Varner, Derrick F. PhD, PA-C; Foutch, Brian K. OD, PhD
According to a 2006 healthcare morale survey, 9 out of 10 physicians have witnessed severe morale problems among their colleagues and many report that they themselves are “close to the end of their ropes.”1 Combine the smaller percentages of medical students choosing careers in primary care and the unique set of stressors present in a military primary care setting, and one may ask whether US Air Force family medicine providers can sustain, without emotional consequence, the rhythm of administering comprehensive top quality healthcare to their patients. Although depression and burnout have been studied in numerous nonmilitary family medicine healthcare professional populations, this is the first study to compare the risk of depression and burnout among active duty Air Force family medicine physicians, physician assistants (PAs), and nurse practitioners (NPs) working in a primary care setting.
Depression, by definition, manifests with at least five of the following nine symptoms present most of the day nearly every day for a minimum of 2 consecutive weeks:
* depressed mood
* loss of interests and pleasure
* changes in sleep habits or patterns
* changes in appetite or weight
* change in psychomotor activity
* trouble concentrating
* thoughts of worthlessness or guilt
* thoughts about death or suicide
* loss of energy.2
Burnout, on the other hand, is work-related health impairment and consists of three dimensions:
* Emotional exhaustion, the state of being depleted of one's emotional resources
* Depersonalization, also referred to as cynicism, characterized by a negative, cynical, and detached approach to people under one's care
* Reduced personal accomplishment, a sense of low self-efficacy and negative feelings toward one's self.2
To examine the prevalence of depression and burnout symptoms among US Air Force family medicine providers in a military healthcare setting, we applied a conceptual framework by Linzer and colleagues that was originally validated when exploring burnout among Dutch and US physicians.3
Study design This study used a cross-sectional survey design to measure and quantify the effects of sociodemographic factors on depression and burnout symptoms. Volunteer participants completed electronically encrypted anonymous self-directed surveys that combined the Maslach Burnout Inventory (MBI) and Patient Health Questionnaire 9 (PHQ-9)—two validated and reliable survey instruments frequently used in behavioral science research. Kellerman and Herold stated, “Physicians play a key role in the rapidly changing healthcare and public health system, and it is essential to study their attitudes, beliefs, and concerns. One of the most effective ways of doing this is through the use of surveys.”4 The entire questionnaire contained 67 questions that took 10 to 15 minutes to answer after participants completed consent forms. An institutional review board (IRB) waiver was granted for exemption status by the Air Force Research Laboratory and Trident University International IRBs listed under the Department of Health and Human Services regulation 32 CFR 219.
Population The sampling frame consisted of 726 active-duty Air Force family medicine providers—387 family medicine physicians, 278 PAs, and 61 NPs. The study sample included female and male providers ages 20 to 60 years who were engaged in direct patient care (full-time or part-time) at their local military treatment facilities. Providers located at austere deployed locations were afforded the same opportunity to participate in this random anonymous cross-sectional study. Potential subjects were excluded if they were not family medicine providers (for example, pediatricians, general medical officers, flight medicine, or internal medicine providers), not on active duty (for example, contract employees, Reservists, and National Guard), not engaged with any direct patient care, or not credentialed.
Response rate The overall return rate was 21% (150 responses), which was a little less than expected. However, we believe it is representative of the sample population studied and is consistent with the range of 5% to 45% in the current literature of research involving family medicine specialty and military members. Shanafelt and colleagues conducted a national study of burnout among physicians via survey, and the response rate was 26% (7,288 of 27,726 physicians).5 Another recent study to assess factors influencing PA specialty selection demonstrated a 7% (2,020 of 30,000) response rate.6
The encrypted surveys were sent to 726 active duty Air Force FMPs across the globe. Of the 150 respondents, 57% (86) were family medicine physicians, 41% (61) were PAs, and 2% (3) were NPs. Due to the low participation by NPs, comparisons of professional designation were only made between physicians and PAs.
