Abstract
Keywords
“At a single academic center, LM was integrated at three levels of resident well-being efforts, suggesting that LM can be integrated into residency well-being efforts at multiple levels, and that doing so is feasible in routine training environments.”
Introduction
Burnout is defined as a work-related stress syndrome that manifests as increased energy depletion, emotional exhaustion, depersonalization, and decreased personal accomplishment.1,2 Burnout is commonly seen in the medical community and is especially prevalent among resident doctors and attending physicians in their first few years of training.3,4 Residents report lower levels of job autonomy, and this is associated with higher levels of burnout compared to attendings/specialists. 5 This problem negatively affects not only healthcare workers’ health and well-being but also patient care outcomes, 6 highlighting the need for consistent, program-relevant approaches to monitoring well-being.3-6
A systems lens for clinician well-being is recommended by national bodies, and U.S. graduate medical education standards emphasize psychological, emotional, and physical well-being as essential to training.7,8 In this manuscript, we use well-being in the ACGME sense (encompassing psychological, emotional, and physical domains that are essential for competent, caring, and resilient physicians) to include those domains. 8 Resilience refers to adaptive responses to stress and adversity and is relevant to resident training environments.9,10 Validated screening instruments exist for resident distress and well-being, including the Resident and Fellow Well-Being Index (RF-WBI), which can identify residents in distress.11,12
Lifestyle medicine provides a behavioral foundation relevant to both individuals and systems. Consensus statements describe 6 core pillars: healthy eating, regular physical activity, restorative sleep, stress management, positive social connection, and avoidance of risky substances.10,13-15 Organization-directed strategies are a key part of reducing clinician distress, with meta-analytic evidence supporting system-level changes. 16
Accordingly, we present a three-level framework for LM-aligned well-being efforts at a single academic center: (A) institution-wide programming that incorporates LM pillars across residency programs, (B) program-level systems changes with pragmatic outcomes such as on-time graduation and attrition, and (C) an individual Plan of Action for Resilience paired with a validated resident well-being index (RF-WBI). Figure 1 summarizes the alignment across A/B/C. Three-level lifestyle medicine framework.
At Loma Linda University, LM education and protected time were introduced at the program level while residents engaged in individualized Plan of Action for Resilience (PAR) planning linked to a brief validated well-being index, allowing coordinated action across institutional, program, and individual levels.10,11,14
Objectives
(1) Summarize the presence and content of residency well-being activities using LM pillars across programs (A). (2) Examine program outcomes (on-time graduation, late graduation, attrition) before vs after systems changes that included LM education and protected time (B). (3) Evaluate feasibility and resident feedback on an LM-aligned PAR and compare RF-WBI distress scores among residents who engaged with PAR follow-up vs those who did not (C).
11
Methods
Overall Design and Oversight
This single-institution project comprises 3 complementary implementations at the institutional, program, and individual levels within an academic medical center. The work was conducted as a coordinated evaluation of resident well-being efforts organized around LM pillars. Primary outcomes by level were: (A) presence and content of well-being activities using LM pillars across residency programs; (B) on-time graduation, late graduation, and attrition before vs after systems changes that included LM education and protected time; and (C) resident distress measured with the RF-WBI alongside usability and uptake of a PAR.
Institution-Wide Programming (CWR/Coordinator Reports)
We requested information from chief wellness residents (CWRs) and program coordinators regarding each program’s well-being activities during the academic year. The templated query asked whether activities were present, their frequency and format (events/retreats, peer support, social gatherings, resource dissemination, check-ins), and which LM pillars were used. Responses were collated descriptively across residency programs.
Program-Level Outcomes (Retention and Graduation)
We conducted a retrospective analysis of on-time graduation, late graduation, and attrition in the former combined Family-Preventive Medicine program (2006-2022) and its successor 1-year “Plus one” Preventive Medicine fellowship (2022-2024). Residency administrative records provided counts by matriculating cohort year. The longer narrative on program evolution is provided in Appendix A.
