A Scoping Review of Physical Activity Interventions for Physician Wellness

Abstract

Training and practicing physicians continue to experience burnout, poor job satisfaction / engagement, high stress, and even depression and suicide. With an understanding of effective and ineffective methods to improve wellness, our team aimed to collect publications with data on the effectiveness of physical activity interventions to address burnout and well-being in physicians. Comprehensive searches for articles related to the impact of exercise on physicians’ well-being were undertaken in March 2022 in PubMed, PsycINFO, and EMBASE. Of 2496 records identified in the databases, the team screened 1,287 and finalized 12 papers for inclusion and review. This study reviews the overall positive benefits of physical activity and exercise interventions in physicians. Despite the wide-ranging utility of an exercise prescription and despite clear evidence that physicians' personal exercise habits affect their patient counseling about exercise, this review shows that (often overburdened) physicians have insufficient physical activity, and are rarely included in studies targeting exercise. Many of the included studies had small sample sizes, leaving possibilities for future research aiming to augment physician wellness.

“Busy physicians have difficulty maintaining a time allotment for exercise, but we should serve as examples to our patients and the public by incorporating healthy movement habits into a daily routine.”

Introduction

Training and practicing physicians continue to experience burnout, poor job satisfaction/engagement, high stress, and even depression and suicide. Doctors should serve as examples of optimal health, but instead they are often worn down physically and mentally by long hours, sleep deprivation, poor work-life balance, and moral distress. ¹ Providers who experience burnout cannot care for patients; they make medical errors but also have difficulty maintaining compassion for complex patients. ² Extensive literature has described burnout, but relatively few publications present convincing data on how to prevent burnout and enhance wellness/flourishing.

A robust and consistent collection of literature supports the assertion that physical activity and exercise enhance mental and physical well-being. ³ Exercise may be as effective as antidepressant medications and appears to have a preventive effect in those predisposed to mental illness. ⁴ Physical activity has a causal connection to longer lifespan and decreased morbidity. ⁵ Developed countries such as the US have concerning and increasing rates of overweight, obesity, and poor cardiorespiratory fitness. ⁶ It appears that physicians and other healthcare professionals have similarly alarming rates of overweight and poor fitness. ⁷ Residents and trainees, though younger than practicing physicians, appear to suffer from the same problems.^8,9 Physicians may underestimate the impact their own health practices can have on patients they treat. ¹⁰

Exercise has the potential to treat at least 26 chronic diseases, though the modern medical paradigm often prioritizes pharmacologic management over lifestyle. ¹¹ Improvement in level of physical activity and cardiorespiratory fitness results in improved quality of life, cardiovascular and all-cause mortality. ¹² Numerous publications describe interventions to reduce or prevent burnout in practicing and resident physicians; ^{13 , 14} a scant but increasing amount of research is devoted to the study of wellness and thriving. With an understanding of effective and ineffective methods to reduce burnout, our team aimed to collect publications with data on the effectiveness of physical activity interventions to address burnout and well-being in physicians, which should translate to better health outcomes for patients.

Methods

Comprehensive searches for articles related to the impact of exercise on physicians’ well-being were undertaken in March 2022 in PubMed, PsycINFO, and EMBASE. The following search string was employed in PubMed: “(Exercise[Mesh] OR Running[Mesh] OR Jogging[Mesh] OR Sports[Mesh] OR Yoga[Mesh] OR Qigong[Mesh] OR ‘Tai Ji’[Mesh] OR ‘Weight Lifting’[Mesh] OR ‘Resistance Training’[Mesh] OR exercise OR ‘physical activity’ OR running OR jogging OR sport* OR athletic* OR yoga OR qigong OR ‘tai ji’ OR ‘tai chi’ OR Pilates OR ‘weightlifting’ OR ‘weight lifting’ OR ‘lifting weight*’ OR ‘weight training’ OR ‘Resistance training’ OR ‘strength training’) AND (Physicians[Mesh] OR physician*[ti] OR doctor*[ti] OR surgeon*[ti] OR pediatrician*[ti] OR Resident*[ti] OR trainee*[ti] OR attending*[ti]) AND (Body Weight[Mesh] OR Body Mass Index[Mesh] OR ‘Body weight’ OR ‘body mass index’ OR BMI OR ‘Blood Pressure’[Mesh] OR Hypertension[Mesh] OR ‘Depressive Disorder’[Mesh] OR ‘Burnout, Psychological’[Mesh] OR ‘Anxiety Disorders’[Mesh] OR ‘blood pressure’ OR hypertension OR burnout OR anxiety OR depression) NOT (Editorial[pt] OR Letter[pt] OR Review[pt] OR Meta-analysis[pt] OR ‘systematic review’[pt]).”