The PHQ-9 portion consisted of nine questions for evaluating depression symptoms and called for a response on a 4-point Likert scale. The PHQ-9 score could range from 0 to 27, and the subjects' depression symptoms were categorized as follows:
* 0 = no depression
* 1 to 4 = minimal
* 5 to 9 = mild
* 10 to 14 = moderate
* 15 to 19 = moderately severe
* 20 to 27 = severe.
For burnout, we employed the revised MBI-GS survey with 16 items covering emotional exhaustion, depersonalization/cynicism, and professional satisfaction efficacy. Scores could range from 0 to 30 for emotional exhaustion and depersonalization and from 0 to 36 for professional satisfaction. However, subjects' burnout symptoms were classified the same for each category: 0-14 = low, 15-24 = medium, and 25 and greater = high.
Of the 150 responses, 84% (126 respondents) scored positive for some degree of depression symptoms. The mean PHQ-9 of 4.76 (95% CI, 4.03–5.49) indicated average reported depression symptoms in the minimum to mild range (Figure 1). Overall burnout symptoms based on MBI scores showed that family medicine providers in our sample had negative burnout symptoms in two of the three categories (Figure 2). For emotional exhaustion symptoms, the mean score was 16.04 (95% CI, 14.83–17.25), with 59% (89) of family medicine providers reporting at least medium levels of emotional exhaustion. For depersonalization or cynicism, the overall mean score was 12.67 (95% CI, 11.48–13.86). Although only 34% (51) of family medicine providers had at least medium level of symptoms, nearly 66% experienced at least low levels of depersonalization. The mean score for professional satisfaction efficacy (positive gratification effect) was 26.85 (95% CI, 25.92–27.79). An overwhelming majority of 143 family medicine providers (95%) had at least medium or high levels of professional satisfaction with practicing primary care medicine.
As shown in Table 1, we analyzed PHQ-9 scores between physicians and PAs, the mean score for PAs were 5.38 (95% CI, 4.12–6.64) in the mild range in comparison to physician colleagues minimal score. As for depersonalization, physicians' low levels measured 13.24 (95% CI, 11.56–14.92) slightly greater in relationship to PAs. Both physicians and PAs experienced high levels of professional satisfaction, with the mean PA score of 27.18 (95% CI, 25.73–28.63). Nonetheless, none of these effects were statistically significant.
For comparisons of professional designation, sex, length of time practiced, and age, the three NP respondents were dropped and not used further in the analysis. In addition, we split length of time practiced (0-8 years versus more than 8 years) and age (25-44 years versus over 44 years) for analysis. After doing so, we performed one-way analyses of variance with professional designation, sex, length of time practiced, and age as predictors of depression (PHQ-9) and the three burnout (MBI) scores. No interaction effect was found between professional designation and any other predictor of depression or burnout scores, so we analyzed physicians and PAs together (Figure 3).
As shown in Figures 3 to 6, neither professional designation, sex, length of time practiced, nor age produced differences in depression (PHQ-9) scores. However, Figure 3 shows that physicians demonstrated a negative trend in all burnout (MBI) scores. That is, physicians reported slightly higher emotional exhaustion and depersonalization but lower professional satisfaction symptom scores. However, none of these effects were statistically significant. Similarly, length of time practiced was a negative predictor of burnout symptoms. Although neither trend was significant, providers with more than 8 years of practice had higher emotional exhaustion and depersonalization scores than those with 8 or fewer years (Figure 4). Female respondents also tended to report higher emotional exhaustion and depersonalization but lower professional satisfaction scores than male respondents (Figure 5). Again, none of these trends were statistically significant.
The singular significant predictor of burnout symptoms in this study was age. As shown in Figure 6, respondents over age 44 years reported lower emotional exhaustion and depersonalization but higher personal satisfaction scores than their younger counterparts. The trends for emotional exhaustion and professional satisfaction were statistically significant.
