Burnout During Residency Training: A Literature Review

Burnout in health care professionals has gained significant attention over the last several years. Given the intense emotional demands of the work environment, clinicians are particularly susceptible to developing burnout above and beyond usual workplace stress. Residency training, in particular, can cause a significant degree of burnout, leading to interference with individuals' ability to establish rapport, sort through diagnostic dilemmas, and work though complex treatment decision making. Overall, burnout is associated with a variety of negative consequences including depression, risk of medical errors, and negative effects on patient safety. The goal of this review is to provide medical educators and leaders with an overview of the existing factors that contribute to burnout, the impact of burnout, interspecialty variation, and suggestions for interventions to decrease burnout.

The term burnout was coined by psychologist Herbert Freudenberger1 in 1974 in an article entitled “Staff Burnout” in which he discussed job dissatisfaction precipitated by work-related stress. A broadly applicable description defines burnout as a state of mental and physical exhaustion related to work or caregiving activities. A long-standing conceptual and operational definition characterized burnout as a triad of emotional exhaustion (emotional overextension and exhaustion), depersonalization (negative, callous, and detached responses to others), and reduced personal accomplishment (feelings of competence and achievement in one's work).2 In the World Health Organization International Classification of Diseases, 10^th revision, burnout is defined as a “state of vital exhaustion.”3 Although no specific diagnosis of burnout is mentioned in the Diagnostic and Statistical Manual of Mental Disorders,4 burnout is a clear syndrome with significant consequences.

Burnout may be associated with decreased productivity and decreased job satisfaction. The rates of depression,5 suicidal ideation, plans, and attempts6 were noted to be high in burnout states and tended to decline with recovery from it. Other risks include cardiovascular disease7 and increased inflammation biomarkers.8 Physical symptoms may take many different forms, including insomnia, appetite changes, fatigue, colds or flu, headaches, and gastrointestinal distress. Physical symptoms alone may interfere with one's sense of well-being and ability to function fully at work.9–11 Psychological symptoms such as low or irritable mood, cynicism, and decreased concentration can negatively affect productivity and rapport.12 A 2008 preliminary study surveying 178 matched pairs of physicians and patients who had been hospitalized within the previous year revealed that (after controlling for illness severity and demographic factors) the depersonalization aspect of physician burnout was associated with lower patient satisfaction and longer postdischarge recovery time.13 Additional components of burnout may include daydreaming while with patients, excessive cancellations, procrastination, and delaying paperwork and vocational tasks.14 Burnout may also lead to increased alcohol or drug use, which can also impact patient care.15,16 

We searched MEDLINE and PsychINFO from 1974 to 2009 using the following keywords: “burnout” AND “residents or residency.” Fifty-one studies were identified that addressed issues related to burnout. Key findings were derived from the full text of the published articles after reviewing the abstracts.

The Maslach Burnout Inventory (MBI)17 is the most commonly used questionnaire to measure burnout in research studies. The MBI human services survey is a self-administered, 22-item questionnaire that was developed to measure burnout in human services workers and is regarded to be the “gold standard” in measuring burnout.18 The MBI items are rated on a Likert scale from 0 to 6 (0  =  never, 1  =  a few times per year, 2  =  once a month, 3  =  a few times per month, 4  =  once a week, 5  =  a few times per week, and 6  =  every day) and score sample items such as: “I feel emotionally drained from my work.” It is designed to assess the 3 primary dimensions of burnout: emotional exhaustion, depersonalization, and personal accomplishment. Burnout is detected using cutoff scores of high emotional exhaustion (≥27), high depersonalization (≥10), and low personal accomplishment (≤33), based on normative data from 1104 medical professionals.19 Many studies using the MBI define burnout as high emotional exhaustion or depersonalization. The personal accomplishment scores are less commonly included because they are thought to correlate less with psychological strain.18 

For medical professionals, the seeds of burnout may be planted as early as medical school. The literature to date seems to support the notion that there are a variety of factors during medical school that contribute to burnout in physicians, and that burnout is a phenomenon that develops cumulatively over an extended time period.20–27 Burnout rates in medical students range from 28% to 45%.20,25 Evidence shows that the factors contributing to burnout include environmental aspects such as stress during medical school, as well as inherent personality traits such as introversion and neuroticism.22,24 Burnout is a phenomenon that may present during medical schools, and may develop or continue to exist in residents and practicing physicians. Several studies have explored possible reasons for burnout in residency training. In these studies, residents report that time demands, lack of control over time management, work planning, work organization, inherently difficult job situations, and interpersonal relationships are stressors that may contribute to burnout.16,28,29 

Multiple studies have looked at burnout rates among resident physicians. A 2006 study by Rosen et al30 showed that at the beginning of intern year, 4.3% of internal medicine residents met criteria for burnout as measured by the MBI. By the end of the first year, the rates had increased to 55.3%, with a significant increase in both the depersonalization and emotional exhaustion subscales. Another study of internal medicine residents at the University of Washington found that 76% met criteria for burnout as measured by the MBI, regardless of postgraduate year.31 A notable nationwide study by Collier et al32 looked at cynicism and humanism in internal medicine residents across the United States. Sixty-one percent of the 4128 residents who responded admitted to becoming more cynical and 23% reported becoming less humanistic during medical training. Although this study did not employ a formally validated burnout scale, it can be assumed that cynicism and decreased humanism can lead to emotional exhaustion and depersonalization, thus contributing to burnout.