Filters for publication date, language, and human studies were applied to the results set. The Search results were saved in EndNote 20 and deduplicated. The inclusion criteria were articles published in the English language within the previous ten years (since 2012) that investigated the impact of any type of exercise of physical activity on the physical or mental health of physicians. Exclusion criteria were commentaries, previous review articles, articles that did not report outcomes measures, and articles about medical students or healthcare professionals other than physicians.

Deduplicated results were uploaded into Rayyan, and two authors (MH and JS) independently screened the titles and abstracts, with conflicts resolved through discussion. The full texts of potentially relevant articles were then obtained, and the same two authors completed full-text screening. In the event of disagreement, final inclusion decisions were arrived at through discussion.

Results

Of 2496 records identified in the databases, the team screened 1287 and finalized 12 papers for inclusion and review (see Figure 1 PRISMA 2020 flow diagram). Four of the 12 describe research with practicing physicians as subjects; the remaining eight conducted interventions in residents/trainees.

Figure 1.

PRISMA 2020 flow diagram for new systematic reviews which include searches of databases and registers only. Notes. * Consider if feasible to do so reporting the number of records identified from each data base or register searched (rather than the total number across all databases/registers). ** If automation tools were used indicates how many records were excluded by a human and how many were excluded by automation a tools.

Yoga

Five of the thirteen papers describing physical activity interventions in physicians implemented yoga exercises. Four of these five studied effects on residents/trainees.

The intervention for faculty members published by Scheid et al took place at The Brigham and Women’s Hospital (BWH) OBGYN and Pediatric Newborn Medicine Departments, inviting 93 total attending physicians. Of the 18 subjects who participated in the 6-week (one hour of yoga per week) pilot study, 12 completed the post-program assessment and 11 the two-month follow-up. The yoga program protocol called RISE consists of resilience, integration, self-awareness, and engagement. Results showed remarkable improvement in multiple metrics: engagement, stress, resilience, anxiety, and depression. Prevalence of total burnout changed from 50% to 8% and professional fulfillment from 8.3% to 27.3%. Acknowledging the small sample size, findings reached statistical significance. ¹⁵

Taylor et al conducted a pilot study in 31 junior doctors at Royal Prince Alfred Hospital (RPAH), a teaching hospital in Sydney, Australia. ¹⁶ This single-arm randomized controlled trial (RCT) compared 8 subjects who had group fitness classes (active control) with 10 subjects who had weekly one-hour trauma-informed yoga sessions, one 4-hour workshop, and eHealth homework. Both groups had decreased burnout and improved personal accomplishment after the 8-week program. The yoga group had an increase in compassion satisfaction but no between group differences. Compared to control, the yoga group had reduced depersonalization, more flexibility, attended class more (100% vs 45%), and rated the experience more highly. The yoga arm also had a slight improvement in mindful attention, while the arm fitness arm had a decrease. Results showed no intervention effects on multiple physiologic parameters (blood pressure, resting heart rate, oxygen saturation, heart rate variability). ¹⁶

Loewenthal et al published results from a randomized control trial of a yoga intervention in resident physicians at Brigham and Women’s hospital, again using the RISE protocol (n = 38) compared to a waitlist control (n = 18). ¹⁷ Of 2000 residents invited, 56 were enrolled and 44 completed the protocol. Residents on average attended two of the six weekly sessions. Results showed improvements in mindfulness, stress, burnout, and well-being from baseline to end of program; these findings persisted at the two month follow-up. However, reviewing the figures, results showed no between group statistically significant findings, only within group. Mean feasibility of attendance was 28.9 out of 100-point visual analog scale, but respondents rated feasibility of virtual attendance at 69.2 out of 100. ¹⁷

