lose money, largely because of low payments from Medicaid, the public health insurance program that covers more than 40% of births in the United States and an even greater share in many rural areas.

The loss of labor-and-delivery services hits especially hard for women who lack resources and time to travel for care.

Muscatine, which is on the Mississippi River, has more than 23,000 residents, making it a relatively large town by Iowa standards. But its hospital is one of 41 Iowa facilities that have closed their birthing units since 2000, according to the Iowa Department of Public Health. Most were in rural areas. Just one has reopened, and only 56 Iowa hospitals now have birthing units.

The nurse-midwife team’s work includes crucial prenatal checkups. Most pregnant people are supposed to have a dozen or more such appointments before giving birth. Health care providers use the checkups to track how a pregnancy is progressing and to watch for signs of high blood pressure and other problems that can lead to premature births, stillbirths, or even maternal deaths. The midwives also advise women on how to keep themselves and their babies healthy after birth.

Karen Jefferson, DM, director of midwifery practice for the American College of Nurse-Midwives, said the University of Iowa team’s approach is an innovative way to address needs in rural areas that have lost hospital birthing units. “How wonderful would it be to see a provider in your town, instead of driving 40 miles for your prenatal visits – especially toward the end of pregnancy, when you’re going every week,” said Dr. Jefferson, who lives in rural New York.

Midwives can provide many other types of care for women and for babies. In theory, they could even open rural birthing centers outside of hospitals, Dr. Jefferson said. But they would need to overcome concerns about financing and about the availability of surgeons to do emergency cesarean sections, which she said are rarely needed in low-risk births.

The University of Iowa midwives focus on low-risk pregnancies, referring patients with significant health issues to physician specialists in Iowa City. Often, those specialists can visit with the patients and the midwives via video conference in the small-town clinics.

The loss of a hospital obstetrics unit can make finding local maternity care harder for rural families.

Ms. Tordai can attest that if patients must travel far for prenatal appointments, they’re less likely to get to them all. If she had to go to Iowa City for each of hers, repeatedly taking 3 hours off from her job managing a pizza restaurant would be tough, she said. On that June day her Jeep broke down, she would have canceled her appointment.

Instead, she wound up on an exam table at the Muscatine clinic listening to her baby’s heartbeat on a monitor and watching as Ms. Roman measured her belly.

“Nice job being perfect,” the midwife told her during the checkup.

Ms. Roman asked Ms. Tordai to describe her baby’s movements. “Constant,” she replied with a smile.

Ms. Roman asked whether she planned to breastfeed. Ms. Tordai said she didn’t have much luck with her first daughter, Aspen, now 4.

“Have you thought about a breastfeeding class?” the midwife asked.

“I don’t have time for that,” Ms. Tordai replied. Ms. Roman continued to coax her, noting where a breastfeeding class is available online.

Near the end of the appointment, Ms. Tordai asked Ms. Roman whether she could schedule an induced birth at the University of Iowa hospital. The midwife told her that, in general, letting labor begin on its own is better than artificially starting it.

But there was the matter of unreliable transportation. Ms. Tordai explained that scheduling the birth would help her arrange to have her mother drive her to the hospital in Iowa City. Ms. Roman agreed that transportation is a legitimate reason and arranged for an induced labor on Aug. 10.

The University of Iowa midwife team started offering services in 2020 in a clinic about 2 miles from Trinity Muscatine hospital. The hospital is owned by UnityPoint Health, a large nonprofit hospital system that blamed a lack of available obstetricians for the closure of the Muscatine birthing unit. At the time, UnityPoint leaders said they hoped to reopen the unit if they could recruit new obstetricians to the area.

Kristy Phillipson, a UnityPoint Health spokesperson, told KHN in June that the company has continued to try to recruit physicians, including for the Muscatine hospital. Although it has not reopened the birthing unit, the company regularly sends an obstetrician and other staff members to provide prenatal care and related services, she said.

Most pregnant patients from the area who choose UnityPoint for their care wind up giving birth at the system’s hospital in Bettendorf, a 45-minute drive to the east.

The University of Iowa midwife team has no plans to open its own birthing centers, but it hopes to expand its rural clinic service to other underserved towns. To do so, the university would need to hire more nurse-midwives, which could be a challenge. According to the Iowa Board of Nursing, 120 licensed nurse-midwives live in the state of 3 million people.

The University of Iowa plans to address that by starting the state’s first nurse-midwife training program in 2023. The master’s degree program, which will emphasize rural service, will train registered nurses to become nurse-midwives. It eventually could graduate eight people per year, said Amber Goodrich, a University of Iowa midwife helping lead the effort.

Those graduates could fill gaps throughout rural areas, where even more hospitals may shutter their birthing units in the coming years.

“This crisis is going nowhere fast,” Ms. Goodrich said.

KHN (Kaiser Health News) is a national newsroom that produces in-depth journalism about health issues. Together with Policy Analysis and Polling, KHN is one of the three major operating programs at KFF (Kaiser Family Foundation). KFF is an endowed nonprofit organization providing information on health issues to the nation.

MUSCATINE, IOWA – Bailee Tordai, who was 33 weeks into her pregnancy, barely made it to the prenatal checkup. Her clunky old Jeep couldn’t complete the 2-mile trip from her house to the University of Iowa’s outreach clinic in her southeastern Iowa hometown. It was a hot June day, and a wiring problem made the Jeep conk out in the street.

A passerby helped Ms.Tordai, 22, push her stricken vehicle off the road and into a parking lot. Then she called her stepdad for a ride to the clinic.

Jaclyn Roman, a nurse-midwife, walked into the exam room. “I heard your car broke down.”

“Yup. You want to buy it? Five bucks!” Ms. Tordai joked.

Her lack of reliable transportation won’t be a laughing matter in August, when her baby is due. She will need to arrange for someone to drive her about 40 miles northwest to the University of Iowa Hospitals and Clinics in Iowa City. She can’t give birth at Muscatine’s hospital because it shuttered its birthing unit in 2020.

Ms. Roman is part of an unusual effort to minimize the harm caused by such closures. She’s one of 11 certified nurse-midwives from the University of Iowa who travel regularly to Muscatine and Washington, another southeastern Iowa town where the local hospital closed its birthing unit. The university’s pilot project, which is supported by a federal grant, doesn’t aim to reopen shuttered birthing units. Instead, the midwife team helps ensure area women receive related services. Last year, it served more than 500 patients in Muscatine and Washington.

Muscatine is one of hundreds of rural areas in the United States where hospitals have dropped birthing services during the past 2 decades, often because they lack obstetricians and other specialized staff members.

Hospital industry leaders say birthing units also tend to lose money, largely because of low payments from Medicaid, the public health insurance program that covers more than 40% of births in the United States and an even greater share in many rural areas.

The loss of labor-and-delivery services hits especially hard for women who lack resources and time to travel for care.

Muscatine, which is on the Mississippi River, has more than 23,000 residents, making it a relatively large town by Iowa standards. But its hospital is one of 41 Iowa facilities that have closed their birthing units since 2000, according to the Iowa Department of Public Health. Most were in rural areas. Just one has reopened, and only 56 Iowa hospitals now have birthing units.

The nurse-midwife team’s work includes crucial prenatal checkups. Most pregnant people are supposed to have a dozen or more such appointments before giving birth. Health care providers use the checkups to track how a pregnancy is progressing and to watch for signs of high blood pressure and other problems that can lead to premature births, stillbirths, or even maternal deaths. The midwives also advise women on how to keep themselves and their babies healthy after birth.

Karen Jefferson, DM, director of midwifery practice for the American College of Nurse-Midwives, said the University of Iowa team’s approach is an innovative way to address needs in rural areas that have lost hospital birthing units. “How wonderful would it be to see a provider in your town, instead of driving 40 miles for your prenatal visits – especially toward the end of pregnancy, when you’re going every week,” said Dr. Jefferson, who lives in rural New York.

Midwives can provide many other types of care for women and for babies. In theory, they could even open rural birthing centers outside of hospitals, Dr. Jefferson said. But they would need to overcome concerns about financing and about the availability of surgeons to do emergency cesarean sections, which she said are rarely needed in low-risk births.

The University of Iowa midwives focus on low-risk pregnancies, referring patients with significant health issues to physician specialists in Iowa City. Often, those specialists can visit with the patients and the midwives via video conference in the small-town clinics.

The loss of a hospital obstetrics unit can make finding local maternity care harder for rural families.

Ms. Tordai can attest that if patients must travel far for prenatal appointments, they’re less likely to get to them all. If she had to go to Iowa City for each of hers, repeatedly taking 3 hours off from her job managing a pizza restaurant would be tough, she said. On that June day her Jeep broke down, she would have canceled her appointment.

Instead, she wound up on an exam table at the Muscatine clinic listening to her baby’s heartbeat on a monitor and watching as Ms. Roman measured her belly.

“Nice job being perfect,” the midwife told her during the checkup.

Ms. Roman asked Ms. Tordai to describe her baby’s movements. “Constant,” she replied with a smile.

Ms. Roman asked whether she planned to breastfeed. Ms. Tordai said she didn’t have much luck with her first daughter, Aspen, now 4.

“Have you thought about a breastfeeding class?” the midwife asked.

“I don’t have time for that,” Ms. Tordai replied. Ms. Roman continued to coax her, noting where a breastfeeding class is available online.

Near the end of the appointment, Ms. Tordai asked Ms. Roman whether she could schedule an induced birth at the University of Iowa hospital. The midwife told her that, in general, letting labor begin on its own is better than artificially starting it.

But there was the matter of unreliable transportation. Ms. Tordai explained that scheduling the birth would help her arrange to have her mother drive her to the hospital in Iowa City. Ms. Roman agreed that transportation is a legitimate reason and arranged for an induced labor on Aug. 10.

The University of Iowa midwife team started offering services in 2020 in a clinic about 2 miles from Trinity Muscatine hospital. The hospital is owned by UnityPoint Health, a large nonprofit hospital system that blamed a lack of available obstetricians for the closure of the Muscatine birthing unit. At the time, UnityPoint leaders said they hoped to reopen the unit if they could recruit new obstetricians to the area.

Kristy Phillipson, a UnityPoint Health spokesperson, told KHN in June that the company has continued to try to recruit physicians, including for the Muscatine hospital. Although it has not reopened the birthing unit, the company regularly sends an obstetrician and other staff members to provide prenatal care and related services, she said.

Most pregnant patients from the area who choose UnityPoint for their care wind up giving birth at the system’s hospital in Bettendorf, a 45-minute drive to the east.

The University of Iowa midwife team has no plans to open its own birthing centers, but it hopes to expand its rural clinic service to other underserved towns. To do so, the university would need to hire more nurse-midwives, which could be a challenge. According to the Iowa Board of Nursing, 120 licensed nurse-midwives live in the state of 3 million people.

The University of Iowa plans to address that by starting the state’s first nurse-midwife training program in 2023. The master’s degree program, which will emphasize rural service, will train registered nurses to become nurse-midwives. It eventually could graduate eight people per year, said Amber Goodrich, a University of Iowa midwife helping lead the effort.

Those graduates could fill gaps throughout rural areas, where even more hospitals may shutter their birthing units in the coming years.

“This crisis is going nowhere fast,” Ms. Goodrich said.

KHN (Kaiser Health News) is a national newsroom that produces in-depth journalism about health issues. Together with Policy Analysis and Polling, KHN is one of the three major operating programs at KFF (Kaiser Family Foundation). KFF is an endowed nonprofit organization providing information on health issues to the nation.

The addition of social determinants of health (SDOH) to machine-learning risk-prediction models improved forecasts of in-hospital mortality in Black adults hospitalized for heart failure (HF) but didn’t show similar ability in non-Black patients, in a study based in part on the American Heart Association–sponsored Get with the Guidelines in Heart Failure (GWTG-HF) registry.

