Improving physician well‐being and reducing burnout using a peer‐to‐peer recognition program

Drivers of physician burnout include an intricate interplay between health care organizational structures, societal influences, and individual‐level factors. In the traditional workforce, peer‐to‐peer recognition programs (PRPs) have reduced burnout by building a sense of community and effectively creating a “wellness culture.” We implemented a PRP in an emergency medicine (EM) residency and determined its impact on subjective symptoms of burnout and wellness.


INTRODUC TI ON
Burnout is an occupational phenomenon defined by the International Classification of Disease (ICD-11) as "a syndrome resulting from chronic workplace stress that has not been successfully managed; feeling of energy depletion, an increase in mental distance from one's job, and a reduction in professional efficacy." 1 In 2019, leading health care organizations labeled physician burnout as a public health crisis. 2 Drivers of physician burnout are an intricate interplay between health care organizational structures, societal influences, and individual-level factors. 3 Physicians suffering from burnout tend to make more medical errors and have increased self-reported frequency of providing suboptimal patient care. [4][5][6] In addition, there is an increase rate of suicidal thoughts, depression, and substance abuse with physician burnout. [7][8][9] Among all medical specialties, emergency medicine (EM) has one of the highest rates of physician burnout. 10 According to the 2022 Medscape Lifestyle Report, nearly 60% of all EM physicians endorsed some features of burnout. 10 In 2017, the National EM Resident Wellness Survey study showed a prevalence of burnout as high as 76.1% for EM residents. 11 More recent studies have shown that burnout is seen as early as in medical trainees, medical students, and residents. 9 The Accreditation Council for Graduate Medical Education (ACGME) in response to these data has initiated requirements for resident wellness initiatives.
Interventions that can decrease burnout are imperative for the future of health care. 12 Fortunately, burnout is reversible and even preventable. A systematic review showed that interventions that worked on improving communication skills and interdisciplinary teamwork led to a reduction in self-reported burnout. 13 Inexpensive and easily implementable programs such as face-to-face workshops with colleagues, meditation seminars with experts, and open forums on well-being/self-care have also shown to build resilience and increase long-term mental health among staff. 13,14 Studies have shown that residency programs have been implementing different approaches to improve resident burnout, but there is insufficient evidence that there is any efficacy in their initiatives. 15 In the traditional workforce, peer-to-peer recognition programs (PRP) have shown great success in improving employee retention and increasing work place engagement. 16,17 Since then, there have been extensive efforts to extend this intervention into the hospital setting among various departments such as nursing and pharmacy. [17][18][19][20] These studies surmise that recognition programs elevate the quality of patient care by empowering staff, encouraging leadership qualities, and rewarding good patient care. Preliminary studies have shown that this "wellness culture" can also be used to improve physician burnout. 21 Gribben et al. 21 created their own "golden ticket project" in which pediatric residents can award another coresident a "golden ticket" for any behavior they were grateful for. They concluded that a supportive working environment can be protective against physician burnout. 21 A vital contributing factor to EM physician burnout has been the lack of respect from administrators/employers, colleagues, and staff. 21 A PRP would implement continuous feedback throughout the year, both from residents/colleagues and from the attendings/leadership. Unlike workshops that are only for a few hours to a day in time, a PRP would be daily affirmations that all residents could participate in and led in.
The focus of this study was to evaluate if the use of a PRP can reduce physician burnout and improve subjective well-being within an EM residency program. The PRP's goal was to improve physician wellness by having others value and acknowledge their hard work, to encourage gratitude, camaraderie, and praise. We hypothesized that the addition of the PRP to the wellness initiatives in our EM residency program would result in a better "wellness culture." Our goal was to evaluate if the PRP would improve the EM residents' professional fulfillment index (PFI) and residency wellness.