Hoffert et al conducted a study of Yoga in resident physicians at Henry Ford Hospital in Detroit, MI. ¹⁸ The design involved three groups, one mandatory attendance group (one session per week), one voluntary attendance (one session per week), and an unassigned walk-in group (one day per week) available to all 680 trainees on campus. The choice of mandatory and voluntary groups depended on rapidity of response by program directors. The mandatory group (who received protected time for the classes) were all dermatology residents and the voluntary group were all pulmonary/critical care trainees. The required yoga group saw no improvement in the primary outcome, mean Resident Physician Well-Being Index (RPWBI). Authors reported little to no data on the voluntary yoga and walk-in groups, and provided no group comparisons. The potential benefit of 6-month follow-up surveys was somewhat nullified by a declining response rate over time. ¹⁸

Finally, Babbar et al report results from a relatively robust 8-week departmental quality improvement initiative. ¹⁹ All 24 OB/GYN residents and five fellows received activity trackers and had access to weekly yoga classes, nutrition challenges, and physical challenges. Yoga classes occurred during protected didactic time. Nutrition and physical challenges took place independently, with one unique challenge each week. Physical challenges were based on data from the activity trackers. Objective parameters were collected before and after the 8-week intervention. 90% of the subjects attended at least one yoga class, while 68% made it to half or more. 80% participated in one or more nutrition challenges, 60% in one or more physical challenges. Post-program results showed significant improvement in depersonalization (no change in emotional exhaustion or personal accomplishment), anxiety (no significant change in depression or stress), and blood pressure. The blood pressure benefit was more marked in those who attended more than half of the yoga sessions. Survey data showed effects on comradery, appreciation, and motivation. ¹⁹

Structured Exercise

Three of the papers in this review conducted structured exercise interventions. Vuong et al present data from a one-day Addiction and Mental Health Sports Day. ²⁰ At Alberta Health Services in Edmonton Zone, participants engaged in sports events, laughter yoga, family fun, lawn games. Researchers sent a survey to 243 staff and physicians (out of the 522 total individuals who registered for the event), receiving 66 responses. Ninety-five percent of respondents were satisfied with the event overall, with >80% satisfaction for all activities other than laughter yoga. Physicians and staff reported positive effects on physical and mental well-being, improved relationships and sense of community. ²⁰

A 2014 publication from Thompson et al reported improved physician activity levels with the implementation of treadmill desks. ²¹ Seventeen overweight, sedentary Internal Medicine physicians (age 25-70) at the Mayo Clinic participated in a crossover trial using treadmill desks, accelerometers, and exercise counseling. Physicians used the desk when completing administrative work, not while seeing patients, averaging 90 minutes per day. The intervention of the treadmill desk increased calorie expenditure by a statistically significant 197 kcal per day. The desk led to weight loss of mean 1.85 kg, and a 1.9% decrease in body fat percentage. 13 of the 17 physicians reported intention to continue using a treadmill desk if available. ²¹

Finally, Weight et al studied quality of life and burnout in the setting of a team-based, incentivized exercise program. ¹ All 1060 residents and fellows at the Mayo Clinic in Rochester received an invitation to join an elective, 12-week experience using institutional exercise facilities. All residents and fellows were invited to have baseline and three month follow-up measurements of quality of life, burnout, and physical activity. Of the 628 residents and fellows who answered the first survey, 194 (31%) met national guidelines for physical activity. Of respondents, 245 (23%) took part in the exercise program, not differing significantly from nonparticipants. At conclusion of the intervention, participants were significantly more likely to meet the minimum physical activity recommendation (48% vs 23%), performing more moderate aerobic, vigorous aerobic, and strength training than nonparticipants. Program participants also had higher quality of life (75 vs 68) and lower (but not statistically significant) burnout (24% vs 29%). Results showed no difference in subcategories of Maslach burnout score. ¹