The novel risk-prediction tool bolstered by SDOH at the zip-code level – including household income, number of adults without a high-school degree, poverty and unemployment rates, and other factors – stratified risk more sharply in Black patients than more standard models, including some based on multivariable logistic regression.

“Traditional risk models that exist for heart failure assign lower risks to Black individuals if everything else is held constant,” Ambarish Pandey, MD, MSCS, University of Texas Southwestern Medical Center, Dallas, told this news organization.

“I think that is problematic, because if Black patients are considered lower risk, they may not get appropriate risk-based therapies that are being provided. We wanted to move away from this approach and use a more race-agnostic approach,” said Dr. Pandey, who is senior author on the study published  in JAMA Cardiology, with lead author Matthew W. Segar, MD, Texas Heart Institute, Houston.

The training dataset for the prediction model consisted of 123,634 patients hospitalized with HF (mean age, 71 years), of whom 47% were women, enrolled in the GWTG-HF registry from 2010 through 2020.

The machine-learning models showed “excellent performance” when applied to an internal subset cohort of 82,420 patients, with a C statistic of 0.81 for Black patients and 0.82 for non-Black patients, the authors report, and in a real-world cohort of 553,506 patients, with C statistics of 0.74 and 0.75, respectively. The models performed similarly well, they write, in an external validation cohort derived from the ARIC registry, with C statistics of 0.79 and 0.80, respectively.

The machine-learning models’ performance surpassed that of the GWTG-HF risk-score model, C statistics 0.69 for both Black and non-Black patients, and other logistic regression models in which race was a covariate, the authors state.

“We also observed significant race-specific differences in the population-attributable risk of in-hospital mortality associated with the SDOH, with a significantly greater contribution of these parameters to the overall in-hospital mortality risk in Black patients versus non-Black patients,” they write.

For Black patients, five of the SDOH parameters were among the top 20 covariate predictors of in-hospital mortality: mean income level, vacancy and unemployment rates, proportion of the population without a high school degree, and proportion older than 65 years. Together they accounted for 11.6% of population-attributable risk for in-hospital death.

Only one SDOH parameter – percentage of population older than 65 years – made the top 20 for non-Black patients, with a population-attributable risk of 0.5%, the group reports.

“I hope our work spurs future investigations to better understand how social determinants contribute to risk and how they can be incorporated in management of these patients,” Dr. Pandey said.

“I commend the authors for attempting to address SDOH as a potential contributor to some of the differences in outcomes among patients with heart failure,” writes Eldrin F. Lewis, MD, MPH, Stanford University School of Medicine, Palo Alto, Calif., in an accompanying editorial.

“It is imperative that we use these newer techniques to go beyond simply predicting which groups are at heightened risk and leverage the data to create solutions that will reduce those risks for the individual patient,” Dr. Lewis states.

“We should use these tools to reduce racial and ethnic differences in the operations of health care systems, potential bias in management decisions, and inactivity due to the difficulty in getting guideline-directed medical therapy into the hands of people who may have limited resources with minimal out-of-pocket costs,” he writes.

The models assessed in the current report “set a new bar for risk prediction: Integration of a comprehensive set of demographics, comorbidities, and social determinants with machine learning obviates race and ethnicity in risk prediction,” contend JAMA Cardiology deputy editor Clyde W. Yancy, MD, and associate editor Sadiya S. Khan, MD, both from Northwestern University Feinberg School of Medicine, Chicago, in an accompanying editor’s note.

“This more careful incorporation of individual-level, neighborhood-level, and hospital-level social factors,” they conclude, “is now a candidate template for future risk models.”

Dr. Pandey discloses grant funding from Applied Therapeutics and Gilead Sciences; consulting for or serving as an advisor to Tricog Health, Eli Lilly, Rivus, and Roche Diagnostics; receiving nonfinancial support from Pfizer and Merck; and research support from the Texas Health Resources Clinical Scholarship, the Gilead Sciences Research Scholar Program, the National Institute on Aging GEMSSTAR Grant, and Applied Therapeutics. Dr. Segar discloses receiving nonfinancial support from Pfizer and Merck. Other disclosures are in the report. Dr. Lewis reported no disclosures. Dr. Yancy and Dr. Khan had no relevant disclosures.

A version of this article first appeared on Medscape.com.

The addition of social determinants of health (SDOH) to machine-learning risk-prediction models improved forecasts of in-hospital mortality in Black adults hospitalized for heart failure (HF) but didn’t show similar ability in non-Black patients, in a study based in part on the American Heart Association–sponsored Get with the Guidelines in Heart Failure (GWTG-HF) registry.

The novel risk-prediction tool bolstered by SDOH at the zip-code level – including household income, number of adults without a high-school degree, poverty and unemployment rates, and other factors – stratified risk more sharply in Black patients than more standard models, including some based on multivariable logistic regression.

“Traditional risk models that exist for heart failure assign lower risks to Black individuals if everything else is held constant,” Ambarish Pandey, MD, MSCS, University of Texas Southwestern Medical Center, Dallas, told this news organization.

“I think that is problematic, because if Black patients are considered lower risk, they may not get appropriate risk-based therapies that are being provided. We wanted to move away from this approach and use a more race-agnostic approach,” said Dr. Pandey, who is senior author on the study published  in JAMA Cardiology, with lead author Matthew W. Segar, MD, Texas Heart Institute, Houston.

The training dataset for the prediction model consisted of 123,634 patients hospitalized with HF (mean age, 71 years), of whom 47% were women, enrolled in the GWTG-HF registry from 2010 through 2020.

The machine-learning models showed “excellent performance” when applied to an internal subset cohort of 82,420 patients, with a C statistic of 0.81 for Black patients and 0.82 for non-Black patients, the authors report, and in a real-world cohort of 553,506 patients, with C statistics of 0.74 and 0.75, respectively. The models performed similarly well, they write, in an external validation cohort derived from the ARIC registry, with C statistics of 0.79 and 0.80, respectively.

The machine-learning models’ performance surpassed that of the GWTG-HF risk-score model, C statistics 0.69 for both Black and non-Black patients, and other logistic regression models in which race was a covariate, the authors state.

“We also observed significant race-specific differences in the population-attributable risk of in-hospital mortality associated with the SDOH, with a significantly greater contribution of these parameters to the overall in-hospital mortality risk in Black patients versus non-Black patients,” they write.

For Black patients, five of the SDOH parameters were among the top 20 covariate predictors of in-hospital mortality: mean income level, vacancy and unemployment rates, proportion of the population without a high school degree, and proportion older than 65 years. Together they accounted for 11.6% of population-attributable risk for in-hospital death.

Only one SDOH parameter – percentage of population older than 65 years – made the top 20 for non-Black patients, with a population-attributable risk of 0.5%, the group reports.

“I hope our work spurs future investigations to better understand how social determinants contribute to risk and how they can be incorporated in management of these patients,” Dr. Pandey said.

“I commend the authors for attempting to address SDOH as a potential contributor to some of the differences in outcomes among patients with heart failure,” writes Eldrin F. Lewis, MD, MPH, Stanford University School of Medicine, Palo Alto, Calif., in an accompanying editorial.

“It is imperative that we use these newer techniques to go beyond simply predicting which groups are at heightened risk and leverage the data to create solutions that will reduce those risks for the individual patient,” Dr. Lewis states.

“We should use these tools to reduce racial and ethnic differences in the operations of health care systems, potential bias in management decisions, and inactivity due to the difficulty in getting guideline-directed medical therapy into the hands of people who may have limited resources with minimal out-of-pocket costs,” he writes.

The models assessed in the current report “set a new bar for risk prediction: Integration of a comprehensive set of demographics, comorbidities, and social determinants with machine learning obviates race and ethnicity in risk prediction,” contend JAMA Cardiology deputy editor Clyde W. Yancy, MD, and associate editor Sadiya S. Khan, MD, both from Northwestern University Feinberg School of Medicine, Chicago, in an accompanying editor’s note.

“This more careful incorporation of individual-level, neighborhood-level, and hospital-level social factors,” they conclude, “is now a candidate template for future risk models.”

Dr. Pandey discloses grant funding from Applied Therapeutics and Gilead Sciences; consulting for or serving as an advisor to Tricog Health, Eli Lilly, Rivus, and Roche Diagnostics; receiving nonfinancial support from Pfizer and Merck; and research support from the Texas Health Resources Clinical Scholarship, the Gilead Sciences Research Scholar Program, the National Institute on Aging GEMSSTAR Grant, and Applied Therapeutics. Dr. Segar discloses receiving nonfinancial support from Pfizer and Merck. Other disclosures are in the report. Dr. Lewis reported no disclosures. Dr. Yancy and Dr. Khan had no relevant disclosures.

A version of this article first appeared on Medscape.com.

The addition of social determinants of health (SDOH) to machine-learning risk-prediction models improved forecasts of in-hospital mortality in Black adults hospitalized for heart failure (HF) but didn’t show similar ability in non-Black patients, in a study based in part on the American Heart Association–sponsored Get with the Guidelines in Heart Failure (GWTG-HF) registry.

The novel risk-prediction tool bolstered by SDOH at the zip-code level – including household income, number of adults without a high-school degree, poverty and unemployment rates, and other factors – stratified risk more sharply in Black patients than more standard models, including some based on multivariable logistic regression.

“Traditional risk models that exist for heart failure assign lower risks to Black individuals if everything else is held constant,” Ambarish Pandey, MD, MSCS, University of Texas Southwestern Medical Center, Dallas, told this news organization.

“I think that is problematic, because if Black patients are considered lower risk, they may not get appropriate risk-based therapies that are being provided. We wanted to move away from this approach and use a more race-agnostic approach,” said Dr. Pandey, who is senior author on the study published  in JAMA Cardiology, with lead author Matthew W. Segar, MD, Texas Heart Institute, Houston.

The training dataset for the prediction model consisted of 123,634 patients hospitalized with HF (mean age, 71 years), of whom 47% were women, enrolled in the GWTG-HF registry from 2010 through 2020.

The machine-learning models showed “excellent performance” when applied to an internal subset cohort of 82,420 patients, with a C statistic of 0.81 for Black patients and 0.82 for non-Black patients, the authors report, and in a real-world cohort of 553,506 patients, with C statistics of 0.74 and 0.75, respectively. The models performed similarly well, they write, in an external validation cohort derived from the ARIC registry, with C statistics of 0.79 and 0.80, respectively.

The machine-learning models’ performance surpassed that of the GWTG-HF risk-score model, C statistics 0.69 for both Black and non-Black patients, and other logistic regression models in which race was a covariate, the authors state.

“We also observed significant race-specific differences in the population-attributable risk of in-hospital mortality associated with the SDOH, with a significantly greater contribution of these parameters to the overall in-hospital mortality risk in Black patients versus non-Black patients,” they write.

For Black patients, five of the SDOH parameters were among the top 20 covariate predictors of in-hospital mortality: mean income level, vacancy and unemployment rates, proportion of the population without a high school degree, and proportion older than 65 years. Together they accounted for 11.6% of population-attributable risk for in-hospital death.

Only one SDOH parameter – percentage of population older than 65 years – made the top 20 for non-Black patients, with a population-attributable risk of 0.5%, the group reports.

“I hope our work spurs future investigations to better understand how social determinants contribute to risk and how they can be incorporated in management of these patients,” Dr. Pandey said.

“I commend the authors for attempting to address SDOH as a potential contributor to some of the differences in outcomes among patients with heart failure,” writes Eldrin F. Lewis, MD, MPH, Stanford University School of Medicine, Palo Alto, Calif., in an accompanying editorial.

“It is imperative that we use these newer techniques to go beyond simply predicting which groups are at heightened risk and leverage the data to create solutions that will reduce those risks for the individual patient,” Dr. Lewis states.

