Universal Well-Being Assessment Associated With Increased Resident Utilization of Mental Health Resources and Decrease in Professionalism Breaches Open Access

Improving resident physician well-being continues to be an area of focus for residency programs and accreditation bodies alike. Literature has showed that the majority of resident physicians do not seek out mental health care despite a self-reported need for treatment.^1,2  Barriers to seeking care continue to be time constraints, confidentiality, cost, and stigma.^1,2  Effective ways of supporting resident self-care are essential to building a sustainable and successful physician workforce for the future.

In 2018, we published “Implementing a Universal Well-Being Assessment to Mitigate Barriers to Resident Utilization of Mental Health Resources.”³  This article illustrated the benefits of our opt-out mental health program through our Faculty Staff Assistance Program (FSAP). Internal medicine (IM) residents indicated that they felt the program was a positive addition to the well-being curriculum. They also felt it was convenient and decreased stigma of seeking out mental health care. When asked how likely they would be to return to FSAP if they needed assistance, residents indicated they would now be more likely to do so.³  At the time of publication, we had no objective data that residents would self-refer back to FSAP if dealing with depression, anxiety, burnout, life stressors, etc. As a follow-up to our initial study, we examined the rate of resident-initiated visits back to FSAP before and after initiation of the universal well-being FSAP program. We hypothesized that the FSAP program would lead to more resident-initiated visits.

Physician burnout and mental illness have both been tied to professionalism deficits.^4,5  Therefore, we wanted to further investigate if there was any measurable association between utilization of FSAP for residents to better manage burnout and maintain mental health and professionalism deficits. We measured the number of program-mandated visits (required visits due to unprofessional or disruptive behavior in the workplace) and incidences of anonymously reported professionalism breaches before and after the universal well-being FSAP program was initiated. We also looked at correlation with resident compassion satisfaction (CS) and burnout scores using the Professional Quality of Life Scale version 5 (ProQOL5)⁶  over time. We hypothesized that residents would have fewer professionalism breaches and fewer program-mandated visits if they were maintaining adequate mental health self-care and minimizing burnout by utilizing FSAP services.

The universal well-being program at FSAP was initiated at the beginning of the 2015–2016 academic year. Participants were postgraduate year (PGY) 1 and PGY-2 IM residents at West Virginia University, which is a medium-size academic program. The Institutional Review Board reviewed the study and granted exemption based on educational research.

Our previously published educational innovation was composed of universally scheduled well-being visits to FSAP for all residents. The visits were scheduled by the program, and residents could opt-out. PGY-1s were informed of the program at their recruitment visit and during intern orientation. Program administrators and chief residents scheduled all visits and informed the resident of the date and time. The day was considered a “wellness day” and the resident did not have to report to work. No vacation or sick time was charged. PGY-1s were scheduled for 2 visits per year and PGY-2s were scheduled 1 time. PGY-2s were not scheduled until the 2016–2017 academic year. PGY-3 residents were not universally scheduled. The FSAP well-being assessment consisted of a visit with a licensed therapist who was familiar with treating residents. Residents completed the Workplace Outcome Suite, the Adverse Childhood Experience Questionnaire, the Beck Depression Inventory, and the Beck Anxiety Inventory prior to their visit. Data from the completed questionnaires were reviewed with the resident during the visit to aid in assessing and maintaining their self-care and work-life balance. Visit information was kept confidential including all clinical documentation. FSAP maintains a separate health record outside the institution's electronic medical record. Residents had the option of continuing to self-schedule regular or as-needed visits on an unlimited basis at no personal cost. No sick days or vacation days were used for any return visits.³  Program-mandated visits also occurred prior to and after the initiation of the universal well-being FSAP program. Program-mandated visits were required by program leadership because of disruptive or unprofessional behavior in the work environment or because of obvious resident distress.

We measured the proportion of universally scheduled visits actually attended by the residents for academic years 2015–2016 through 2018–2019.