Study design and population
This was a prospective evaluation of an intervention conducted in a single EM residency within a large urban academic program over a 6-month period. The EM residency consisted of 84 EM residents, 21 residents per academic postgraduate year, with 50% identified as female.
The institution's institutional review board deemed this research protocol exempt. No outside funding was used for this project.
A subscription to Bonusly-a PRP-was purchased by the EM residency administration. This platform allowed for residents and attending physicians to acknowledge individual residents for their achievements through public praise and the granting of points that can be redeemed for rewards. All 84 EM resident physicians elected to participate in the study and received access to their own Bonusly accounts. Three additional shared accounts were also created for attending physicians across the three hospital sites in which the residents train. All EM residents were able to give out acknowledgments and praises to their fellow residents only, not attendings. The EM attending accounts were only able to give out acknowledgments but could not receive any themselves or collect tokens for rewards. On the shared feed, residents can see all the acknowledgments that are given across the department. They are able to see who is specifically giving the praises, who is receiving them, and what the acknowledgment is for. The tokens are given to the resident receiving the acknowledgment, no tokens are given for giving an acknowledgment.
Eventually tokens that were received on the Bonusly platform could be redeemed for gift cards or customized experiences. The gift cards were purchased by the EM residency administration and the personalized prizes were voluntarily offered by residents and attendings (i.e., a cooked dinner, a jog in the park together, handcrafted goods).
Rewards were redeemed by residents on a rolling basis throughout the 6 months. A more comprehensive list of the rewards redeemed can be found in the supplement material.
All 84 EM residents of the program were sent a voluntary anonymized pre-PRP survey. The first page of the survey explained to the physician participants that participation was voluntary and that completing and submitting the survey implied consent to use the data. Regardless of survey response, all 84 residents received access to the PRP. After 6 months, an anonymized post-PRP survey was distributed to the same EM residents.

Survey content and administration
The pre-PRP survey had three sections. The first section had three questions that collected demographic information. The second section was six questions specifically related to residency wellness. The third section was the complete 16-question PFI. Respondents were asked to indicate their opinions using a 5-point Likert scale: "not at all true" to "completely true" for residency wellness items and professional fulfillments items and "not at all" to "completely" for work exhaustion and interpersonal disengagement items. Each item is  We chose to use the Stanford PFI as part of the measurement instrument to evaluate and assess professional well-being. We specifically chose the Stanford PFI because it is an instrument that was specifically designed for physicians to investigate professional fulfillment, work exhaustion, and interpersonal disengagement. We felt that this was the best inventory to assess the intervention of a PRP. We had considered other inventories such as the MSI-HSS, Oldenburg Inventory, mini-Z, and Copenhagen Burnout Inventory but felt that the dimensions measured in the Stanford PFI more closely evaluated our measured outcomes.
In addition to the PFI, our survey included six questions related to specific residency wellness. These six questions were partially derived by the questionnaire provided by the PRP company used (Bonusly) and, as such, these parameters were speculated to have a positive change after the intervention as they were already validated by the company's research to have substantial improvements in behavior. These questions were not prescreened for effectiveness prior to the adaptation of our survey. We felt that the proposed questions directly related to inclusion, motivation, community, and positive recognition-all parameters that sound seemingly be affected with the implementation of the PRP program. For comparison, the preset company questions provided by Bonusly can be found in the supplemental material.

Key outcome measurements
The primary outcome of the study was the measure of residency wellness and PFI within the EM residency program. The PFI tool is a validated tool to assess not only burnout but professional fulfillment as well. The PFI is a short, validated, 16-item survey that is relevant to physicians and incorporates both positive aspects, i.e., professional fulfillment as well as negative aspects, i.e., work exhaustion and interpersonal disengagement items. 22 Each of the three PFI subscales questions assessed "respondent" well-being over the 2 weeks before taking the survey. Answers were on a 5-point Likert scale score. For the PFI subscale professional fulfillment the higher the score is more favorable, while for work exhaustion and interpersonal disengagement scales higher score less favorable. The secondary outcome was to measure whether a PRP can improve factors that contribute to residency wellness and improve PFI. We will examine if these outcomes were affected by gender or postgraduate year of the residency.