“We should use these tools to reduce racial and ethnic differences in the operations of health care systems, potential bias in management decisions, and inactivity due to the difficulty in getting guideline-directed medical therapy into the hands of people who may have limited resources with minimal out-of-pocket costs,” he writes.

The models assessed in the current report “set a new bar for risk prediction: Integration of a comprehensive set of demographics, comorbidities, and social determinants with machine learning obviates race and ethnicity in risk prediction,” contend JAMA Cardiology deputy editor Clyde W. Yancy, MD, and associate editor Sadiya S. Khan, MD, both from Northwestern University Feinberg School of Medicine, Chicago, in an accompanying editor’s note.

“This more careful incorporation of individual-level, neighborhood-level, and hospital-level social factors,” they conclude, “is now a candidate template for future risk models.”

Dr. Pandey discloses grant funding from Applied Therapeutics and Gilead Sciences; consulting for or serving as an advisor to Tricog Health, Eli Lilly, Rivus, and Roche Diagnostics; receiving nonfinancial support from Pfizer and Merck; and research support from the Texas Health Resources Clinical Scholarship, the Gilead Sciences Research Scholar Program, the National Institute on Aging GEMSSTAR Grant, and Applied Therapeutics. Dr. Segar discloses receiving nonfinancial support from Pfizer and Merck. Other disclosures are in the report. Dr. Lewis reported no disclosures. Dr. Yancy and Dr. Khan had no relevant disclosures.

A version of this article first appeared on Medscape.com.

Medically reviewed by Jennifer Casarella, MD

Dan Miller has parked his Nissan Altima on the side of the road near a field outside Chicago and is holding a gun to his head. 

Haunted for years by the compounded trauma of tours of duty in the Middle East and his work as a police officer in Chicago, at that moment, Dr. Miller saw no reason to live. And there were troubles at home with his wife and children, who had grown fearful of his behavior.

“My whole world was falling apart,” he says of that dark night in 2014. “It left a hole I didn’t know how to fill.”

He chose not to pull the trigger after a brochure on the passenger seat of his car gave him an unexpected perspective – and launched him on a path to help others in his situation.

Had Mr. Miller taken his life that night, he would have joined thousands of other veterans who died by suicide. About 17 U.S. veterans lose their lives this way each day, on average, according to the Department of Veterans Affairs. In 2019, the last year for which records are available, 6,261 veterans took their own lives – and the suicide rate for veterans was 52% higher than for nonveterans, the agency’s records show. 

The problem has become so severe that the Veterans Health Administration now uses artificial intelligence (AI) to help identify veterans at the highest risk of suicide – and reach out to them before a crisis strikes.

But that wasn’t available when Dan Miller’s life was unraveling.

In the years leading up to his near-suicide, his wife had pushed him to get help. “She said, ‘You’re not the same person you were when you left. The kids are scared of you. The pets are scared of you,” he recalls.

He resisted, even when his wife threatened divorce. Rising through the ranks of the Marines, Mr. Miller had become more emotionally isolated. He feared losing his job and the respect of others if he let anyone know what he was going through.

Finally, he gave the VHA a chance. He went in for an initial consultation in 2010 and didn’t find it helpful. He didn’t like being told what to do. So he stopped. He turned to obsessive exercise and excessive drinking.

That day in 2014, Mr. Miller’s wife told him she was taking the kids out for a playdate. After she left, he was served with divorce papers. Less than an hour later, he was parked in his car with his gun, ready to end his life.

But if it all had happened just a few years later, things might never have gotten to that point.
 

Scanning for suicide risk

In 2017, the VHA piloted its AI program, called REACH VET, that aims to help prevent veterans from dying by suicide.

Every month, a computer scans the electronic health records of all VHA patients who’ve had a health care visit for any reason in the last 2 years. It checks more than 140 variables and weights them to estimate someone’s overall suicide risk at that moment in time.

To build the risk algorithm, a computer combed through the medical records of 6,360 veterans confirmed to have died by suicide between 2009 and 2011. (The VHA continually updates the list of variables from the health records of VHA patients, including those who have died by suicide since then and others.)

Some variables are things you’d expect:

• A past suicide attempt.
• A diagnosis of depression or other mental illness.
• A diagnosis of a terminal illness.

Others are more surprising. For example, a diagnosis of arthritis or diabetes adds weight.

REACH VET flags the riskiest cases – the top 0.1% – for a mental health or primary care provider to review. They reach out to the patient to tell them how and why their record was flagged, discuss any recommended treatment changes, and ask them to come in for a visit.

“It’s an opportunity to talk about their risk factors, which is designed to lead to a conversation about safety planning,” says clinical psychologist Matthew Miller, PhD, national director of the U.S. Department of Veterans Affairs’ Suicide Prevention Program. He’s not related to Dan Miller.

Making a suicide safety plan

A safety plan is a document that outlines how a person can help prevent their own suicide in a crisis. 

The plan may include:

• A list of personal triggers or warning signs.
• What’s helped them in the past.
• Names of people or organizations who can support them.
• Plans to remove means of suicide, such as guns, from their environment.
• Their reasons for living.

In people at risk for suicide, research shows that having a safety plan reduces suicidal thoughts and attempts, lowers rates of depression and hopelessness, and boosts veterans’ engagement with the health care system. It may also help people manage things that trigger their suicidal thoughts.
 

Getting the call

What if REACH VET had been around when Dan Miller was in crisis – and he’d gotten a call from the VHA?

“It absolutely, positively would have helped because one of the biggest things on that day when I got served was feeling completely alone and that I had no one to turn to,” Mr. Miller says. He’s now a speaker for the Wounded Warrior Project, a nonprofit that serves veterans and active-duty service people.

Vets’ reactions to the unexpected VHA phone call, psychologist Dr. Miller says, “run the gamut from ‘Thank you for contacting me. Let’s talk,’ to ‘What are you talking about? Leave me alone!’ ”

Nothing stops all suicides. But REACH VET is having an impact. In a clinical trial, vets contacted through REACH VET had more doctor visits, were more likely to have a written suicide prevention safety plan, and had fewer hospital admissions for mental health, ER visits, and suicide attempts.
 

An assist from AI

Even simple outreach can make a big difference. And there’s research to prove it.

One study included 4,730 veterans recently discharged from psychiatric care at the VHA, a group considered at high risk for suicide. 

Half of them got 13 caring emails from hospital staff in the weeks after leaving the hospital. The emails mentioned personal things the patient had shared, like a love of hiking, and wished them well. The other veterans got routine follow-up but no emails.

Two years later, those who got the caring emails were less likely to have died by suicide than the other vets. The study was published in 2014 in Contemporary Clinical Trials.

Researchers have done studies like this many times: with handwritten notes from the primary care doctor, postcards from the ER, and so forth. The results never vary: The notes reduce suicide risk.

“If we could use AI to identify people to receive notes or phone calls, it would be a very effective and inexpensive way to guide follow-up care,” says Rebecca Bernert, PhD, director and founder of the Suicide Prevention Research Laboratory at Stanford (Calif.) University.
 

AI doesn’t replace clinical judgment.

“AI can capture data that we miss due to the limits of our humanity,” psychologist Dr. Miller says. “There’s suicide prevention processes founded on big data and AI, and there are processes founded in clinical intuition and acumen.”

AI is only as good as the data it’s based on. If that data lacks diversity, it may miss things. And variables that apply to veterans may differ in civilians.
 

Stopping suicidal thoughts

Google is putting AI to work against suicide, too. Its MUM (Multitask Unified Model) technology seeks to understand the intent behind what we google.

MUM powers Google Search. It can often tell the difference between a search for information about suicide for someone writing a research paper on the topic and a search for information on how or where to carry out a suicide.

When Google Search detects that someone in the United States might be in crisis and at risk of suicide, the first search results that person gets are the number for the National Suicide Prevention Lifeline and other resources for people in crisis.

Google Home Assistant works in the same way. When a user makes a query that signals a suicide-related crisis, the gadget serves up resources that offer help.

MUM is working to understand the nuances of crisis language in 75 languages so that Google Search can provide people in crisis with hotlines or other resources in many countries.

“We want to find partners that are accessible to users in terms of hours of operation. We have a strong preference for finding partners that promise confidentiality and privacy to the extent that those are permitted [in that country],” says Anne Merritt, MD, a product manager at Google Search.

Other companies are working on apps that use AI to spot suicide risk in other ways, including voice technology that may notice subtle changes in the voice of someone who’s depressed and may be thinking of suicide. Those are still in development but show promise. Keep in mind that apps do not require government approval, so if you try one, be sure to let your health care provider know.
 

Changing the channel

Seeing a hotline number on your phone or computer screen can help, Dan Miller says. “If I happened to be online, searching maybe for a bridge to jump off of ... and suddenly that pops up on the screen, it’s like it changes the channel.”

It may not work for everyone, he says, but that search result could interrupt someone’s suicidal train of thought.

That’s crucial, psychologist Dr. Miller says, because most suicide attempts escalate from first thought to potentially fatal action in just 1 hour. That’s how fast it happened for Dan Miller in 2014.

“When you’re able to put time and space between the suicidal thought and the access to the method to act on that thought, you save lives,” Dr. Bernert says.
 

Making a different choice

An interruption in Mr. Miller’s thinking is what had saved his life.

Holding the gun to his head, Mr. Miller looked over at the passenger seat at a brochure from Wounded Warrior Project, which he had just learned about. Mr. Miller noticed a photo of a man in a wheelchair, a veteran like him, who had no legs. He thought that the man looked worse off than him but hadn’t given up.

Mr. Miller put down his gun and decided to get help.

Recovering from a near suicide attempt, he says, is a journey. It doesn’t happen overnight. Now, 8 years later, Mr. Miller is planning a brief break from the speaker circuit. He plans to spend 2 weeks in an outpatient counseling program for posttraumatic stress disorder and traumatic brain injury.

“Telling my story to strangers – part of it is healing me in a way, but I’m learning that repeating the story over and over again is also keeping me from letting it go. And I’m still healing.”

A version of this article first appeared on WebMD.com.

Medically reviewed by Jennifer Casarella, MD

Dan Miller has parked his Nissan Altima on the side of the road near a field outside Chicago and is holding a gun to his head. 

Haunted for years by the compounded trauma of tours of duty in the Middle East and his work as a police officer in Chicago, at that moment, Dr. Miller saw no reason to live. And there were troubles at home with his wife and children, who had grown fearful of his behavior.

“My whole world was falling apart,” he says of that dark night in 2014. “It left a hole I didn’t know how to fill.”

He chose not to pull the trigger after a brochure on the passenger seat of his car gave him an unexpected perspective – and launched him on a path to help others in his situation.

Had Mr. Miller taken his life that night, he would have joined thousands of other veterans who died by suicide. About 17 U.S. veterans lose their lives this way each day, on average, according to the Department of Veterans Affairs. In 2019, the last year for which records are available, 6,261 veterans took their own lives – and the suicide rate for veterans was 52% higher than for nonveterans, the agency’s records show. 

The problem has become so severe that the Veterans Health Administration now uses artificial intelligence (AI) to help identify veterans at the highest risk of suicide – and reach out to them before a crisis strikes.

But that wasn’t available when Dan Miller’s life was unraveling.

In the years leading up to his near-suicide, his wife had pushed him to get help. “She said, ‘You’re not the same person you were when you left. The kids are scared of you. The pets are scared of you,” he recalls.

He resisted, even when his wife threatened divorce. Rising through the ranks of the Marines, Mr. Miller had become more emotionally isolated. He feared losing his job and the respect of others if he let anyone know what he was going through.

Finally, he gave the VHA a chance. He went in for an initial consultation in 2010 and didn’t find it helpful. He didn’t like being told what to do. So he stopped. He turned to obsessive exercise and excessive drinking.

That day in 2014, Mr. Miller’s wife told him she was taking the kids out for a playdate. After she left, he was served with divorce papers. Less than an hour later, he was parked in his car with his gun, ready to end his life.