Ergonomics

Two of the papers were included due to focus on ergonomic healthy movement of physicians. The first of these, published in 2020 by Allespach et al., enlisted physical therapists to instruct surgical residents and medical students over one academic year. ²² They provided three modules using a perioperative micro-break model. They also collected survey data to document pain, applicability, and effectiveness. Before the intervention, 87% of participants had pain (most commonly low back and neck), with 80% of those reporting pain in the operating room. Following module two, only residents were assessed (18). 65% of them had less physical discomfort after learning the exercises. Of the 15 residents to answer the survey after module three, 93% of them felt that the training would help them physically perform better in the operating room, and 85% of them had less discomfort after exercises. Of note, none of the reductions in pain reached statistical significance. ²²

Markwell et al describe a pilot intervention at one outpatient surgical center in practicing gastrointestinal endoscopists. ²³ Again, physical therapists helped conduct the study, assessing (based on colonoscopy procedure and desk work) eight endoscopists who agreed to participate. The physical therapists developed personalized pain education and wellness exercises. Data showed twenty-two pain sites among five of the eight physicians studied, most commonly back and neck. The physical therapists detected diverse ergonomic inefficiencies and poor movement patterns. Results showed 63% of pain sites reduced in intensity or completely resolved, 32% of pain sites did not change, but 4% increased in intensity. All eight physicians reported that the postural education and procedure room set of recommendations positively impacted their performance. ²³

Technology Trackers and Apps

Several publications in our search described interventions using technology such as activity trackers or apps. Only three included an intervention with measurable outcomes. NeCamp et al sought to provide mHealth interventions over a six-month period on 1565 medical interns. ²⁴ Subjects wore a FitBit Charge 2 device and installed an app specifically designed for the study. The intervention consisted of push notifications randomly targeted toward one of three different metrics: weekly mood (via daily survey), activity (weekly step count), and sleep (weekly sleep time). Results showed an interaction between the previous weeks step count negatively moderating the current weeks step count. Put more simply, when a participant received a notification during a time of low step count, the next measurement was a higher step count. When the participant had a high step count and received the notification, the next measurement was lower. This may simply represent regression to the mean, although this possibility was not mentioned by the authors. ²⁴

Thorndike et al conducted a two-phase interventional study on 104 medicine residents, looking at steps taken (using Fitbit activity trackers) and other laboratory parameters. ² In phase 1 of the study, half of the participants were blinded to step count. They found no significant difference between groups during phase 1 on step count for compliance. In phase 2, all residents unblinded their devices and researchers instructed them to compete by residency class for the highest average number of steps per week. The phase 2 competition intervention led to more average steps per day than during phase 1 only for the control group. Mean systolic blood pressure and high-density lipoprotein (HDL) measurements showed improvement at the end of the study compared to baseline. Interestingly, authors noted availability of an on-site fitness center, one, hour long personal training sessions per week at no cost, and a healthy meal substituted once a week for their daily catered resident lunches. ²

In a prospective crossover study in resident physicians, Yeung et al enrolled 86 of a possible 104 residents to wear a FitBit activity tracker. ²⁵ 49% of the residents had a body mass index greater than 25 kg/m². All residents were blinded to step count during the first four weeks, then unblinded for the second four-week period, with the additional invitation to join an activity tracking group. The median number of steps among participants during the blinded period (7260) increased to 8266 during the unblinded period. The 26 residents who participated in the activity tracking group had higher median steps per day entire compliance with the device. ²⁵

Discussion

This review presents insight into the effectiveness of physical activity interventions to address burnout and well-being in physicians. Busy physicians have difficulty maintaining a time allotment for exercise, but we should serve as examples to our patients and the public by incorporating healthy movement habits into a daily routine. For instance, breaking exercise into segments throughout the day can efficiently improve health profiles. ²⁶ A physician capable of shifting from sedentary behavior toward an active lifestyle can receive rewarding gains including additional years of lifespan and “healthspan.” ²⁷ The evidence suggests that this would translate to more effective care for patients, and more convincing recommendations for lifestyle changes in patients. Exercise prescriptions have evidence of benefit for a variety of populations: elderly patients, individuals experiencing poor mental health, Parkinson’s patients, etc. Despite the wide-ranging utility of an exercise prescription, this review shows that physicians themselves, who face daunting work-related adversities and demands that lead to burnout, underutilize physical activity in their lives and have rare inclusion in studies of exercise interventions.