Definitions: “on-time graduation” was completion by the expected date; “late graduation” was completion after the expected date; “drop-out” was leaving without graduating. Outcomes were compared for cohorts matriculating before systems changes (2006-2014) vs after systems changes (2015-2023). Systems changes included scheduled LM didactics and protected non-clinical time for LM implementation. We compared proportions using two-sided chi-square tests with α = 0.05. Missing outcome data were rare; when present, analyses used available cases.
Individual-Level PAR With RF-WBI
PAR Tool and Rollout
The PAR is a one-page (Figure 2), LM-aligned action plan co-created by faculty and residents after a brief well-being session, modeled visually after condition-specific action plans to encourage early recognition and escalation steps. The tool prompts users to list early warning signs, set LM-pillar SMART goals, identify accountability partners, and outline contingency steps. The PAR was distributed with completion instructions during scheduled sessions. The longer background on PAR rollout is provided in Appendix A. Plan of action for resilience (PAR) template.
RF-WBI Measurement
Residents completed the RF-WBI at baseline and were invited to complete follow-up at 6-12 months. Surveys were administered via Qualtrics with a short, resident-generated code to enable anonymous matching across time points. Higher RF-WBI scores (>3) indicate greater distress; the index has evidence for identifying residents at risk for distress that may affect quality of care.11,12
Post-PAR Survey
Immediately after the PAR session, residents completed a brief survey assessing usability, likelihood of future use and recommendation, and confidence in recognizing and addressing signs of distress in oneself or others. A 6-12-month follow-up survey attempted longitudinal code-matched reassessment of RF-WBI and PAR use.
Analysis
We summarized uptake and feedback descriptively. RF-WBI scores were compared between residents who completed PAR-linked follow-up and those who did not using two-sided unpaired t-tests (α = 0.05). No adjustments were made for multiple comparisons.
Results
Overview
Three-Level Lifestyle Medicine Framework and Measures.
Institution-Wide Programming
Of 34 residency programs contacted, 31 responded (91%). The most frequently reported activity formats were wellness events and retreats (49%), support activities (36%), social gatherings (26%), and resource dissemination (26%), with check-ins with program directors (23%) also reported. All 31 responding programs reported using at least one LM pillar. The most used pillars were social connection (31%), stress management (25%), and physical activity (17%). Sleep (17%), nutrition (11%), and avoiding risky substances (4%) were less commonly reported.
Residency programs most often used monthly interventions (38%), as compared to quarterly (25%), weekly (22%) or annual (15%) schedules. 24% of lifestyle interventions were implemented in didactic sessions, while the other 76% were conducted informally. Examples included (direct quotations): • Each year, we also have 1 day near the end of the year that is a “Resident Skip Day.” • As a program, we participate in a group activity for our residency members to spend the day together. Prior events have included going to an arcade, an amusement park, watching movies, etc. • Annual resident retreat when residents spend 1 weekend day as a group with our PD to do team-building activities and discuss program goals/strategies to study during residency. • Department holiday party for the entire staff to celebrate (including residents and their families). • Small stipends (separate from GME) are allotted to each resident that can be spent on travel for conferences, textbooks, test prep, or other academic purchases. • Each resident has individual monthly meetings 1-on-1 with our PD to discuss anything—work, family life, future goals, career planning, etc. • Our department has a monthly birthday celebration with cake for all staff members who have birthdays that month.
Program-Level Outcomes
Pre-implementation (matriculating 2006-2014) vs post-implementation (matriculating 2015-2023), on-time graduation increased from 47% to 77%, while attrition decreased from 35% to 10%. Late graduation was 18% pre-implementation and 14% post-implementation. The decrease in drop-out vs graduation of any type reached statistical significance (chi-square P < .03).