But if it all had happened just a few years later, things might never have gotten to that point.
 

Scanning for suicide risk

In 2017, the VHA piloted its AI program, called REACH VET, that aims to help prevent veterans from dying by suicide.

Every month, a computer scans the electronic health records of all VHA patients who’ve had a health care visit for any reason in the last 2 years. It checks more than 140 variables and weights them to estimate someone’s overall suicide risk at that moment in time.

To build the risk algorithm, a computer combed through the medical records of 6,360 veterans confirmed to have died by suicide between 2009 and 2011. (The VHA continually updates the list of variables from the health records of VHA patients, including those who have died by suicide since then and others.)

Some variables are things you’d expect:

• A past suicide attempt.
• A diagnosis of depression or other mental illness.
• A diagnosis of a terminal illness.

Others are more surprising. For example, a diagnosis of arthritis or diabetes adds weight.

REACH VET flags the riskiest cases – the top 0.1% – for a mental health or primary care provider to review. They reach out to the patient to tell them how and why their record was flagged, discuss any recommended treatment changes, and ask them to come in for a visit.

“It’s an opportunity to talk about their risk factors, which is designed to lead to a conversation about safety planning,” says clinical psychologist Matthew Miller, PhD, national director of the U.S. Department of Veterans Affairs’ Suicide Prevention Program. He’s not related to Dan Miller.

Making a suicide safety plan

A safety plan is a document that outlines how a person can help prevent their own suicide in a crisis. 

The plan may include:

• A list of personal triggers or warning signs.
• What’s helped them in the past.
• Names of people or organizations who can support them.
• Plans to remove means of suicide, such as guns, from their environment.
• Their reasons for living.

In people at risk for suicide, research shows that having a safety plan reduces suicidal thoughts and attempts, lowers rates of depression and hopelessness, and boosts veterans’ engagement with the health care system. It may also help people manage things that trigger their suicidal thoughts.
 

Getting the call

What if REACH VET had been around when Dan Miller was in crisis – and he’d gotten a call from the VHA?

“It absolutely, positively would have helped because one of the biggest things on that day when I got served was feeling completely alone and that I had no one to turn to,” Mr. Miller says. He’s now a speaker for the Wounded Warrior Project, a nonprofit that serves veterans and active-duty service people.

Vets’ reactions to the unexpected VHA phone call, psychologist Dr. Miller says, “run the gamut from ‘Thank you for contacting me. Let’s talk,’ to ‘What are you talking about? Leave me alone!’ ”

Nothing stops all suicides. But REACH VET is having an impact. In a clinical trial, vets contacted through REACH VET had more doctor visits, were more likely to have a written suicide prevention safety plan, and had fewer hospital admissions for mental health, ER visits, and suicide attempts.
 

An assist from AI

Even simple outreach can make a big difference. And there’s research to prove it.

One study included 4,730 veterans recently discharged from psychiatric care at the VHA, a group considered at high risk for suicide. 

Half of them got 13 caring emails from hospital staff in the weeks after leaving the hospital. The emails mentioned personal things the patient had shared, like a love of hiking, and wished them well. The other veterans got routine follow-up but no emails.

Two years later, those who got the caring emails were less likely to have died by suicide than the other vets. The study was published in 2014 in Contemporary Clinical Trials.

Researchers have done studies like this many times: with handwritten notes from the primary care doctor, postcards from the ER, and so forth. The results never vary: The notes reduce suicide risk.

“If we could use AI to identify people to receive notes or phone calls, it would be a very effective and inexpensive way to guide follow-up care,” says Rebecca Bernert, PhD, director and founder of the Suicide Prevention Research Laboratory at Stanford (Calif.) University.
 

AI doesn’t replace clinical judgment.

“AI can capture data that we miss due to the limits of our humanity,” psychologist Dr. Miller says. “There’s suicide prevention processes founded on big data and AI, and there are processes founded in clinical intuition and acumen.”

AI is only as good as the data it’s based on. If that data lacks diversity, it may miss things. And variables that apply to veterans may differ in civilians.
 

Stopping suicidal thoughts

Google is putting AI to work against suicide, too. Its MUM (Multitask Unified Model) technology seeks to understand the intent behind what we google.

MUM powers Google Search. It can often tell the difference between a search for information about suicide for someone writing a research paper on the topic and a search for information on how or where to carry out a suicide.

When Google Search detects that someone in the United States might be in crisis and at risk of suicide, the first search results that person gets are the number for the National Suicide Prevention Lifeline and other resources for people in crisis.

Google Home Assistant works in the same way. When a user makes a query that signals a suicide-related crisis, the gadget serves up resources that offer help.

MUM is working to understand the nuances of crisis language in 75 languages so that Google Search can provide people in crisis with hotlines or other resources in many countries.

“We want to find partners that are accessible to users in terms of hours of operation. We have a strong preference for finding partners that promise confidentiality and privacy to the extent that those are permitted [in that country],” says Anne Merritt, MD, a product manager at Google Search.

Other companies are working on apps that use AI to spot suicide risk in other ways, including voice technology that may notice subtle changes in the voice of someone who’s depressed and may be thinking of suicide. Those are still in development but show promise. Keep in mind that apps do not require government approval, so if you try one, be sure to let your health care provider know.
 

Changing the channel

Seeing a hotline number on your phone or computer screen can help, Dan Miller says. “If I happened to be online, searching maybe for a bridge to jump off of ... and suddenly that pops up on the screen, it’s like it changes the channel.”

It may not work for everyone, he says, but that search result could interrupt someone’s suicidal train of thought.

That’s crucial, psychologist Dr. Miller says, because most suicide attempts escalate from first thought to potentially fatal action in just 1 hour. That’s how fast it happened for Dan Miller in 2014.

“When you’re able to put time and space between the suicidal thought and the access to the method to act on that thought, you save lives,” Dr. Bernert says.
 

Making a different choice

An interruption in Mr. Miller’s thinking is what had saved his life.

Holding the gun to his head, Mr. Miller looked over at the passenger seat at a brochure from Wounded Warrior Project, which he had just learned about. Mr. Miller noticed a photo of a man in a wheelchair, a veteran like him, who had no legs. He thought that the man looked worse off than him but hadn’t given up.

Mr. Miller put down his gun and decided to get help.

Recovering from a near suicide attempt, he says, is a journey. It doesn’t happen overnight. Now, 8 years later, Mr. Miller is planning a brief break from the speaker circuit. He plans to spend 2 weeks in an outpatient counseling program for posttraumatic stress disorder and traumatic brain injury.

“Telling my story to strangers – part of it is healing me in a way, but I’m learning that repeating the story over and over again is also keeping me from letting it go. And I’m still healing.”

A version of this article first appeared on WebMD.com.

Medically reviewed by Jennifer Casarella, MD

Dan Miller has parked his Nissan Altima on the side of the road near a field outside Chicago and is holding a gun to his head. 

Haunted for years by the compounded trauma of tours of duty in the Middle East and his work as a police officer in Chicago, at that moment, Dr. Miller saw no reason to live. And there were troubles at home with his wife and children, who had grown fearful of his behavior.

“My whole world was falling apart,” he says of that dark night in 2014. “It left a hole I didn’t know how to fill.”

He chose not to pull the trigger after a brochure on the passenger seat of his car gave him an unexpected perspective – and launched him on a path to help others in his situation.

Had Mr. Miller taken his life that night, he would have joined thousands of other veterans who died by suicide. About 17 U.S. veterans lose their lives this way each day, on average, according to the Department of Veterans Affairs. In 2019, the last year for which records are available, 6,261 veterans took their own lives – and the suicide rate for veterans was 52% higher than for nonveterans, the agency’s records show. 

The problem has become so severe that the Veterans Health Administration now uses artificial intelligence (AI) to help identify veterans at the highest risk of suicide – and reach out to them before a crisis strikes.

But that wasn’t available when Dan Miller’s life was unraveling.

In the years leading up to his near-suicide, his wife had pushed him to get help. “She said, ‘You’re not the same person you were when you left. The kids are scared of you. The pets are scared of you,” he recalls.

He resisted, even when his wife threatened divorce. Rising through the ranks of the Marines, Mr. Miller had become more emotionally isolated. He feared losing his job and the respect of others if he let anyone know what he was going through.

Finally, he gave the VHA a chance. He went in for an initial consultation in 2010 and didn’t find it helpful. He didn’t like being told what to do. So he stopped. He turned to obsessive exercise and excessive drinking.

That day in 2014, Mr. Miller’s wife told him she was taking the kids out for a playdate. After she left, he was served with divorce papers. Less than an hour later, he was parked in his car with his gun, ready to end his life.

But if it all had happened just a few years later, things might never have gotten to that point.
 

Scanning for suicide risk

In 2017, the VHA piloted its AI program, called REACH VET, that aims to help prevent veterans from dying by suicide.

Every month, a computer scans the electronic health records of all VHA patients who’ve had a health care visit for any reason in the last 2 years. It checks more than 140 variables and weights them to estimate someone’s overall suicide risk at that moment in time.

To build the risk algorithm, a computer combed through the medical records of 6,360 veterans confirmed to have died by suicide between 2009 and 2011. (The VHA continually updates the list of variables from the health records of VHA patients, including those who have died by suicide since then and others.)

Some variables are things you’d expect:

• A past suicide attempt.
• A diagnosis of depression or other mental illness.
• A diagnosis of a terminal illness.

Others are more surprising. For example, a diagnosis of arthritis or diabetes adds weight.

REACH VET flags the riskiest cases – the top 0.1% – for a mental health or primary care provider to review. They reach out to the patient to tell them how and why their record was flagged, discuss any recommended treatment changes, and ask them to come in for a visit.

“It’s an opportunity to talk about their risk factors, which is designed to lead to a conversation about safety planning,” says clinical psychologist Matthew Miller, PhD, national director of the U.S. Department of Veterans Affairs’ Suicide Prevention Program. He’s not related to Dan Miller.

Making a suicide safety plan

A safety plan is a document that outlines how a person can help prevent their own suicide in a crisis. 

The plan may include:

• A list of personal triggers or warning signs.
• What’s helped them in the past.
• Names of people or organizations who can support them.
• Plans to remove means of suicide, such as guns, from their environment.
• Their reasons for living.

In people at risk for suicide, research shows that having a safety plan reduces suicidal thoughts and attempts, lowers rates of depression and hopelessness, and boosts veterans’ engagement with the health care system. It may also help people manage things that trigger their suicidal thoughts.
 

Getting the call

What if REACH VET had been around when Dan Miller was in crisis – and he’d gotten a call from the VHA?

“It absolutely, positively would have helped because one of the biggest things on that day when I got served was feeling completely alone and that I had no one to turn to,” Mr. Miller says. He’s now a speaker for the Wounded Warrior Project, a nonprofit that serves veterans and active-duty service people.

Vets’ reactions to the unexpected VHA phone call, psychologist Dr. Miller says, “run the gamut from ‘Thank you for contacting me. Let’s talk,’ to ‘What are you talking about? Leave me alone!’ ”

Nothing stops all suicides. But REACH VET is having an impact. In a clinical trial, vets contacted through REACH VET had more doctor visits, were more likely to have a written suicide prevention safety plan, and had fewer hospital admissions for mental health, ER visits, and suicide attempts.
 

An assist from AI

Even simple outreach can make a big difference. And there’s research to prove it.

One study included 4,730 veterans recently discharged from psychiatric care at the VHA, a group considered at high risk for suicide. 

Half of them got 13 caring emails from hospital staff in the weeks after leaving the hospital. The emails mentioned personal things the patient had shared, like a love of hiking, and wished them well. The other veterans got routine follow-up but no emails.

Two years later, those who got the caring emails were less likely to have died by suicide than the other vets. The study was published in 2014 in Contemporary Clinical Trials.

