Intervention Mostly Flops for Health Workers' Well-Being During the Pandemic

A peer-to-peer support intervention did not improve the well-being of healthcare workers overall during the COVID-19 pandemic, but the program did have a positive effect for those ages 30 and younger, a cluster randomized trial showed.

In an intention-to-treat analysis of over 2,000 healthcare workers from 28 hospitals and federally qualified health centers (FQHCs), there was no overall treatment effect for psychological distress score or post-traumatic stress disorder (PTSD) symptom score among those who participated in Stress First Aid, an evidence-informed peer-to-peer support intervention, compared with usual care, reported Lisa S. Meredith, PhD, of the RAND Corporation in Santa Monica, California, and colleagues.

However, in a post-hoc analysis that looked at treatment effects of the intervention by age group, healthcare workers at FQHCs who were 30 or younger had significantly less psychological distress and PTSD after participation in the intervention. In this age group, the psychological distress score decreased by 4.52 points (P=0.01) and the mean PTSD score decreased by 6.77 points (P=0.04) compared with usual care, they noted in JAMA Network Open.

This study highlighted "the importance of collegiality and mutual support in a cataclysmic crisis," wrote Jason H. Wasfy, MD, MPhil, of Massachusetts General Hospital and Harvard Medical School in Boston, and colleagues in an accompanying editorial. "On the basis of the results of their study, this may be especially meaningful for young healthcare workers who have yet to develop the most personally meaningful coping strategies, clinical confidence, and workplace community and networks that a longer career in healthcare may afford."

Co-author Courtney Gidengil, MD, MPH, of the RAND Corporation in Boston, told MedPage Today that she and her team had "hoped to see an intervention effect, given how desperately it is needed with all that healthcare workers have suffered through. But on the other hand, it's really hard to learn to swim when you're already drowning -- healthcare workers had an impossible burden on their plates."

Meredith told MedPage Today that "alleviating stress and burnout among healthcare workers is a bigger concern than can be solved by this study. While there is potential for SFA [Stress First Aid] trainees to gain coping skills by using the SFA tools, what we really need are bigger systemic solutions to these problems."

Wasfy and colleagues also noted that "burnout is a problem not only because of the suffering of our colleagues and their families, but also because of the repercussions for the healthcare system."

"For instance, clinicians who are burned out are more likely to consider leaving the practice of medicine," they wrote. "In a national survey [from] December 2020, 1 in 5 physicians and 2 in 5 nurses reported intending to leave medical practice within 2 years, and 1 in 3 healthcare workers reported intending to reduce their work hours."

"With the growing concern over healthcare worker shortages, the burnout crisis for healthcare clinicians is already likely contributing to a lack of availability of care for patients," they added.

At the beginning of the study, approximately 36% to 40% of all participants at FQHCs reported burnout, 15% to 18% reported serious psychological distress, and 13% to 20% met PTSD diagnostic criteria. Among workers at hospitals, about 50% reported burnout, 10% to 12% reported serious psychological distress, and 17% to 19% met PTSD diagnostic criteria.

The Stress First Aid intervention was originally developed by the U.S. Navy and Marine Corps to help individuals without any mental health training manage stressful circumstances among themselves and peers. It promotes five key elements of safety, calm, connectedness, self- and community-efficacy, and hope. To attain these goals, the intervention encourages seven core actions: check, coordinate, cover, calm, connect, competence, and confidence.

The trial took place from March 2021 to late July 2022. A total of 28 sites were recruited in pairs -- eight hospital pairs and six FQHCs pairs -- so that they were matched based on size, type, and COVID-19 burden. Each site within a pair was randomized to the Stress First Aid intervention or to usual care.

Among enrolled facilities, 2,077 healthcare workers participated in the study and completed a pre-intervention and post-intervention survey. Of the 1,649 healthcare workers from hospitals and 428 from FQHCs, 862 individuals were included in the intervention arm and 1,215 were included in the usual care arm. Approximately 80% of all participants were women, and most were ages 31 to 50.

The authors noted that at site-level randomization, baseline characteristics were not balanced among individual healthcare workers. For example, healthcare workers at FQHCs were younger and more likely to have been working in healthcare for at least 5 years compared with their counterparts at hospitals. Healthcare workers at FQHCs were also more likely to be Black or Hispanic or Latino/Latina than those at hospitals. Also, across all the sites, more healthcare workers participating in the intervention were administrators, assistants, or technicians, rather than physicians, nurses or nurse practitioners, or physician assistants.

The researchers identified at least one site champion for every 50 healthcare workers, who watched 4 hours of video training and participated in a 2-hour virtual peer-to-peer support training session. The champions then trained their peers at their own organizations. Sites that were randomized to usual care continued to provide any healthcare worker support programs that were already in place.

General psychological distress was measured with the Kessler 6 instrument (range 0-24, with the highest scores indicating higher distress), and PTSD was measured with the PTSD Checklist (possible range 0-80, with the highest scores indicating greater symptom severity). Baseline mean psychological distress score was 5.86, and the baseline mean PTSD score was 16.11.

In addition to the primary outcome of psychological distress and PTSD, secondary outcomes included sleep problems, workplace stress, burnout, resilience, and moral distress. There were no statistically significant differences between the intervention group and the usual care group for these outcomes.

The authors noted several limitations to the study. Implementation of the intervention was occasionally paused due to "enormous clinical demands" on participating healthcare facilities during the pandemic. Training in the intervention may not have been intensive enough, and the follow-up period may have been too short. Also, participants who had an improvement in mental health may have been less likely to complete follow-up surveys.

Comment