Data analysis
Descriptive data are presented as mean with standard deviation (SD) and percentages with 95% confidence intervals (95% CIs) where appropriate. Items of the residency wellness questions were scored 0 to 4 and their mean score for pre-PRP and post-PRP were determined and compared using a Student t-test. The residency wellness differences were analyzed by subcategories of gender and resident's postgraduate year.
The professional fulfillment score was tabulated on a scale of 0-4 based on averaging of survey responses to the six Stanford PFI professional fulfillment questions. A resultant score of 3 and above is considered to be professionally fulfilled. 22 The work exhaustion score was tabulated on a scale of 0-4 based on the averaging of survey responses to the four Stanford PFI work exhaustion questions. A resultant score of 1.33 and above is considered to be exhausted at work. 22 The interpersonal disengagement was tabulated on a scale of 0-4 based on the averaging of survey responses to the six Stanford PFI interpersonal disengagement questions. A resultant score of 1.33 and above is considered to be interpersonally disengaged. 22 The burnout score was tabulated on a scale of 0-4 based on the averaging of survey responses to the 10 Stanford PFI work exhaustion and interpersonal disengagement questions. A resultant score of 1.33 and above is considered to be burned out. 22 The 95% CI and p-values were calculated. We used SAS (v. 15.1) to perform all analyses.

RE SULTS
The study was conducted from July 2021 to January 2022. Bonusly, a PRP, was implemented into the EM residency department July reported that they would continue using this platform in the future. Table 1 shows the residents' response to usage of the PRP platform.
Although the PGY-2 and females reported giving and receiving higher percentage of recognitions, it was not statistically significant.
Resident physicians reported improvement in two items of the resident wellness questions: "feeling recognized for accomplishments at work"' and "my residency fosters a comfortable and supportive work environment" (see Table 2). When asked whether they "felt recognized for their accomplishments at work" on the pre-PRP survey, 45% of the residents answered that the statement was true, with a significant 18% increase in the post-PRP survey to 63%. Then when asked whether "my residency fostered a comfortable and supportive work environment," 68% of residents answered that the statement was true on the pre-PRP survey and it increased by 17% to 85% in the post-PRP survey. Further analysis stratified by gender on these two measures revealed a larger incremental change in these two variables in female-identifying residents: 70% from 38%, a 32% difference, and 92% from 67%, a 25% difference, respectively.
In addition, female-identifying residents had an improvement This is reflected in their answers to the statements "I feel recognized for my accomplishments at work," "there is a sense of community in my residency program," "my residency fosters a comfortable and supportive work environment," and "my values and the residency's values are aligned." Furthermore, the PGY-2 residents self-reported a decrease in burnout from their work from 76% to 47%, −29% difference. The PGY-1 residents, those who just started their residency at the beginning of this study, July 1, 2021, had the lowest burnout rate compared to the senior residents: 14% versus 76%, 67%, and 57% incremental years of residency, respectively. There was no significant effect in the Stanford PFI as a result of this PRP intervention (Table 4).