Researchers have done studies like this many times: with handwritten notes from the primary care doctor, postcards from the ER, and so forth. The results never vary: The notes reduce suicide risk.

“If we could use AI to identify people to receive notes or phone calls, it would be a very effective and inexpensive way to guide follow-up care,” says Rebecca Bernert, PhD, director and founder of the Suicide Prevention Research Laboratory at Stanford (Calif.) University.
 

AI doesn’t replace clinical judgment.

“AI can capture data that we miss due to the limits of our humanity,” psychologist Dr. Miller says. “There’s suicide prevention processes founded on big data and AI, and there are processes founded in clinical intuition and acumen.”

AI is only as good as the data it’s based on. If that data lacks diversity, it may miss things. And variables that apply to veterans may differ in civilians.
 

Stopping suicidal thoughts

Google is putting AI to work against suicide, too. Its MUM (Multitask Unified Model) technology seeks to understand the intent behind what we google.

MUM powers Google Search. It can often tell the difference between a search for information about suicide for someone writing a research paper on the topic and a search for information on how or where to carry out a suicide.

When Google Search detects that someone in the United States might be in crisis and at risk of suicide, the first search results that person gets are the number for the National Suicide Prevention Lifeline and other resources for people in crisis.

Google Home Assistant works in the same way. When a user makes a query that signals a suicide-related crisis, the gadget serves up resources that offer help.

MUM is working to understand the nuances of crisis language in 75 languages so that Google Search can provide people in crisis with hotlines or other resources in many countries.

“We want to find partners that are accessible to users in terms of hours of operation. We have a strong preference for finding partners that promise confidentiality and privacy to the extent that those are permitted [in that country],” says Anne Merritt, MD, a product manager at Google Search.

Other companies are working on apps that use AI to spot suicide risk in other ways, including voice technology that may notice subtle changes in the voice of someone who’s depressed and may be thinking of suicide. Those are still in development but show promise. Keep in mind that apps do not require government approval, so if you try one, be sure to let your health care provider know.
 

Changing the channel

Seeing a hotline number on your phone or computer screen can help, Dan Miller says. “If I happened to be online, searching maybe for a bridge to jump off of ... and suddenly that pops up on the screen, it’s like it changes the channel.”

It may not work for everyone, he says, but that search result could interrupt someone’s suicidal train of thought.

That’s crucial, psychologist Dr. Miller says, because most suicide attempts escalate from first thought to potentially fatal action in just 1 hour. That’s how fast it happened for Dan Miller in 2014.

“When you’re able to put time and space between the suicidal thought and the access to the method to act on that thought, you save lives,” Dr. Bernert says.
 

Making a different choice

An interruption in Mr. Miller’s thinking is what had saved his life.

Holding the gun to his head, Mr. Miller looked over at the passenger seat at a brochure from Wounded Warrior Project, which he had just learned about. Mr. Miller noticed a photo of a man in a wheelchair, a veteran like him, who had no legs. He thought that the man looked worse off than him but hadn’t given up.

Mr. Miller put down his gun and decided to get help.

Recovering from a near suicide attempt, he says, is a journey. It doesn’t happen overnight. Now, 8 years later, Mr. Miller is planning a brief break from the speaker circuit. He plans to spend 2 weeks in an outpatient counseling program for posttraumatic stress disorder and traumatic brain injury.

“Telling my story to strangers – part of it is healing me in a way, but I’m learning that repeating the story over and over again is also keeping me from letting it go. And I’m still healing.”

A version of this article first appeared on WebMD.com.

Iron accumulation in the brain as a result of alcohol consumption may explain why even moderate drinking is linked to compromised cognitive function.

Results of a large observational study suggest brain iron accumulation is a “plausible pathway” through which alcohol negatively affects cognition, study Anya Topiwala, MD, PhD, senior clinical researcher, Nuffield Department of Population Health, University of Oxford, England, said in an interview.

Study participants who drank 56 grams of alcohol a week had higher brain iron levels. The U.K. guideline for “low risk” alcohol consumption is less than 14 units weekly, or 112 grams.

“We are finding harmful associations with iron within those low-risk alcohol intake guidelines,” said Dr. Topiwala.

The study was published online  in PLOS Medicine.
 

Early intervention opportunity?

Previous research suggests higher brain iron may be involved in the pathophysiology of Alzheimer’s and Parkinson’s diseases. However, it’s unclear whether deposition plays a role in alcohol’s effect on the brain and if it does, whether this could present an opportunity for early intervention with, for example, chelating agents.

The study included 20,729 participants in the UK Biobank study, which recruited volunteers from 2006 to 2010. Participants had a mean age of 54.8 years, and 48.6% were female.

Participants self-identified as current, never, or previous alcohol consumers. For current drinkers, researchers calculated the total weekly number of U.K. units of alcohol consumed. One unit is 8 grams. A standard drink in the United States is 14 grams. They categorized weekly consumption into quintiles and used the lowest quintile as the reference category.

Participants underwent MRI to determine brain iron levels. Areas of interest were deep brain structures in the basal ganglia.

Mean weekly alcohol consumption was 17.7 units, which is higher than U.K. guidelines for low-risk consumption. “Half of the sample were drinking above what is recommended,” said Dr. Topiwala.

Alcohol consumption was associated with markers of higher iron in the bilateral putamen (beta, 0.08 standard deviation; 95% confidence interval, 0.06-0.09; P < .001), caudate (beta, 0.05; 95% CI, 0.04-0.07; P < .001), and substantia nigra (beta, 0.03; 95% CI; 0.02-0.05; P < .001).
 

Poorer performance

Drinking more than 7 units (56 grams) weekly was associated with higher susceptibility for all brain regions, except the thalamus.

Controlling for menopause status did not alter associations between alcohol and susceptibility for any brain region. This was also the case when excluding blood pressure and cholesterol as covariates.

There were significant interactions with age in the bilateral putamen and caudate but not with sex, smoking, or Townsend Deprivation Index, which includes such factors as unemployment and living conditions.

To gather data on liver iron levels, participants underwent abdominal imaging at the same time as brain imaging. Dr. Topiwala explained that the liver is a primary storage center for iron, so it was used as “a kind of surrogate marker” of iron in the body.

The researchers showed an indirect effect of alcohol through systemic iron. A 1 SD increase in weekly alcohol consumption was associated with a 0.05 mg/g (95% CI, 0.02-0.07; P < .001) increase in liver iron. In addition, a 1 mg/g increase in liver iron was associated with a 0.44 (95% CI, 0.35-0.52; P < .001) SD increase in left putamen susceptibility.

In this sample, 32% (95% CI, 22-49; P < .001) of alcohol’s total effect on left putamen susceptibility was mediated via higher systemic iron levels.

To minimize the impact of other factors influencing the association between alcohol consumption and brain iron – and the possibility that people with more brain iron drink more – researchers used Mendelian randomization that considers genetically predicted alcohol intake. This analysis supported findings of associations between alcohol consumption and brain iron.

Participants completed a cognitive battery, which included trail-making tests that reflect executive function, puzzle tests that assess fluid intelligence or logic and reasoning, and task-based tests using the “Snap” card game to measure reaction time.

Investigators found the more iron that was present in certain brain regions, the poorer participants’ cognitive performance.

Patients should know about the risks of moderate alcohol intake so they can make decisions about drinking, said Dr. Topiwala. “They should be aware that 14 units of alcohol per week is not a zero risk.”
 

Novel research

Commenting for this news organization, Heather Snyder, PhD, vice president of medical and scientific relations, Alzheimer’s Association, noted the study’s large size as a strength of the research.

She noted previous research has shown an association between higher iron levels and alcohol dependence and worse cognitive function, but the potential connection of brain iron levels, moderate alcohol consumption, and cognition has not been studied to date.

“This paper aims to look at whether there is a potential biological link between moderate alcohol consumption and cognition through iron-related pathways.”

The authors suggest more work is needed to understand whether alcohol consumption impacts iron-related biologies to affect downstream cognition, said Dr. Snyder. “Although this study does not answer that question, it does highlight some important questions.”

Study authors received funding from Wellcome Trust, UK Medical Research Council, National Institute for Health Research (NIHR) Oxford Biomedical Research Centre, BHF Centre of Research Excellence, British Heart Foundation, NIHR Cambridge Biomedical Research Centre, U.S. Department of Veterans Affairs, China Scholarship Council, and Li Ka Shing Centre for Health Information and Discovery. Dr. Topiwala has disclosed no relevant financial relationships.

A version of this article first appeared on Medscape.com.

Iron accumulation in the brain as a result of alcohol consumption may explain why even moderate drinking is linked to compromised cognitive function.

Results of a large observational study suggest brain iron accumulation is a “plausible pathway” through which alcohol negatively affects cognition, study Anya Topiwala, MD, PhD, senior clinical researcher, Nuffield Department of Population Health, University of Oxford, England, said in an interview.

Study participants who drank 56 grams of alcohol a week had higher brain iron levels. The U.K. guideline for “low risk” alcohol consumption is less than 14 units weekly, or 112 grams.

“We are finding harmful associations with iron within those low-risk alcohol intake guidelines,” said Dr. Topiwala.

The study was published online  in PLOS Medicine.
 

Early intervention opportunity?

Previous research suggests higher brain iron may be involved in the pathophysiology of Alzheimer’s and Parkinson’s diseases. However, it’s unclear whether deposition plays a role in alcohol’s effect on the brain and if it does, whether this could present an opportunity for early intervention with, for example, chelating agents.

The study included 20,729 participants in the UK Biobank study, which recruited volunteers from 2006 to 2010. Participants had a mean age of 54.8 years, and 48.6% were female.

Participants self-identified as current, never, or previous alcohol consumers. For current drinkers, researchers calculated the total weekly number of U.K. units of alcohol consumed. One unit is 8 grams. A standard drink in the United States is 14 grams. They categorized weekly consumption into quintiles and used the lowest quintile as the reference category.

Participants underwent MRI to determine brain iron levels. Areas of interest were deep brain structures in the basal ganglia.

Mean weekly alcohol consumption was 17.7 units, which is higher than U.K. guidelines for low-risk consumption. “Half of the sample were drinking above what is recommended,” said Dr. Topiwala.

Alcohol consumption was associated with markers of higher iron in the bilateral putamen (beta, 0.08 standard deviation; 95% confidence interval, 0.06-0.09; P < .001), caudate (beta, 0.05; 95% CI, 0.04-0.07; P < .001), and substantia nigra (beta, 0.03; 95% CI; 0.02-0.05; P < .001).
 

Poorer performance

Drinking more than 7 units (56 grams) weekly was associated with higher susceptibility for all brain regions, except the thalamus.

Controlling for menopause status did not alter associations between alcohol and susceptibility for any brain region. This was also the case when excluding blood pressure and cholesterol as covariates.

There were significant interactions with age in the bilateral putamen and caudate but not with sex, smoking, or Townsend Deprivation Index, which includes such factors as unemployment and living conditions.

To gather data on liver iron levels, participants underwent abdominal imaging at the same time as brain imaging. Dr. Topiwala explained that the liver is a primary storage center for iron, so it was used as “a kind of surrogate marker” of iron in the body.

The researchers showed an indirect effect of alcohol through systemic iron. A 1 SD increase in weekly alcohol consumption was associated with a 0.05 mg/g (95% CI, 0.02-0.07; P < .001) increase in liver iron. In addition, a 1 mg/g increase in liver iron was associated with a 0.44 (95% CI, 0.35-0.52; P < .001) SD increase in left putamen susceptibility.

In this sample, 32% (95% CI, 22-49; P < .001) of alcohol’s total effect on left putamen susceptibility was mediated via higher systemic iron levels.

To minimize the impact of other factors influencing the association between alcohol consumption and brain iron – and the possibility that people with more brain iron drink more – researchers used Mendelian randomization that considers genetically predicted alcohol intake. This analysis supported findings of associations between alcohol consumption and brain iron.