To objectively assess resident-initiated visits, we obtained all visit counts from FSAP for IM residents from academic years 2014–2015 through 2018–2019. By 2017, all residents in the program had a chance to participate unless they opted out. We compared the number of universally scheduled initial and return visits, resident-initiated return visits, and program-mandated initial visits. Program-mandated visits may have been at FSAP or at alternative sites. Follow-up visits that occurred as a result of a program-mandated initial visit were not counted because data were not obtainable if care was provided outside of FSAP. No identifying data were provided to the program.

In order to assess for association with incidences of unprofessional resident behavior, we accessed our institution's confidential graduate medical education (GME) professionalism online reporting system. Reports can be filed anonymously by anyone in the institution, including nurses, support staff, other residents, and faculty. A narrative of the incident is then sent to program leadership to address. We looked only at the number attributed to the IM program and total numbers of resident reports. Narrative reports, resident names, and resident programs (excluding IM) were not analyzed to maintain confidentiality. We compared the proportion of professionalism reports attributed to IM residents from calendar years 2014–2019 to see if there was any correlation with the initiation of the universal well-being FSAP program.

In order to monitor overall burnout and compassion satisfaction (CS) within the program over time, residents voluntarily and anonymously completed the ProQOL5.⁶  Surveying with the ProQOL⁶  began in spring 2017.

Numbers of universally scheduled visits that residents actually attended were counted for academic years 2015–2019 and plotted graphically. The percentages of professionalism reports attributed to IM residents were calculated. Total reports and IM-attributed reports were plotted graphically along with the percentage attributed to IM residents for the calendar years 2014–2019. A 2-sided Mann-Kendall trend test was applied to the resident-initiated and program-mandated visits as well as the percent of reports attributed to IM residents. Sen's slope was used to estimate the slope of the trends. Mean ProQOL⁶  scores for CS and burnout were plotted graphically from spring 2017 through winter 2020. A Student's t test was used to compare mean burnout and CS scores from spring 2017 and winter 2020.

Data for attendance rates at the universally scheduled well-being FSAP visits for each academic year are illustrated in Figure 1. The overall proportion of universally scheduled FSAP visits that were not attended (opted-out) by residents was 8 out of 239 (3.3%). It is unknown if the same PGY-1 residents who did not attend the universally scheduled initial visit also did not attend the follow-up visit that same year or if they did or did not attend their subsequent PGY-2 visit.

Resident-initiated visits, all of which were return visits, significantly increased over time (slope = 6.5; P = .027; 95% CI [1.0, 8.0]). Program-mandated initial visits significantly declined over time (slope= -2.4; P = .027; 95% CI [-3.0, -1.0]). Graphical data is displayed in Figure 2.

Percentage of professionalism reports attributed to IM residents decreased significantly over time. (slope = -5.7%; P = .024; 95% CI [-11.6%, -0.6%]). There was no statistically significant change in the number of non-IM professionalism reports (slope = < 0.01; P = .71; 95% CI [-0.01, 0.02]) Graphical data for each year are displayed in Figure 3.

The ProQOL5⁶ was administered 7 times from the spring of 2017 until the winter of 2020. Response rates are noted in the online supplementary data. Mean CS and burnout scores showed no statistical difference between spring 2017 and winter 2020 using a 2-tailed Student's t test (P = .88 and P = .39, respectively). Mean burnout scores ranged from 18.5 to 22.0 (≤ 22 low burnout, 23–41 average burnout, ≥ 42 high burnout) during each measured interval (provided as online supplementary data). Mean CS scores ranged from 38.7 to 42.0 (≤ 22 = low CS, 23–41 average CS, ≥ 42 high CS; provided as online supplementary data).