Individual-Level PAR With RF-WBI
The PAR was distributed to 619 resident physicians across 48 programs. At baseline, 75 residents completed the RF-WBI and 42 completed the immediate PAR survey. Residents who completed PAR-linked follow-up reported lower mean RF-WBI distress, 1.2 compared with those without PAR follow-up 2.9 (unpaired t-test P < .004). However, only six of the initial participants completed the 6-month follow-up survey, and none reported identifier codes which could be linked to the original surveys, thereby limiting the ability to analyze longitudinal changes in well-being metrics over time. Qualitative feedback indicated perceived usefulness of PAR and suggested providing an online editable version, digital version (application) and/or reminders to support sustained engagement. In descriptive reporting, 1 resident with the highest RF-WBI distress score (7.0) expressed dissatisfaction with workload and uncertainty about corrective steps; comments suggested that a clearer escalation pathway and scheduled check-ins could improve follow-through.
Discussion
Principal Findings
At a single academic center, LM was integrated at 3 levels of resident well-being efforts, suggesting that LM can be integrated into residency well-being efforts at multiple levels, and that doing so is feasible in routine training environments. Institution-wide, most programs already used LM pillars, especially social connection, stress management, and physical activity, consistent with accreditation guidance that emphasizes psychological, emotional, and physical well-being as essential in training (ACGME). 8 Program-level systems changes that included LM education and protected time coincided with higher on-time graduation and lower attrition; we report these as program outcomes, not substitutes for burnout measurement. Individual-level action planning with a PAR, paired with the RF-WBI, was associated with lower distress among residents who completed follow-up. The RF-WBI has validation evidence for identifying residents in distress that may affect quality of care,11,12 making it a practical screening and monitoring tool in training settings.
Comparison With Prior Literature
The overall pattern is consistent with a systems lens on clinician well-being advocated by national bodies and reviews.7,8 Organization-directed strategies tend to show greater impact on clinician burnout than individual-only approaches in controlled studies and meta-analyses, 16 supporting our focus on program- and institution-level design alongside individual tools. Linking system design to downstream outcomes aligns with reviews summarizing contributors, consequences, and solutions for physician burnout 9 and LM-related approaches to supporting clinician well-being.10,13,14,17 Lifestyle behaviors themselves have supportive evidence bases: for example, multiple reviews report that physical activity is often associated with a lower risk of burnout in health care workers, particularly for emotional exhaustion and depersonalization, 18 and sleep health is an important and modifiable factor in physician distress. 19
One study reveals residents reporting high levels of burnout also reporting improvement from providing lifestyle medicine care to patients, even when they did not report similar improvements from lifestyle curriculum alone. This suggests that clinical LM exposure may have a different impact on different groups of trainees and may be more impactful than LM teaching at reducing burnout after it develops. Being able to use the knowledge of LM for the benefit of patients appears to be as beneficial at preventing burnout as LM teaching. 17 This provides motivation for programs to consider clinical LM experiences for trainees to help prevent or reduce resident burnout.
Interpretation of Program Outcomes
Improvements in retention and timely graduation are institutionally meaningful outcomes that align with learning-environment quality. They are not proxies for burnout, but their movement alongside system redesign is compatible with literature linking burnout and dissatisfaction to changes in professional work effort and organizational performance.20-22 Our findings therefore complement, rather than replace, validated measures of resident distress. For individual monitoring and early action, brief instruments such as the RF-WBI are practical in GME settings and can be paired with a simple plan like the PAR to translate results into specific, accountable steps like sleep routines, activity goals, stress-management skills, and social support. 11
Implications for Programs
A multilevel design allows institutions to track program-relevant outcomes like retention and on-time graduation while simultaneously using a validated well-being instrument to monitor distress and to trigger supportive action planning. First, programs can align wellness infrastructure with LM pillars that many already use, while ensuring protected time and operational support for participation. Second, institutions can monitor program outcomes such as on-time graduation and attrition alongside validated, brief well-being measures. Third, simple tools like PAR can help translate screening into action by encouraging early recognition, SMART goals, escalation steps, and accountability partners, which mirrors clinical “action-plan” logic and reflects LM competencies. 15
Limitations
This was a single-institution evaluation and not a randomized trial, meaning that causal inference is limited. For the program-level analysis, retention and graduation were not intended as burnout measures, they were analyzed as pragmatic program outcomes. In component C, no participants reported identifier codes that could be linked to the original surveys, thereby limiting the ability to analyze longitudinal changes. Self-report may introduce response bias, and PAR uptake varied across programs.