Participants completed a cognitive battery, which included trail-making tests that reflect executive function, puzzle tests that assess fluid intelligence or logic and reasoning, and task-based tests using the “Snap” card game to measure reaction time.

Investigators found the more iron that was present in certain brain regions, the poorer participants’ cognitive performance.

Patients should know about the risks of moderate alcohol intake so they can make decisions about drinking, said Dr. Topiwala. “They should be aware that 14 units of alcohol per week is not a zero risk.”
 

Novel research

Commenting for this news organization, Heather Snyder, PhD, vice president of medical and scientific relations, Alzheimer’s Association, noted the study’s large size as a strength of the research.

She noted previous research has shown an association between higher iron levels and alcohol dependence and worse cognitive function, but the potential connection of brain iron levels, moderate alcohol consumption, and cognition has not been studied to date.

“This paper aims to look at whether there is a potential biological link between moderate alcohol consumption and cognition through iron-related pathways.”

The authors suggest more work is needed to understand whether alcohol consumption impacts iron-related biologies to affect downstream cognition, said Dr. Snyder. “Although this study does not answer that question, it does highlight some important questions.”

Study authors received funding from Wellcome Trust, UK Medical Research Council, National Institute for Health Research (NIHR) Oxford Biomedical Research Centre, BHF Centre of Research Excellence, British Heart Foundation, NIHR Cambridge Biomedical Research Centre, U.S. Department of Veterans Affairs, China Scholarship Council, and Li Ka Shing Centre for Health Information and Discovery. Dr. Topiwala has disclosed no relevant financial relationships.

A version of this article first appeared on Medscape.com.

Iron accumulation in the brain as a result of alcohol consumption may explain why even moderate drinking is linked to compromised cognitive function.

Results of a large observational study suggest brain iron accumulation is a “plausible pathway” through which alcohol negatively affects cognition, study Anya Topiwala, MD, PhD, senior clinical researcher, Nuffield Department of Population Health, University of Oxford, England, said in an interview.

Study participants who drank 56 grams of alcohol a week had higher brain iron levels. The U.K. guideline for “low risk” alcohol consumption is less than 14 units weekly, or 112 grams.

“We are finding harmful associations with iron within those low-risk alcohol intake guidelines,” said Dr. Topiwala.

The study was published online  in PLOS Medicine.
 

Early intervention opportunity?

Previous research suggests higher brain iron may be involved in the pathophysiology of Alzheimer’s and Parkinson’s diseases. However, it’s unclear whether deposition plays a role in alcohol’s effect on the brain and if it does, whether this could present an opportunity for early intervention with, for example, chelating agents.

The study included 20,729 participants in the UK Biobank study, which recruited volunteers from 2006 to 2010. Participants had a mean age of 54.8 years, and 48.6% were female.

Participants self-identified as current, never, or previous alcohol consumers. For current drinkers, researchers calculated the total weekly number of U.K. units of alcohol consumed. One unit is 8 grams. A standard drink in the United States is 14 grams. They categorized weekly consumption into quintiles and used the lowest quintile as the reference category.

Participants underwent MRI to determine brain iron levels. Areas of interest were deep brain structures in the basal ganglia.

Mean weekly alcohol consumption was 17.7 units, which is higher than U.K. guidelines for low-risk consumption. “Half of the sample were drinking above what is recommended,” said Dr. Topiwala.

Alcohol consumption was associated with markers of higher iron in the bilateral putamen (beta, 0.08 standard deviation; 95% confidence interval, 0.06-0.09; P < .001), caudate (beta, 0.05; 95% CI, 0.04-0.07; P < .001), and substantia nigra (beta, 0.03; 95% CI; 0.02-0.05; P < .001).
 

Poorer performance

Drinking more than 7 units (56 grams) weekly was associated with higher susceptibility for all brain regions, except the thalamus.

Controlling for menopause status did not alter associations between alcohol and susceptibility for any brain region. This was also the case when excluding blood pressure and cholesterol as covariates.

There were significant interactions with age in the bilateral putamen and caudate but not with sex, smoking, or Townsend Deprivation Index, which includes such factors as unemployment and living conditions.

To gather data on liver iron levels, participants underwent abdominal imaging at the same time as brain imaging. Dr. Topiwala explained that the liver is a primary storage center for iron, so it was used as “a kind of surrogate marker” of iron in the body.

The researchers showed an indirect effect of alcohol through systemic iron. A 1 SD increase in weekly alcohol consumption was associated with a 0.05 mg/g (95% CI, 0.02-0.07; P < .001) increase in liver iron. In addition, a 1 mg/g increase in liver iron was associated with a 0.44 (95% CI, 0.35-0.52; P < .001) SD increase in left putamen susceptibility.

In this sample, 32% (95% CI, 22-49; P < .001) of alcohol’s total effect on left putamen susceptibility was mediated via higher systemic iron levels.

To minimize the impact of other factors influencing the association between alcohol consumption and brain iron – and the possibility that people with more brain iron drink more – researchers used Mendelian randomization that considers genetically predicted alcohol intake. This analysis supported findings of associations between alcohol consumption and brain iron.

Participants completed a cognitive battery, which included trail-making tests that reflect executive function, puzzle tests that assess fluid intelligence or logic and reasoning, and task-based tests using the “Snap” card game to measure reaction time.

Investigators found the more iron that was present in certain brain regions, the poorer participants’ cognitive performance.

Patients should know about the risks of moderate alcohol intake so they can make decisions about drinking, said Dr. Topiwala. “They should be aware that 14 units of alcohol per week is not a zero risk.”
 

Novel research

Commenting for this news organization, Heather Snyder, PhD, vice president of medical and scientific relations, Alzheimer’s Association, noted the study’s large size as a strength of the research.

She noted previous research has shown an association between higher iron levels and alcohol dependence and worse cognitive function, but the potential connection of brain iron levels, moderate alcohol consumption, and cognition has not been studied to date.

“This paper aims to look at whether there is a potential biological link between moderate alcohol consumption and cognition through iron-related pathways.”

The authors suggest more work is needed to understand whether alcohol consumption impacts iron-related biologies to affect downstream cognition, said Dr. Snyder. “Although this study does not answer that question, it does highlight some important questions.”

Study authors received funding from Wellcome Trust, UK Medical Research Council, National Institute for Health Research (NIHR) Oxford Biomedical Research Centre, BHF Centre of Research Excellence, British Heart Foundation, NIHR Cambridge Biomedical Research Centre, U.S. Department of Veterans Affairs, China Scholarship Council, and Li Ka Shing Centre for Health Information and Discovery. Dr. Topiwala has disclosed no relevant financial relationships.

A version of this article first appeared on Medscape.com.

Two drugs have emerged as the optimal medications for treating insomnia based on the “best-available evidence,” but there are caveats.

In a comprehensive comparative-effectiveness analysis, lemborexant and eszopiclone showed the best efficacy, acceptability, and tolerability for acute and long-term insomnia treatment.

However, eszopiclone may cause substantial side effects – and safety data on lemborexant were inconclusive, the researchers note.

Not surprisingly, short-acting, intermediate-acting, and long-acting benzodiazepines were effective in the acute treatment of insomnia, but they have unfavorable tolerability and safety profiles, and there are no long-term data on these issues.

For many insomnia medications, there is a “striking” and “appalling” lack of long-term data, study investigator Andrea Cipriani, MD, PhD, professor of psychiatry, University of Oxford, United Kingdom, noted during a press briefing.

“This is a call for regulators to raise the bar and ask for long-term data when companies submit an application for licensing insomnia drugs,” Dr. Cipriani said.

The findings were published online  in The Lancet.
 

Prevalent, debilitating

Insomnia is highly prevalent, affecting up to 1 in 5 adults, and can have a profound impact on health, well-being, and productivity.

Sleep hygiene and cognitive-behavioral therapy for insomnia (CBT-I) are recommended first-line treatments, but they are often unavailable, which often leads patients and clinicians to turn to medications.

However, “insomnia drugs are not all created equal. Even within the same drug class there are differences,” Dr. Cipriani said.

In a large-scale systematic review and network meta-analysis, the researchers analyzed data from 154 double-blind, randomized controlled trials of medications (licensed or not) used for acute and long-term treatment of insomnia in 44,089 adults (mean age, 51.7 years; 63% women).

Results showed, for the acute treatment of insomnia, benzodiazepines, doxylamine, eszopiclone, lemborexant, seltorexant, zolpidem, and zopiclone were more effective than placebo (standardized mean difference range, 0.36-0.83; high-to-moderate certainty of evidence).

In addition, benzodiazepines, eszopiclone, zolpidem, and zopiclone were more effective than melatonin, ramelteon, and zaleplon (SMD, 0.27-0.71; moderate-to-very low certainty of evidence).

“Our results show that the melatonergic drugs melatonin and ramelteon are not really effective. The data do not support the regular use of these drugs,” co-investigator Phil Cowen, PhD, professor of psychopharmacology, University of Oxford, said at the briefing.
 

Best available evidence

What little long-term data is available suggest eszopiclone and lemborexant are more effective than placebo. Plus, eszopiclone is more effective than ramelteon and zolpidem but with “very low” certainty of evidence, the researchers report.

“There was insufficient evidence to support the prescription of benzodiazepines and zolpidem in long-term treatment,” they write.

Another problem was lack of data on other important outcomes, they add.  

“We wanted to look at hangover effects, daytime sleepiness, [and] rebound effect, but often there was no data reported in trials. We need to collect data about these outcomes because they matter to clinicians and patients,” Dr. Cipriani said.

Summing up, the researchers note the current findings represent the “best available evidence base to guide the choice about pharmacological treatment for insomnia disorder in adults and will assist in shared decisionmaking between patients, carers, and their clinicians, as well as policy makers.”

They caution, however, that all statements comparing the merits of one drug with another “should be tempered by the potential limitations of the current analysis, the quality of the available evidence, the characteristics of the patient populations, and the uncertainties that might result from choice of dose or treatment setting.”

In addition, it is important to also consider nonpharmacologic treatments for insomnia disorder, as they are supported by “high-quality evidence and recommended as first-line treatment by guidelines,” the investigator write.
 

Shared decisionmaking

In an accompanying editorial, Myrto Samara, MD, University of Thessaly, Larissa, Greece, agrees with the researchers that discussion with patients is key.  

“For insomnia treatment, patient-physician shared decisionmaking is crucial to decide when a pharmacological intervention is deemed necessary and which drug [is] to be given by considering the trade-offs for efficacy and side effects,” Dr. Samara writes.  

The study was funded by the UK National Institute for Health Research (NIHR) Oxford Health Biomedical Research Center. Dr. Cipriani has received research and consultancy fees from the Italian Network for Pediatric Trials, CARIPLO Foundation, and Angelini Pharma, and is the chief and principal investigator of two trials of seltorexant in depression that are sponsored by Janssen. Dr. Samara has reported no relevant financial relationships.

A version of this article first appeared on Medscape.com.

Two drugs have emerged as the optimal medications for treating insomnia based on the “best-available evidence,” but there are caveats.

In a comprehensive comparative-effectiveness analysis, lemborexant and eszopiclone showed the best efficacy, acceptability, and tolerability for acute and long-term insomnia treatment.

However, eszopiclone may cause substantial side effects – and safety data on lemborexant were inconclusive, the researchers note.

Not surprisingly, short-acting, intermediate-acting, and long-acting benzodiazepines were effective in the acute treatment of insomnia, but they have unfavorable tolerability and safety profiles, and there are no long-term data on these issues.

For many insomnia medications, there is a “striking” and “appalling” lack of long-term data, study investigator Andrea Cipriani, MD, PhD, professor of psychiatry, University of Oxford, United Kingdom, noted during a press briefing.

“This is a call for regulators to raise the bar and ask for long-term data when companies submit an application for licensing insomnia drugs,” Dr. Cipriani said.