Respondents provided information that, together with the formal analysis of PHQ-9 and MBI scores, could suggest that family medicine physicians and PAs may be vulnerable for developing burnout and/or depression symptoms while working in the military family medicine environment. Despite challenges encountered by family medicine providers working in primary care settings, 98% (147) of the participants “strongly agree” that they consistently provide high-quality care to each patient.
Measures to reduce burnout and/or depression should be targeted at all family medicine providers regardless of their professional designation or training background. Interestingly, 96% (144) of active-duty Air Force family medicine providers agreed that if they were empowered or were in a position to make changes in family medicine, they would do so. Some of the suggested measures by family medicine providers to enhance their well-being or improve practice efficiency were smaller patient empanelment, more clinical staffing to manage the demands of the patient population load, improving medical team continuity, decreasing metrics for the number of patients to be seen per day, focusing on quality versus quantity of care provided, making the electronic health record more efficient and provider-friendly to allow seamless documentation, having more time with patients and not feeling rushed, and allowing comprehensive care.
One of the goals of this study was to increase awareness for Air Force family medicine providers, minimizing negative outcomes and improving retention and recruitment strategies. To do so, family medicine providers are continually informed of the confidentiality and effectiveness of mental health services at their disposal. As noted by Mushet and Donaldson, “Even those who managed to use the service were often struggling with great difficulty in accepting their human vulnerability and receiving appropriate help and time off work.”7 Therefore, a nonreprisal attitude at all leadership levels is vital.
The primary limitation for cross-sectional designs is the inherent inability to infer causality as compared with experimental or cohort-longitudinal designs. Another limitation is the inability of our sample population (active-duty military only) to generalize to PAs in civilian practice environments.
The relatively low response rate is another limitation. We expected that a small percentage of family medicine providers might not fully answer or return the survey if they were experiencing a high degree of burnout or severe depressive symptoms. The low return rate could also be attributed to the unique challenges of being an active-duty military member, including fearing reprisal from leadership or a stigma related to the sensitive nature of the topic being studied. We suspect the frequency and timing of peak military move cycles (that is, moving from one military installation to another) also may have prevented some family medicine providers from participating in the study due to limited or no access to their Air Force secure e-mails.
The nation is in the midst of a supply crisis for primary healthcare providers. Future research should include providers in pediatrics, internal medicine, and flight medicine, perhaps in collaboration with other branches of the military (such as the Army, Navy, and Coast Guard) that provide family medicine services. This would cast a broader net to a comprehensive audience to assist in identifying and/or confirming specific personal, social, or organizational risk elements that may contribute to depression and/or burnout. We acknowledge that some of the findings may be due to the military setting rather than professional status as a PA.
Also, future research could explore personal and public stigmas or barriers that prevent family medicine providers from seeking mental healthcare. Continued research in the form of prospective or longitudinal studies will be required to elucidate the consequences of these conditions and identify risk factors. In essence, as Linzer and colleagues suggested, “periodic review of physicians can provide insight into what background and mediating factors organizations can focus on to improve the medical professional's quality of life, (which) translates into better patient care.”3
1. Weber DO. Physicians offer prescriptions to boost low morale. Physician Exec. 2006;32(6):16–22.
2. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Arlington, VA: American Psychiatric Association; 2013.
3. Linzer M, Visser MR, Oort FJ, et al. Predicting and preventing physician burnout: results from the United States and the Netherlands. Am J Med. 2001;111(2):170–175.
4. Kellerman SE, Herold J. Physician response to surveys. A review of the literature. Am J Prev Med. 2001;20(1):61–67.
5. Shanafelt TD, Boone S, Tan L, et al. Burnout and satisfaction with work-life balance among US physicians relative to the general US population. Arch Intern Med. 2012;172(18):1377–1385.
6. Halasy MP, Leafman J, Mathieson K, et al. A descriptive analysis of factors influencing physician assistant specialty selection. JAAPA. 2012;25(7):52.
7. Mushet G, Donaldson L. A psychotherapist in the house: a service for distressed junior doctors. Br J Med Psychol. 2000;73(Pt 3):377–380.
© 2014 American Academy of Physician Assistants.