Universal well-being visits using an opt-out approach has continued to be successful in terms of resident utilization of the available mental health services. Almost all residents attended their universally scheduled FSAP appointments. Residents opted out of only 3.3% of universally scheduled visits over the course of the last 4 years. A similar universal intervention utilizing an institutional Employee Assistance Program (EAP) at a different institution also found successful outcomes in terms of resident attendance and mitigation of barriers.⁷ 

Residents had indicated in a previously published survey that they would be more likely to return to FSAP if they needed assistance with depression, anxiety, burnout, or other life stressors.³  This new data objectively show that resident-initiated visits to FSAP significantly increased. We extrapolate that universal well-being visits using an opt-out method results in higher rates of resident utilization of mental health resources by removing barriers such as stigma, scheduling conflicts, confidentiality concerns, and cost. In their commentary, Batra et al cited the success of opt-out strategies throughout the health care system and have called for more institutions to implement similar programs in order to more accurately measure the impact on physician utilization of mental health resources.⁸  Our data show that in our small resident population it has had a major effect.

The number of program-mandated visits declined as the number of resident-initiated visits increased. Program-mandated visits usually were part of a remediation plan for disruptive or unprofessional behavior in the workplace. Since mental health factors often contributed to the need for resident professionalism remediation,⁹  the fewer program mandated visits may be because residents are more freely accessing FSAP and are better coping with issues like burnout, life stressors, and mental illness. As a result, there are fewer episodes of obvious decompensation that would require a mandatory referral by program leadership.

One of the reasons for continued attention to physician well-being is the association between burnout and outcomes such as professionalism.^4,5  Our study used data from our institution's anonymous reporting system for unprofessional resident behavior. After implementation of our universal well-being FSAP program, we noted a sustained decrease in the proportion of professionalism reports attributed to our program's residents. This could be due in part to increased utilization of FSAP by the residents, resulting in better mental health self-care and fewer episodes of unprofessional behavior in the workplace.

This study was completed at a single residency program, which limits its generalizability to other settings. Other programs at our institution have implemented this model, and we hope to collect their data for future information on generalizability. Because residents may have sought follow-up care elsewhere, we may have missed other mental health visits, which may have resulted in underestimation of self-initiated mental health visits. We also are unaware of the number of residents who had mental health appointments already in place prior to the universally scheduled FSAP visits. If residents opted out, we had no way of knowing the reason because of the anonymity of the study. Since we discuss the FSAP program with all residency applicants who interview at our program, we may preferentially attract residents who already place higher emphasis on their own well-being than the general population of residents.

Tardiness, inappropriate or late medical documentation, and deficiencies in administrative tasks were not included as a measure of professionalism in this study, and we do not know if there was any associated change with the initiation of the universal well-being FSAP program. The decrease over time with professionalism breaches and implementation of the universal well-being FSAP program establishes correlation but not causation. It is unknown if other factors may have played into the decrease in proportion of professionalism breaches. We also do not have baseline scores of resident burnout and CS to compare to scores after the program was initiated. We measured CS and burnout 2 to 3 times yearly via the ProQOL5.⁶ CS scores have remained average to high, and burnout scores have remained low⁶  since implementation of the FSAP program but have not showed a statistically significant change. Our response rate for the ProQOL5⁶ was also low with an average of 27%. Residents who have lower levels of burnout and higher CS may also be more likely to complete the survey, thereby biasing the results toward the preferred scores.

Despite limitations, our data do support the hypothesis that a universal well-being program using an opt-out approach increases resident utilization of the services. Most institutions have programs like FSAP or EAP that could be utilized in a similar manner. The Accreditation Council for Graduate Medical Education professionalism competency includes “maintaining emotional, physical, and mental health, and pursuing continual personal and professional growth.”¹⁰  Training programs should ensure that tools are in place to teach residents effective and sustainable self-care. Destigmatizing and removing barriers to mental health care is the first step. Our study shows that utilizing an opt-out system increased utilization of the available resource. Additionally, by treating mental illness and burnout, residents might have been less likely to have instances of unprofessional behavior in the workplace.

Our findings demonstrate that a universal well-being FSAP program with an opt-out strategy increased resident utilization of mental health resources and was associated with fewer professionalism breaches in the workplace.