The findings were published online  in The Lancet.
 

Prevalent, debilitating

Insomnia is highly prevalent, affecting up to 1 in 5 adults, and can have a profound impact on health, well-being, and productivity.

Sleep hygiene and cognitive-behavioral therapy for insomnia (CBT-I) are recommended first-line treatments, but they are often unavailable, which often leads patients and clinicians to turn to medications.

However, “insomnia drugs are not all created equal. Even within the same drug class there are differences,” Dr. Cipriani said.

In a large-scale systematic review and network meta-analysis, the researchers analyzed data from 154 double-blind, randomized controlled trials of medications (licensed or not) used for acute and long-term treatment of insomnia in 44,089 adults (mean age, 51.7 years; 63% women).

Results showed, for the acute treatment of insomnia, benzodiazepines, doxylamine, eszopiclone, lemborexant, seltorexant, zolpidem, and zopiclone were more effective than placebo (standardized mean difference range, 0.36-0.83; high-to-moderate certainty of evidence).

In addition, benzodiazepines, eszopiclone, zolpidem, and zopiclone were more effective than melatonin, ramelteon, and zaleplon (SMD, 0.27-0.71; moderate-to-very low certainty of evidence).

“Our results show that the melatonergic drugs melatonin and ramelteon are not really effective. The data do not support the regular use of these drugs,” co-investigator Phil Cowen, PhD, professor of psychopharmacology, University of Oxford, said at the briefing.
 

Best available evidence

What little long-term data is available suggest eszopiclone and lemborexant are more effective than placebo. Plus, eszopiclone is more effective than ramelteon and zolpidem but with “very low” certainty of evidence, the researchers report.

“There was insufficient evidence to support the prescription of benzodiazepines and zolpidem in long-term treatment,” they write.

Another problem was lack of data on other important outcomes, they add.  

“We wanted to look at hangover effects, daytime sleepiness, [and] rebound effect, but often there was no data reported in trials. We need to collect data about these outcomes because they matter to clinicians and patients,” Dr. Cipriani said.

Summing up, the researchers note the current findings represent the “best available evidence base to guide the choice about pharmacological treatment for insomnia disorder in adults and will assist in shared decisionmaking between patients, carers, and their clinicians, as well as policy makers.”

They caution, however, that all statements comparing the merits of one drug with another “should be tempered by the potential limitations of the current analysis, the quality of the available evidence, the characteristics of the patient populations, and the uncertainties that might result from choice of dose or treatment setting.”

In addition, it is important to also consider nonpharmacologic treatments for insomnia disorder, as they are supported by “high-quality evidence and recommended as first-line treatment by guidelines,” the investigator write.
 

Shared decisionmaking

In an accompanying editorial, Myrto Samara, MD, University of Thessaly, Larissa, Greece, agrees with the researchers that discussion with patients is key.  

“For insomnia treatment, patient-physician shared decisionmaking is crucial to decide when a pharmacological intervention is deemed necessary and which drug [is] to be given by considering the trade-offs for efficacy and side effects,” Dr. Samara writes.  

The study was funded by the UK National Institute for Health Research (NIHR) Oxford Health Biomedical Research Center. Dr. Cipriani has received research and consultancy fees from the Italian Network for Pediatric Trials, CARIPLO Foundation, and Angelini Pharma, and is the chief and principal investigator of two trials of seltorexant in depression that are sponsored by Janssen. Dr. Samara has reported no relevant financial relationships.

A version of this article first appeared on Medscape.com.

Two drugs have emerged as the optimal medications for treating insomnia based on the “best-available evidence,” but there are caveats.

In a comprehensive comparative-effectiveness analysis, lemborexant and eszopiclone showed the best efficacy, acceptability, and tolerability for acute and long-term insomnia treatment.

However, eszopiclone may cause substantial side effects – and safety data on lemborexant were inconclusive, the researchers note.

Not surprisingly, short-acting, intermediate-acting, and long-acting benzodiazepines were effective in the acute treatment of insomnia, but they have unfavorable tolerability and safety profiles, and there are no long-term data on these issues.

For many insomnia medications, there is a “striking” and “appalling” lack of long-term data, study investigator Andrea Cipriani, MD, PhD, professor of psychiatry, University of Oxford, United Kingdom, noted during a press briefing.

“This is a call for regulators to raise the bar and ask for long-term data when companies submit an application for licensing insomnia drugs,” Dr. Cipriani said.

The findings were published online  in The Lancet.
 

Prevalent, debilitating

Insomnia is highly prevalent, affecting up to 1 in 5 adults, and can have a profound impact on health, well-being, and productivity.

Sleep hygiene and cognitive-behavioral therapy for insomnia (CBT-I) are recommended first-line treatments, but they are often unavailable, which often leads patients and clinicians to turn to medications.

However, “insomnia drugs are not all created equal. Even within the same drug class there are differences,” Dr. Cipriani said.

In a large-scale systematic review and network meta-analysis, the researchers analyzed data from 154 double-blind, randomized controlled trials of medications (licensed or not) used for acute and long-term treatment of insomnia in 44,089 adults (mean age, 51.7 years; 63% women).

Results showed, for the acute treatment of insomnia, benzodiazepines, doxylamine, eszopiclone, lemborexant, seltorexant, zolpidem, and zopiclone were more effective than placebo (standardized mean difference range, 0.36-0.83; high-to-moderate certainty of evidence).

In addition, benzodiazepines, eszopiclone, zolpidem, and zopiclone were more effective than melatonin, ramelteon, and zaleplon (SMD, 0.27-0.71; moderate-to-very low certainty of evidence).

“Our results show that the melatonergic drugs melatonin and ramelteon are not really effective. The data do not support the regular use of these drugs,” co-investigator Phil Cowen, PhD, professor of psychopharmacology, University of Oxford, said at the briefing.
 

Best available evidence

What little long-term data is available suggest eszopiclone and lemborexant are more effective than placebo. Plus, eszopiclone is more effective than ramelteon and zolpidem but with “very low” certainty of evidence, the researchers report.

“There was insufficient evidence to support the prescription of benzodiazepines and zolpidem in long-term treatment,” they write.

Another problem was lack of data on other important outcomes, they add.  

“We wanted to look at hangover effects, daytime sleepiness, [and] rebound effect, but often there was no data reported in trials. We need to collect data about these outcomes because they matter to clinicians and patients,” Dr. Cipriani said.

Summing up, the researchers note the current findings represent the “best available evidence base to guide the choice about pharmacological treatment for insomnia disorder in adults and will assist in shared decisionmaking between patients, carers, and their clinicians, as well as policy makers.”

They caution, however, that all statements comparing the merits of one drug with another “should be tempered by the potential limitations of the current analysis, the quality of the available evidence, the characteristics of the patient populations, and the uncertainties that might result from choice of dose or treatment setting.”

In addition, it is important to also consider nonpharmacologic treatments for insomnia disorder, as they are supported by “high-quality evidence and recommended as first-line treatment by guidelines,” the investigator write.
 

Shared decisionmaking

In an accompanying editorial, Myrto Samara, MD, University of Thessaly, Larissa, Greece, agrees with the researchers that discussion with patients is key.  

“For insomnia treatment, patient-physician shared decisionmaking is crucial to decide when a pharmacological intervention is deemed necessary and which drug [is] to be given by considering the trade-offs for efficacy and side effects,” Dr. Samara writes.  

The study was funded by the UK National Institute for Health Research (NIHR) Oxford Health Biomedical Research Center. Dr. Cipriani has received research and consultancy fees from the Italian Network for Pediatric Trials, CARIPLO Foundation, and Angelini Pharma, and is the chief and principal investigator of two trials of seltorexant in depression that are sponsored by Janssen. Dr. Samara has reported no relevant financial relationships.

A version of this article first appeared on Medscape.com.

When the COVID-19 pandemic first began, the general thought was that once people were infected, they were then protected from the virus.

But a new analysis by ABC News shows that more and more Americans are getting the virus again.

It’s hard to say how many. The ABC News analysis found at least 1.6 million reinfections in 24 states, but the actual number is probably a lot higher.

“These are not the real numbers because many people are not reporting cases,” Ali Mokdad, MD, an epidemiologist with the University of Washington, Seattle, told ABC.

The latest variant, BA.5, has become the dominant strain in the United States, making up more than 65% of all COVID-19 cases as of July 13, according to data from the CDC.

Prior infections and vaccines aren’t providing as much protection against the newly dominant BA.5 strain as they did against earlier variants.

But evidence doesn’t show this subvariant of Omicron to be more harmful than earlier, less transmissible versions.

Several factors are contributing to rising reinfections, experts say. For example, fewer people are wearing masks than in the first year or so of the pandemic. Dr. Mokdad said just 18% of Americans reported always wearing a mask in public at the end of May, down from 44% the year before.

The emergence of the Omicron variant, of which BA.5 is a subvariant, is indicating that less protection is being offered by prior infections.

A version of this article first appeared on WebMD.com.

When the COVID-19 pandemic first began, the general thought was that once people were infected, they were then protected from the virus.

But a new analysis by ABC News shows that more and more Americans are getting the virus again.

It’s hard to say how many. The ABC News analysis found at least 1.6 million reinfections in 24 states, but the actual number is probably a lot higher.

“These are not the real numbers because many people are not reporting cases,” Ali Mokdad, MD, an epidemiologist with the University of Washington, Seattle, told ABC.

The latest variant, BA.5, has become the dominant strain in the United States, making up more than 65% of all COVID-19 cases as of July 13, according to data from the CDC.

Prior infections and vaccines aren’t providing as much protection against the newly dominant BA.5 strain as they did against earlier variants.

But evidence doesn’t show this subvariant of Omicron to be more harmful than earlier, less transmissible versions.

Several factors are contributing to rising reinfections, experts say. For example, fewer people are wearing masks than in the first year or so of the pandemic. Dr. Mokdad said just 18% of Americans reported always wearing a mask in public at the end of May, down from 44% the year before.

The emergence of the Omicron variant, of which BA.5 is a subvariant, is indicating that less protection is being offered by prior infections.

A version of this article first appeared on WebMD.com.

When the COVID-19 pandemic first began, the general thought was that once people were infected, they were then protected from the virus.

But a new analysis by ABC News shows that more and more Americans are getting the virus again.

It’s hard to say how many. The ABC News analysis found at least 1.6 million reinfections in 24 states, but the actual number is probably a lot higher.

“These are not the real numbers because many people are not reporting cases,” Ali Mokdad, MD, an epidemiologist with the University of Washington, Seattle, told ABC.

The latest variant, BA.5, has become the dominant strain in the United States, making up more than 65% of all COVID-19 cases as of July 13, according to data from the CDC.

Prior infections and vaccines aren’t providing as much protection against the newly dominant BA.5 strain as they did against earlier variants.

But evidence doesn’t show this subvariant of Omicron to be more harmful than earlier, less transmissible versions.

Several factors are contributing to rising reinfections, experts say. For example, fewer people are wearing masks than in the first year or so of the pandemic. Dr. Mokdad said just 18% of Americans reported always wearing a mask in public at the end of May, down from 44% the year before.

The emergence of the Omicron variant, of which BA.5 is a subvariant, is indicating that less protection is being offered by prior infections.

A version of this article first appeared on WebMD.com.

NEW YORK – On the basis of MRI images of the hand, a neural network has been trained to distinguish seronegative and seropositive rheumatoid arthritis (RA) from psoriatic arthritis (PsA) as well as from each other, according to a study that was presented at the annual meeting of the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis.

In the work so far, the neural network was correct about 70% of the time in the absence of any further clinical analyses, according to David Simon, MD, a rheumatologist in the department of internal medicine at Friedrich-Alexander University, Erlangen, Germany.

Ted Bosworth/MDedge News

Previous to this work, “there has been no study that has exclusively used hand MRI data and deep learning without requiring further expert input for the classification of arthritides,” Dr. Simon said.