In 2004, Martini et al33 did a unique study that compared burnout rates among the different specialties using the MBI. The overall burnout rate was 50% and ranged from 27% to 75% among different specialties. This variation among specialties was not statistically significant; however, burnout rates were as follows: 75% in obstetrics-gynecology followed by 63% in internal medicine, 63% in neurology, 60% in ophthalmology, 50% in dermatology, 40% in general surgery, 40% in psychiatry, and 27% in family medicine. However, this variation among specialties was not statistically significant. Being in one's first year in residency, mood fluctuation, dissatisfaction with clinical faculty, recent family stress, and being unmarried were all associated with increased likelihood to meet burnout criteria.33 Psychiatry residents were noted to have additional stressors including fear and exposure to patient violence and suicide.34–38 

The table offers a summary of the findings from available studies performed with residents in different specialties and details the percentage of residents experiencing burnout on the 2 most frequently-used subscales: emotional exhaustion and depersonalization.

There are a wide variety of inherent and environmental factors that have been shown to affect burnout rates in resident physicians. The effect of gender on burnout shows conflicting results. Some studies showed that female residents scored significantly lower than male residents on the depersonalization subscale, emotional exhaustion, and personal accomplishment subscales, whereas other studies have shown the opposite.29,31,45 Woodside et al36 evaluated the association between age and burnout, and showed an inverse correlation between depersonalization scores and age of residents. However, 3 other studies showed no significant correlation between age and burnout.37,45,46 

Marriage and parenting have also been examined in relation to burnout. Martini et al33 showed that 65.2% of single, divorced, or unmarried residents met the criteria for burnout compared with 40.0% of married individuals (P < .01). Other studies report no correlation between marriage and burnout.31,47 Although it could be assumed that having the added responsibility of caring for children would add to burnout, research has shown that parenting can act as a protective factor against burnout. Parenting has a possible humanizing effect on residents, resulting in less detachment and depersonalization.31 Collier et al32 showed that having children during residency resulted in lower rates of depression and cynicism as well as an increase in humanistic feelings. However, other studies showed that parenting has no effect on burnout.46 Interestingly, residents from other cultures who train in the United States experienced significantly less burnout, with significantly lower scores on both emotional exhaustion and depersonalization, than those who were born in the United States.31,48 Finally, in 2004 Thomas49 reviewed the literature and showed that the intensity of the workday and its interference with the resident's home life plays a more influential role in burnout than inadequate sleep. The same review also showed that studies on the relationship between burnout and personality traits were inconclusive.

The impact of burnout on patient care has been investigated in several studies. Some studies have shown that residents reporting burnout were more inclined to self-report suboptimal patient care and practices and medical errors than those without burnout.31,50 However, in a study by Fahrenkopf et al43 no actual correlation was found between burnout and the number of medical errors seen in collected data. One plausible explanation may be that residents reporting symptoms of burnout may be more likely to overreport their errors.

The effect of work hour limitations on residents has been researched widely as an important environmental consideration in the development of burnout. In July 2003, the Accreditation Council for Graduate Medical Education implemented work hour limitations, cutting down the workweek to 80 hours and no more than 24 consecutive hours, with an additional 6 hours for transfer of care and educational activities per call. Gelfand et al44 looked at the effect of the 80-hour workweek restriction on resident burnout. Surgical residents and faculty completed the MBI 1 week before and 6 months after the mandated time restrictions took effect in 2003. Despite a reduction of approximately 18 work hours per week, there was no statistical difference in emotional exhaustion, depersonalization, or personal accomplishment scores, although others have commented on methodological flaws in the study.51 Martini et al47 also looked at burnout rates in all interns and residents in a variety of different specialties at Wayne State University School of Medicine, before and after the work hour limits were placed. Residents who reported working more than 80 hours had higher rates of burnout (69.2%) compared with a burnout rate of 38.5% after the time restriction was in effect. Overall internal medicine residents reported that the work hour limits have had a positive effect with a decrease in the amount of teaching by attending physicians, as well as “having to cut corners” on both patient care and educational activities.52 

Current data on interventions for physician burnout are insufficient to recommend particular measures. Although a number of interventions have been studied in other caregiving and health care professionals and not necessarily residents, some of the same interventions may or may not be effective. More studies are needed to examine the applicability and utility of these interventions in resident physicians. Interventions to address burnout fall into 2 categories: workplace-driven interventions and individual-driven behavioral, social, and physical activities.