Almost half of studies included in this review utilized yoga as an intervention in physicians. This ancient practice has become increasingly popular in Western society due to its ability to bring balance to the physical, mental, emotional, and spiritual facets of one’s health. ²⁸ Yoga has demonstrated positive psychological effects such as improved mood and life satisfaction and decreased emotional distress and anxiety. Research on yoga has also established benefits to physical fitness and cellular/neurocognitive parameters. ²⁹ The five studies included in this review focusing on yoga, four of which targeted residents/interns, showed positive benefits of yoga within the physician population.

Two studies focused on general exercise interventions. One paper studied overweight physicians, revealing positive effects of a treadmill desk intervention that may guide future work and recommendations. While some have focused on specific thresholds (e.g., 30 minutes a day), it is important to acknowledge that health benefits can accrue for busy professionals who attain less than the current physical activity recommendations. ³⁰ The findings from these two studies indicate that while interventions may not improve all outcomes significantly, there appears to be no indication that an exercise or physical activity intervention would negatively impact a physician.

Two studies focused on ergonomics training, demonstrating positive effects in physicians’ pain and ability to perform job duties more effectively. These studies could be extended to operative/procedural specialties such as General Surgery or OB-GYN.

Three included studies focused on wearable activity tracking technology. Unfortunately, several similar papers encountered in the literature search did not provide granular data related to activity level or step count, nor did most perform a true intervention intended to increase physical activity or exercise.^{31,32,33,34,35,36} Technological advancements have potential to motivate individuals toward a healthy lifestyle through the support of self-monitoring behaviors (steps, calories, etc.) as well as building self-regulatory skills such as goal-setting and action plans. ³⁷

This work is important as practicing healthy behaviors not only leads to higher well-being within themselves as physicians but will spillover to patients’ health as well. Frank & Elon discovered that physicians’ ability to motivate patients to prioritize healthy habits can be augmented when they portray healthy habits themselves. ³⁸ Another study found that freshmen medical students believed it would be highly relevant to counsel patients in future practice about exercise. ³⁹ Medical trainees appreciate medical schools being attentive to their own health which may positively influence clinical practice. ⁴⁰ Thus, medical schools could adopt healthy interventions for trainees to create a healthier environment to improve patient counseling. ⁴¹ Research completed by Emory showed that students who received an intervention on health behavior counseling had 50% greater odds of providing counseling on exercise and diet compared to controls. ⁴⁰ Ultimately, if medical schools can increase the attentiveness towards trainees maintaining regular habits such as physical activity, it may improve the quality of wellness counseling completed by doctors. ⁴²

In consideration of the included studies, the evidence indicates numerous positive effects of exercise or physical activity interventions for physicians. As with most interventions, resources could be a barrier for future researchers or stakeholders aiming to improve physician well-being. However, the cost of trackers, physician time, or financial/reward systems to encourage physicians to engage in physical activity would likely offset the cost of medical errors, physician distress and turnover, and suicide. Future work examining this phenomenon should be supported through resource allowance to fully tap into the potential impact and determine strategies to improve physician well-being.

Limitations

Similar to any review article, this paper is limited by search strategies and decisions made by the research team to understand physical activity in doctors. There are important studies that consumers of this review may want to be conscious of published prior to 2012. For example, a 2003 study examined exercise counseling and exercise habits of 4501 women physicians. ⁴³ Several similar studies laid the groundwork for research on improving health in physicians (which has intrinsic value) and the impact this has on patients. Studies may have been missed/not included despite utilization of methodologically-focused searches. Additionally, the overall results included relatively few publications, most with low sample sizes, limiting ability to understand the effects of exercise interventions in physicians. Furthermore, of the papers obtained and included in this study, the team did not perform quality assessment of each article.

Conclusion

This study reviews the overall positive benefits of physical activity and exercise interventions in physicians. This review found limited studies and low sample sizes warranting further work to study the relationship between exercise and wellness in doctors. Future investigations should build on foundational studies to fully realize the potential of interventions to address physician burnout and well-being and the beneficial effects on patient outcomes.

ORCID iD

Jacob Shreffler https://orcid.org/0000-0001-6023-0594