In fact, when demographic and clinical data were added, there was no improvement in the performance of patient classification relative to the deep learning classification alone, according to the data presented by Dr. Simon.

The images were evaluated with residual neural networks (ResNet), which represents a sophisticated form of deep learning to facilitate the flow of information across the network layers as they form to improve accuracy in their ability to distinguish one form of disease from the other. The training was performed on images from the T1 coronal, T2 corona1, T1 coronal fat suppressed with contrast, T1 axial fat suppressed with contrast, and T2 fat suppressed axial sequences.

The study included hand MRI scans from 135 patients with seronegative RA, 190 with seropositive RA, 177 with PsA, and 147 with psoriasis. The performance was judged on the basis of area under the receiver operating characteristics curve (AUROC) with and without input of clinical characteristics. Patients who had psoriasis without clinical arthritis were included as a control population.

The AUROC for accuracy was 75% for seropositive RA relative to PsA, 74% for seronegative RA relative to PsA, and 67% for seropositive relative to seronegative RA. Of the patients who had psoriasis without arthritis, 98% were classified as PsA and 2% as RA.

Subsequent to the classification of the patients with psoriasis, 14 of the 147 (9.5%) have developed PsA so far over a relatively short follow-up. All of these were among those identified as PsA by neural network evaluation of the hand MRIs.

This suggests that “a PsA-like pattern may be present early in the course of psoriatic disease,” Dr. Simon said.

In the groups with joint disease, who had mean ages ranging from 56 to 65, the mean disease durations were 2.6 years for those with seropositive RA, 1.3 years for those with seronegative RA, and 0.8 years for those with PsA. The patients with psoriasis were younger (mean age, 40.5 years) but had a longer disease duration (mean 4.2 years).

All of the MRI sequences were relevant for classification, but contrast did not appear to help with accuracy.

“If the images with contrast enhancement were deleted, the loss of performance was only marginal,” Dr. Simon reported.

The accuracy of neural networks increases with data, making it likely that further refinements in methodology will lead to a greater degree of accuracy, according to Dr. Simon. While the methodology is not yet ready for routine use in the clinic, the study demonstrates that neural network analysis of hand MRI to distinguish forms of arthritis “is possible.” Further studies are planned toward the goal of creating a viable clinical tool.

“Of course, if we could create an accurate tool with ultrasound, this would be even more practical,” said Dr. Simon, recognizing the value of an office tool, but he cautioned that this would be far more challenging.

“The precision of MRI is an important factor for effective neural network training,” he said.

Utility: ‘In challenging cases if the accuracy improves’?

A viable method for objectively and rapidly distinguishing inflammatory joint diseases, particularly in patients with an ambiguous clinical presentation, is an unmet need, according to Philip J. Mease, MD, director of rheumatology research at Swedish Medical Center, Seattle.

Although the data presented are promising, Dr. Mease said in an interview that he believes there is a fair amount of work to be done before imaging analysis based on deep learning makes its way into routine clinical care. He is also hoping for methods to distinguish RA from PsA that are easier and less expensive, such as serum biomarkers. However, he agreed that a MRI-based tool could be useful when differentiating disease that is challenging.

“MRI is an expensive way for routine classification of disease, but this approach could be useful in challenging cases if the accuracy improves,” he said.

Meanwhile, other clinical researchers might want to test the principle. “You can try it,” said Dr. Simon, who reported that his team has made the methodology publicly available.

Dr. Simon reported no conflicts of interest. Dr. Mease reported financial relationships with more than 10 pharmaceutical companies, most of which make products used for the treatment of inflammatory joint diseases.

NEW YORK – On the basis of MRI images of the hand, a neural network has been trained to distinguish seronegative and seropositive rheumatoid arthritis (RA) from psoriatic arthritis (PsA) as well as from each other, according to a study that was presented at the annual meeting of the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis.

In the work so far, the neural network was correct about 70% of the time in the absence of any further clinical analyses, according to David Simon, MD, a rheumatologist in the department of internal medicine at Friedrich-Alexander University, Erlangen, Germany.

Ted Bosworth/MDedge News

Previous to this work, “there has been no study that has exclusively used hand MRI data and deep learning without requiring further expert input for the classification of arthritides,” Dr. Simon said.

In fact, when demographic and clinical data were added, there was no improvement in the performance of patient classification relative to the deep learning classification alone, according to the data presented by Dr. Simon.

The images were evaluated with residual neural networks (ResNet), which represents a sophisticated form of deep learning to facilitate the flow of information across the network layers as they form to improve accuracy in their ability to distinguish one form of disease from the other. The training was performed on images from the T1 coronal, T2 corona1, T1 coronal fat suppressed with contrast, T1 axial fat suppressed with contrast, and T2 fat suppressed axial sequences.

The study included hand MRI scans from 135 patients with seronegative RA, 190 with seropositive RA, 177 with PsA, and 147 with psoriasis. The performance was judged on the basis of area under the receiver operating characteristics curve (AUROC) with and without input of clinical characteristics. Patients who had psoriasis without clinical arthritis were included as a control population.

The AUROC for accuracy was 75% for seropositive RA relative to PsA, 74% for seronegative RA relative to PsA, and 67% for seropositive relative to seronegative RA. Of the patients who had psoriasis without arthritis, 98% were classified as PsA and 2% as RA.

Subsequent to the classification of the patients with psoriasis, 14 of the 147 (9.5%) have developed PsA so far over a relatively short follow-up. All of these were among those identified as PsA by neural network evaluation of the hand MRIs.

This suggests that “a PsA-like pattern may be present early in the course of psoriatic disease,” Dr. Simon said.

In the groups with joint disease, who had mean ages ranging from 56 to 65, the mean disease durations were 2.6 years for those with seropositive RA, 1.3 years for those with seronegative RA, and 0.8 years for those with PsA. The patients with psoriasis were younger (mean age, 40.5 years) but had a longer disease duration (mean 4.2 years).

All of the MRI sequences were relevant for classification, but contrast did not appear to help with accuracy.

“If the images with contrast enhancement were deleted, the loss of performance was only marginal,” Dr. Simon reported.

The accuracy of neural networks increases with data, making it likely that further refinements in methodology will lead to a greater degree of accuracy, according to Dr. Simon. While the methodology is not yet ready for routine use in the clinic, the study demonstrates that neural network analysis of hand MRI to distinguish forms of arthritis “is possible.” Further studies are planned toward the goal of creating a viable clinical tool.

“Of course, if we could create an accurate tool with ultrasound, this would be even more practical,” said Dr. Simon, recognizing the value of an office tool, but he cautioned that this would be far more challenging.

“The precision of MRI is an important factor for effective neural network training,” he said.

Utility: ‘In challenging cases if the accuracy improves’?

A viable method for objectively and rapidly distinguishing inflammatory joint diseases, particularly in patients with an ambiguous clinical presentation, is an unmet need, according to Philip J. Mease, MD, director of rheumatology research at Swedish Medical Center, Seattle.

Although the data presented are promising, Dr. Mease said in an interview that he believes there is a fair amount of work to be done before imaging analysis based on deep learning makes its way into routine clinical care. He is also hoping for methods to distinguish RA from PsA that are easier and less expensive, such as serum biomarkers. However, he agreed that a MRI-based tool could be useful when differentiating disease that is challenging.

“MRI is an expensive way for routine classification of disease, but this approach could be useful in challenging cases if the accuracy improves,” he said.

Meanwhile, other clinical researchers might want to test the principle. “You can try it,” said Dr. Simon, who reported that his team has made the methodology publicly available.

Dr. Simon reported no conflicts of interest. Dr. Mease reported financial relationships with more than 10 pharmaceutical companies, most of which make products used for the treatment of inflammatory joint diseases.

NEW YORK – On the basis of MRI images of the hand, a neural network has been trained to distinguish seronegative and seropositive rheumatoid arthritis (RA) from psoriatic arthritis (PsA) as well as from each other, according to a study that was presented at the annual meeting of the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis.

In the work so far, the neural network was correct about 70% of the time in the absence of any further clinical analyses, according to David Simon, MD, a rheumatologist in the department of internal medicine at Friedrich-Alexander University, Erlangen, Germany.

Ted Bosworth/MDedge News

Previous to this work, “there has been no study that has exclusively used hand MRI data and deep learning without requiring further expert input for the classification of arthritides,” Dr. Simon said.

In fact, when demographic and clinical data were added, there was no improvement in the performance of patient classification relative to the deep learning classification alone, according to the data presented by Dr. Simon.

The images were evaluated with residual neural networks (ResNet), which represents a sophisticated form of deep learning to facilitate the flow of information across the network layers as they form to improve accuracy in their ability to distinguish one form of disease from the other. The training was performed on images from the T1 coronal, T2 corona1, T1 coronal fat suppressed with contrast, T1 axial fat suppressed with contrast, and T2 fat suppressed axial sequences.

The study included hand MRI scans from 135 patients with seronegative RA, 190 with seropositive RA, 177 with PsA, and 147 with psoriasis. The performance was judged on the basis of area under the receiver operating characteristics curve (AUROC) with and without input of clinical characteristics. Patients who had psoriasis without clinical arthritis were included as a control population.

The AUROC for accuracy was 75% for seropositive RA relative to PsA, 74% for seronegative RA relative to PsA, and 67% for seropositive relative to seronegative RA. Of the patients who had psoriasis without arthritis, 98% were classified as PsA and 2% as RA.

Subsequent to the classification of the patients with psoriasis, 14 of the 147 (9.5%) have developed PsA so far over a relatively short follow-up. All of these were among those identified as PsA by neural network evaluation of the hand MRIs.

This suggests that “a PsA-like pattern may be present early in the course of psoriatic disease,” Dr. Simon said.

In the groups with joint disease, who had mean ages ranging from 56 to 65, the mean disease durations were 2.6 years for those with seropositive RA, 1.3 years for those with seronegative RA, and 0.8 years for those with PsA. The patients with psoriasis were younger (mean age, 40.5 years) but had a longer disease duration (mean 4.2 years).

All of the MRI sequences were relevant for classification, but contrast did not appear to help with accuracy.

“If the images with contrast enhancement were deleted, the loss of performance was only marginal,” Dr. Simon reported.

The accuracy of neural networks increases with data, making it likely that further refinements in methodology will lead to a greater degree of accuracy, according to Dr. Simon. While the methodology is not yet ready for routine use in the clinic, the study demonstrates that neural network analysis of hand MRI to distinguish forms of arthritis “is possible.” Further studies are planned toward the goal of creating a viable clinical tool.

“Of course, if we could create an accurate tool with ultrasound, this would be even more practical,” said Dr. Simon, recognizing the value of an office tool, but he cautioned that this would be far more challenging.

“The precision of MRI is an important factor for effective neural network training,” he said.

Utility: ‘In challenging cases if the accuracy improves’?

A viable method for objectively and rapidly distinguishing inflammatory joint diseases, particularly in patients with an ambiguous clinical presentation, is an unmet need, according to Philip J. Mease, MD, director of rheumatology research at Swedish Medical Center, Seattle.

Although the data presented are promising, Dr. Mease said in an interview that he believes there is a fair amount of work to be done before imaging analysis based on deep learning makes its way into routine clinical care. He is also hoping for methods to distinguish RA from PsA that are easier and less expensive, such as serum biomarkers. However, he agreed that a MRI-based tool could be useful when differentiating disease that is challenging.

“MRI is an expensive way for routine classification of disease, but this approach could be useful in challenging cases if the accuracy improves,” he said.

Meanwhile, other clinical researchers might want to test the principle. “You can try it,” said Dr. Simon, who reported that his team has made the methodology publicly available.

Dr. Simon reported no conflicts of interest. Dr. Mease reported financial relationships with more than 10 pharmaceutical companies, most of which make products used for the treatment of inflammatory joint diseases.