DISCUSS ION
This study found the self-reported burnout rate in our EM residency program was 54% in the start of the study and after 6 months of implementation of the PRP the burnout percentage increased to 56%. After the implementation of the PRP, there was improvement of specific residency wellness items: "feeling recognized for accomplishments at work" and "my residency is a comfortable and supportive work environment." The PFI and the subscales profession fulfillment, work exhaustion, and professional disengagement did not show an improvement with the implementation of the PRP.
Compared with prior studies, our study finds a similarly high frequency of EM physician burnout at 54%. To our knowledge, this is the first prospective evaluation of a peer-to-peer recognition intervention study in an EM residency program. Comparably, we found that the implementation of Bonusly improved residents' selfreported recognition of their accomplishments at work and that it fosters a more comfortable and supportive work environment. Both aspects can be protective against burnout in residency training.
Interestingly, the overall rate of self-reported burnout increased in the postintervention survey at 56%. The most likely driver of this reflection is the increase of burnout among the PGY-1 class from 14% to 56%. This phenomenon appears to start early on for some doctors, 9,23 where the rigors of practicing medicine for the first time can add to stress-related burnout later on.
Interestingly, our study found a significant improvement in wellness contributors in female-identifying residents compared to their and preference in "working with others.". 24 Men were driven more by "financial matters," "altruism," and "power." 24 This could explain why there was a significant increase in reported value alignment, community, and recognition and supportive work environment among our female-identifying residents.
In addition, we found that the percentage of reward redemption was low among the residency. It should be noted that whereas reward redemption remained low by the conclusion of our study at 6 months, it significantly increased in the latter half of the year and exponential increased by the end of the academic year. It is surmised that residents were hoping to accumulate more tokens for higher ticket items. Future studies to elucidate any correlation between reward redemption and burnout would be worthwhile.  Note: Percentage of resident physicians who believe that these statements are moderately, very, and completely true.
each year of training. 25 The residents enter the program had the  This study centered around identifying burnout within an intensive, urban EM residency and showed that a PRP can help build a comfortable work environment and help residents feel recognized for their work, both barriers to burnout. By using a platform that allowed residents and attending physicians to acknowledge achievements through praise on a public forum and the provision of points that can be redeemed as meaningful rewards, we showed that a supportive work environment can be built within a program by easily implementable measures. In just 6 months, the initiative led to improved support from coworkers, a sense of accomplishment at work, and ultimately improved physician well-being. Although more needs to be done, this is a pivotal first step toward reducing physician burnout and is our hope that endeavors such as these will be extended into residency programs across the nation to help build a future culture of wellness within emergency departments. A workplace wellness program resulted in a significantly greater rate of some positive self-reporting but no significant difference in clinical measures after 6 months.

LI M ITATI O N S
There are several limitations for this study. The study took place in a large urban EM residency program that may not be generalizable to other smaller residency programs and to other medical specialties. Since we chose to protect the anonymity of the residents during this study for them to feel comfortable in answering the questions on a sensitive topic, we lost the ability to assess the impact of the intervention on individual residents. Thus, we had to generalize the effects to the greater groups, i.e., the postgraduate year and gender.
Also, the distribution of tokens and redeemable rewards could also be a confounder. However, this phenomenon is deemed to be minis- The use of PFI in longitudinal studies has not been validated, so it is unclear if PFI is a reliable tool that can be used to compare wellness and burnout over time.

CON CLUS IONS
There were no significant effects on the professional fulfillment index scores or self-reported burnout after 6 months. A peer-topeer recognition program initiated among ED residents resulted in improvements in a few factors that drive physician burnout: feeling recognized for accomplishments at work and a comfortable and supportive work environment. A peer-to-peer recognition program initiative may need more than 6 months to see its effects on professional fulfillment index scores and residency wellness items.
Physician's wellness is an important issue and peer-to-peer recognition program may be one initiative that could be added to the ar- supportive work environment in the emergency medicine residency will not be a "one-size fits all" solution, an extended initiative with the addition of a peer-to-peer recognition program may still need to be determined if beneficial.

AUTH O R CO NTR I B UTI O N S
Jenny Chang-study concept and design, acquisition of the data, analysis and interpretation of the data, drafting of the manuscript, critical revision of the manuscript for important intellectual content.
Vinay Saggar-acquisition of the data, drafting of the manuscript,

FU N D I N G I N FO R M ATI O N
The cost of the peer-to-peer recognition program, Bonusly, was paid for by the Jacobi/Montefiore residency administration.

CO N FLI C T O F I NTER E S T S TATEM ENT
The authors declare no conflicts of interest.