Loneliness, Burnout, and Other Types of Emotional Distress Among Family Medicine Physicians

Results From a National Survey

Samuel Ofei-Dodoo, PhD, MPA, MA; Rebecca Mullen, MD, MPH; Andrew Pasternak, MD, MS; Christina M. Hester, PhD, MPH; Elisabeth Callen, PhD, GStat; Edward James Bujold, MD; Jennifer K. Carroll, MD, MPH; Kim S. Kimminau, PhD


J Am Board Fam Med. 2021;34(3):531-541. 

In This Article


Study Design and Participants

The study was a cross-sectional survey of physicians who were members of the American Academy of Family Physicians (AAFP) and AAFP National Research Network (NRN) between December 7, 2019, and January 20, 2020. The study participants (n = 401) completed an anonymous, 30-item survey that included questions regarding loneliness, burnout, symptoms of depression, fatigue, in addition to personal demographic information. The survey link was emailed to all members of the AAFP NRN and the AAFP Member Insight Exchange, a subset of AAFP members who agree to be contacted for surveys. The AAFP Institutional Review Board gave this study exempt status on review.

Table 1 presents the demographic information of the respondents. The mean age of respondents was 50.2 years (standard deviation [SD] = 11.1). A slight majority were female (53.1%), most were employed full-time (84.4%), worked in urban areas (85.0%), and were members of AAFP Member Insight Exchange (71.8%). The mean years in clinical practice among the physicians was 18.9 years (SD = 11.3).

Study Instruments

Loneliness. We assessed the physicians' level of loneliness using a 3-item University of California, Los Angeles (UCLA) Loneliness Scale, a validated shortened screening tool to measure loneliness.[24,25] The questions are: How often do you feel that you lack companionship? How often do you feel left out? How often do you feel isolated from others? Participants recorded their responses on a 3-point Likert-like scale (1 = Hardly ever, 2 = Sometimes, and 3 = Often). Scores for the 3 questions were summed, with a possible score ranging from 3 to 9. As a standard practice,[25,26] respondents who had a total score of 6 or above were considered "lonely."

Burnout. Assessment of burnout among the respondents used 2 single-item measures of overwhelming exhaustion and depersonalization adapted from the full Maslach Burnout Inventory (MBI-22), which has been validated.[27] The overwhelming exhaustion item ("I feel burnout from my work") and depersonalization item ("I've become more callous toward people since I became a physician") have been shown to be useful screening questions for burnout.[28–30] These 2 items have shown the highest factor loading[27,31,32] and strong correlation[28,33] with their respective emotional exhaustion and depersonalization domains in the MBI-22.[28] The 2 single items have been used in previous studies to measure manifestations of burnout among physicians.[8,34–36] The respondents recorded the degree to which each item applied to them on a 7-point Likert-like scale (0 = Never, 6 = Every day). The scores of each domain were grouped into low, moderate, and high burnout categories using established cutoffs.[8,27,34,36] Higher scores are indicative of greater exhaustion and depersonalization, and greater burnout. Consistent with convention,[8,34,36] we considered respondents who scored high (score of greater than 3) on exhaustion and/or depersonalization domains as having at least 1 manifestation of professional burnout.

Symptoms of Depression and Fatigue. We screened for symptoms of depression using a 2-item Primary Care Evaluation of Mental Disorders Patient Health depression-screening questionnaire (PHQ-2). The questions asked the participants about their feelings of being down, depressed, or hopeless and if they have been bothered by little interest or pleasure in usual activities during the past month. The 2-question screener has a reported 96% sensitivity and 57% specificity for depression.[37,38] The PHQ-2 is an accurate depression screening in adults and used in previous studies to screen for symptoms of depression among physicians.[18,30,39–41]

We measured the physicians' levels of fatigue during the past week using a standardized linear analog scale (0 = as bad as it can be, to 10 = as good as it can be) similar to what is described by West and colleagues.[42] Respondents who scored 5 or less on the scale were considered to have higher levels of fatigue.[30,34,39]

Statistical Analyses

Standard descriptive statistics were used to create a demographic profile and describe the prevalence of loneliness, burnout, depression, and fatigue among the respondents. Mann-Whitney U test/independent samples t-test and/or Kruskal-Wallis test/1-way ANOVA (for continuous variables), and Likelihood Ratio Chi-square (for categorical variables) were used to evaluate the association between variables. Generalized linear mixed models were used to calculate associations between the loneliness classification modeled as a binary outcome against a single fixed effect for independent variables (age, gender, employment status [whether a participant is employed part-time, full-time, fully retired, or not in the workforce for other reasons], practice location, membership [AAFP Member Insight Exchange and AAFP NRN], depression, emotional exhaustion, depersonalization, and fatigue).

Adjusted odds ratios were estimated by modeling all significant independent variables against the loneliness classification, controlling for respondent age, gender, employment status, practice location, and membership. Adequate power (>0.85) to detect significant relationships among the variables with 2 degrees of freedom, P < .05, and 0.21 effect size required a sample size of 350 respondents.[43,44] All analyses were 2-sided with α of 0.05. The IBM SPSS (Statistical Package for the Social Sciences; Armonk, NY), version 26 was used for these analyses.