Suggested interventions in the workplace include developing stress-reduction programs, increasing staff awareness of burnout, enhancing support for health professionals treating challenging populations, and ensuring a reasonable workload.53 Some programs have attempted to manage workload by instituting night float or home call systems. Positive effects include increased opportunities for rest; potential negative effects include decreased clinical and surgical experience and decreased opportunities for development of professionalism and communication skills.54 Furthermore, increasing variety in workplace roles (opportunities to conduct research, teach, and supervise) in addition to performing direct clinical care has been documented to improve satisfaction.55 Mentoring programs in residency training can also be helpful in this regard.48,56 Emotional intelligence training and team building for residents may decrease burnout. Emotional awareness and emotional management abilities help to maintain effective and appropriate relationships within a team, thereby preventing dysfunctional relationships and stress, leading to subsequent burnout.57 The Cedars-Sinai Obstetrics and Gynecology Residency uses a program to develop self-awareness profiles for residents. Through this program, residents reflect on their interpersonal styles, participate in interactive group activities, and have the opportunity to debrief from group activities.58 Process groups have also been used (especially in psychiatry) to provide a protected setting to address issues that are professional and bordering on personal, and to help foster work-related relationships. Some institutions have developed “GME (graduate medical education) Wellness Teams,” or resident-specific employee assistance programs to support a healthy approach to the training environment. Such programs address a wide variety of personal and professional issues while ensuring confidentiality.59 In the Cedars-Sinai Psychiatry Residency Program, a wellness consultant developed modules incorporating time management, relaxation response techniques, focused breathing, and meditation methods. The modules are administered interactively to the residents in order to facilitate heightened awareness of stress response and managing stress through intra-workday rather than post-workday activities.60 It is critical, moving forward, to continue to develop, rigorously study in a randomized controlled fashion, and publish the impact of such workplace-driven interventions on burnout in residency programs.

There are a number of interventions that can be used individually by residents. Peer support around challenging cases can be validating and stress reducing. In a sample of 200 professionals, Maslach61 showed that venting, laughing, and discussing care with colleagues decreased personal anxiety. Participating in professional organizations and attending lectures or conferences can further develop work-related social networks. Meditation has been shown to improve burnout.62 Another study on meditation and stress among health care professionals revealed improved mood and emotional states for those who practice meditation regularly.63 Physical exercise has also been shown to ameliorate depression, anxiety, and mood, making it an ideal intervention for burnout.64 Creating a defined boundary between work and home has also been strongly suggested in the literature.65 Counseling can provide an opportunity to explore occupational challenges and increase self-awareness. Some residency programs are encouraging and supporting personal psychotherapy for residents.66 Other suggested interventions include vacation, mindfulness techniques, yoga, reflective writing, spiritual activities, scheduled daily rest, music, massage, and enjoying nature. It is important for residency training directors and educators to model and demonstrate self-care and awareness; it is also important to support residents in identifying and pursuing the individual techniques that each individual finds most useful.

Maslach67 summarized effective working through burnout by stating: “If all of the knowledge and advice about how to beat burnout could be summed up in 1 word, that word would be balance—balance between giving and getting, balance between stress and calm, balance between work and home.”

Burnout is a triad of emotional exhaustion, depersonalization, and a sense of decreased personal accomplishment. It is a phenomenon that reflects the complex interaction between environmental stressors, genetic vulnerabilities, and coping styles. Burnout can contribute to multiple physical symptoms, psychological symptoms, and substance abuse, all of which can impact a resident's quality of life, ability to provide sustainable and safe patient care, quality of learning and teaching, and the overall morale of a residency program. The studies we reviewed suggest that residents, especially in the early years of training, are particularly vulnerable to burnout, with a prevalence rate ranging from 27% to 75%. The negative impact of burnout on patient care includes risk of medical errors, patient safety risks, and potential compromise of quality of care. Negative consequences of burnout on physicians in training include depression, suicidal tendencies, and medical illnesses. Effective interventions to address burnout should be developed at both the individual and institutional levels. Work hour regulations developed by the Accreditation Council for Graduate Medical Education are only part of the solution to improving patient safety. Although preliminary studies indicate that work hour limitations have improved resident quality of life, there is potential risk to decreased educational opportunities and a “shift” mentality. As a result, individuals and programs must also incorporate other interventions to balance appropriate service responsibilities with the academic and educational training mission. Easy, affordable access to adequate counseling services is an additional tool in creating educational programs that enhance resident performance and provide outstanding, safe, high-quality patient care. New physician generations were shown to have a significant attraction to work and life balance activities,68 thus creating readily accepting participants for burnout interventions.

Burnout is a well-known phenomenon that must be addressed by leadership in academic medicine. As we move toward national health care reform and attempt to reconfigure our approach to training, attention to personal well-being is critical to the successful education of the next generation of physicians.