Future Directions
Building on these observational results, multi-site, prospective studies could test whether layered LM interventions reduce distress and improve program outcomes, using cluster-level designs that compare different intensities of organization-directed support. Trials could also evaluate digital PAR delivery and automated prompts, both of which were requested by residents in feedback.
Practical takeaways: (1) Build on existing LM pillars already present in program activities. (2) Schedule protected time and operational supports so residents can participate meaningfully. (3) Pair brief screening (RF-WBI) with PAR to translate results into specific actions and accountability.
Conclusion
An LM approach implemented across institutional, program, and individual levels was feasible and was associated with favorable outcomes at each level: at the institution level (A), programs reported widespread use of LM pillars; at the program level (B), on-time graduation increased and attrition decreased after systems changes that included LM education and protected time; at the individual level (C), residents who engaged with the PAR tool and follow-up reported lower RF-WBI distress. These findings align with a systems view of resident well-being and support pairing organization-directed strategies with resident-facing tools.
Programs seeking to enhance resident well-being can combine system-level LM education and protected time with individual action planning (PAR) and validated screening (RF-WBI) and evaluate success using both program outcomes and resident well-being metrics. Future work should strengthen longitudinal follow-up and code-matched tracking and test the model across multiple sites.
Footnotes
Author Contributions
K Studer, CM Miller and R Taylor planned the programs and interventions represented in these cases and implemented the studies. CM Miller analyzed study data, with help from W Laguerre and R Taylor. E Fenwick performed the literature review, which was augmented by W Laguerre, and wrote the initial draft of this article. All authors reviewed drafts and contributed to the final version of this article. The authors would like to thank K Kennedy for her help in outlining and editing this article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project and publication was supported by the Health Resources and Services Administration (HRSA), (Grant/Award Number: 5 T34HP42140‐04‐00) of the U.S. Department of Health and Human Services (HHS) as part of an award totaling $2,916,823 with 50 percent financed with non-governmental sources. The contents are those of the author(s) and do not necessarily represent the official views of, nor an endorsement, by HRSA, HHS or the U.S. Government.
Ethical Considerations
Case study A (institution study) underwent IRB determination review and was found not to be human subjects research (IRB #5240122). Case study B (program study) involved analysis of de-identified resident numbers only and did not require review. Case Study C (PAR) involved Human Studies Research and was approved by the Loma Linda University Health Institutional Review Board #5210444. Written informed consent was obtained from all subjects before the study.
Consent to Participate
Case Study A was determined not to be human subjects research; Case Study B involved counting of residency graduates only and did not require individual consent. Case Study C required and obtained informed consent according to Loma Linda University Health Institutional Review Board Policies.
AI Attestation
We acknowledge the use of language model assistance in the preparation of this manuscript. Specifically, we utilized ChatGPT for outlining, assistance with finding published sources that were reviewed for relevance and accuracy, and suggestions for language improvements.
Disclosures
CM Miller and K Studer serve as consultants with the American College of Lifestyle Medicine (ACLM) receiving a part-time stipend for work with the Lifestyle Medicine Residency Curriculum (LMRC) as Research/Grants Consultant and Director of Evaluation, respectively. K Studer is employed full-time and CM Miller is employed part-time by Loma Linda University Health, which is a co-owner of the LMRC along with the ACLM.
Appendix
This appendix expands on two components referenced in the main text: (1) the system-level evolution of the Loma Linda University Health Education Consortium (LLUH) Preventive Medicine training pathway and related well-being initiatives, and (2) the individual-level rollout and evaluation of the Plan of Action for Resilience (PAR). It summarizes timelines, rationale, logistics, and measurement choices that informed the Methods and Results while keeping the main manuscript concise.