Recent healthcare projections suggest that more than 84 million people in the United States live in an area of primary care shortage.1 Although the numbers of physician assistants (PAs) and NPs are expected to increase, the Association of American Medical Colleges anticipates a shortfall of up to 49,000 primary care physicians by 2030.2 A contributing factor to this shortage is the increasing number of physicians retiring early due to burnout and job dissatisfaction.3 A 2014 study found that more than half of US physicians suffer from burnout, a nearly 20% increase from 2011.4 Dissatisfaction and burnout among physicians were also shown to be higher than in the general US professional working population, with implications for quality of care, patient satisfaction, and patient safety.4,5
The projected primary care provider shortage has implications for potential burnout for all healthcare professionals. Burnout can lead to broken relationships, substance abuse, depression, and even suicide.5 Professionally, burnout may result in medical errors and compromised quality of patient care. Additionally, early retirement due to burnout reduces the medical workforce.5 Staffing shortages can lead to an increased workload, which contributes to reduced job satisfaction and more burnout.6 Although burnout has been defined in many ways, this study uses social psychologist Christina Maslach's description of burnout as a state characterized by emotional exhaustion, depersonalization, and lack of personal accomplishment.7 For healthcare workers, this triad of characteristics may manifest as depletion of emotional resources (emotional exhaustion), development of apathetic attitudes toward patients (depersonalization), and feelings of professional uselessness or unproductiveness (lack of personal accomplishment).7
Studies show that rates of burnout differ across medical specialties; some of the highest rates of burnout among physicians are in primary care (family medicine and general internal medicine).8 However, we know little about PAs because most research has focused on burnout in unique subpopulations such as emergency medicine, oncology, or rural practices.9-11 These studies have limited generalizability to primary care or to the PA profession as a whole. One recent nationwide survey looked at the entire PA profession but did not use a standardized tool to assess burnout, so it cannot be directly compared with other literature on burnout.12
The PA profession is regarded as one of the best career options in the United States and has historically been found to foster high levels of job satisfaction.13 The flexibility afforded to PAs to change specialties distinguishes them from physicians and may contribute to a high degree of career satisfaction.14,15 Physician literature has not shown a consistent correlation between satisfaction and burnout, and to the best of our knowledge, little research has examined the link between satisfaction and burnout in the PA profession.8,16
This study sought to describe burnout among PAs in Minnesota as well as identify the specialties in which burnout is the most prevalent. Using Maslach's validated burnout survey tool, we were able to capture three dimensions of burnout: emotional exhaustion, depersonalization, and lack of personal accomplishment.7 Secondary objectives included assessing associations between job satisfaction, flexibility, and burnout. Our findings will expand our understanding of PA burnout in one state and encourage similar research nationwide.
Following approval from the St. Catherine University institutional review board, we initiated an online, cross-sectional survey of PAs practicing in Minnesota. The survey tool consisted of the Maslach Burnout Inventory and original questions. The survey was created and distributed through Qualtrics survey software. We purchased contact information for certified PAs in Minnesota from the Minnesota Board of Medical Practice. This list of 1,166 email addresses represented 52% of the 2,242 licensed PAs in Minnesota.17 We emailed our survey to these 1,166 PAs in November 2016, followed by two reminder emails sent 2 weeks apart. The survey link closed in December 2016.
Demographic characteristics in the survey were age, sex, and years practicing as a PA. Professional characteristics included whether the respondent was practicing as a PA, primary work setting, and primary practice specialty (17 options). With the same 17 specialty options, we asked respondents to mark all current and past specialties, from which we created a new variable called number of specialties, reflecting the total number of specialties in which a clinician had practiced during his or her career. Using the National Commission on Certification of Physician Assistants (NCCPA) report conventions, we created a primary care category combining family practice and general internal medicine.18 All other specialties were collapsed into a nonprimary care category.
We measured burnout using the Maslach Burnout Inventory Human Services Survey (MBI), a validated 22-item questionnaire.19 Each item on the MBI uses a 7-point Likert scale (ranging from 0 = never to 6 = every day) in which respondents reflect on an attitude or feeling statement. A sample statement is “I feel emotionally drained from my work.” The responses were individually scored into three subscales, capturing different dimensions of burnout: emotional exhaustion, depersonalization, and personal accomplishment. Collectively, burnout is characterized by high scores on the emotional exhaustion and depersonalization subscales and a low score on the personal accomplishment subscale. However, in keeping with other studies, the three subscales were assessed independently as opposed to calculating a total score.20 Each subscale also was coded into three levels (high, moderate, and low) using the MBI scoring guidelines.19
We assessed career satisfaction by asking, “Please select your satisfaction level of the PA profession for your career” following a previous survey methodology.21 Response options were 1 = extremely dissatisfied, 2 = mostly dissatisfied, 3 = dissatisfied, 4 = satisfied, 5 = mostly satisfied, and 6 = extremely satisfied. A nearly identical question addressed satisfaction level with one's current position. To clarify, career satisfaction referred to satisfaction with the PA profession as a whole; job satisfaction referred to the clinician's satisfaction with his or her position at the time of the survey. For analysis of the career satisfaction questions, we dichotomized the responses into satisfied (satisfied, mostly satisfied, and extremely satisfied) and dissatisfied (dissatisfied, mostly dissatisfied, and extremely dissatisfied). Perceived benefits of flexibility in the PA profession was assessed by asking, “In your opinion, does the flexibility to change specialties in your practice as a PA help prevent burnout?”
We calculated descriptive statistics to examine demographic and professional data, satisfaction with the PA profession, and career flexibility. Spearman rank correlation coefficients were used to compare the three MBI subscales with age, sex, years of practice, primary work setting, number of specialties, flexibility, satisfaction with profession, and satisfaction with current position. Strength of the relationships was assessed between 0.1 and 0.3 as weak, between 0.4 and 0.6 as moderate, and between 0.7 and 0.9 as strong. Additionally, we used chi-square analysis to compare each level of burnout (low, moderate, and high) against primary care and nonprimary care specialties. Statistical significance for all tests was set at P < .05. All data were analyzed with SPSS version 24.
Of the 1,166 surveys sent to PAs in Minnesota, 366 were returned by the deadline, yielding a 31.4% response rate. Thirty-one (8.5%) of the 366 surveys could not be used due to incomplete responses. Twenty-three respondents (6.3%) indicated they were no longer practicing and were removed from further analysis, leaving a total of 312 completed surveys available for analysis. Table 1 illustrates the demographic and professional characteristics of the final sample. Most respondents were female (68.9%), with a mean age of 43 years. Nearly 60% had practiced for more than 10 years, with a mean of 13.6 years; more than 61% had changed specialties at least once during their careers. Family medicine was the single largest area of practice (28.5%) followed by internal medicine subspecialty (11.9%). Nearly a third (31.7%) of respondents were represented under the umbrella of primary care, which combines family medicine and general internal medicine (excluding internal medicine subspecialty). Group office practice (44.2%) followed by hospital setting (31.4%) were the most common practice settings. Our demographics are consistent with statewide and national figures, which provide a measure of external validity and suggest that we are well-representing our target population. For example, 68.9% of our respondents were female compared with 72% of PAs statewide and 67.7% nationwide; 28.5% identified their current specialty as family medicine compared with 27.8% statewide and 20.6% nationwide.17,18
The distribution of low, moderate, and high scores for each burnout subscale are shown in Table 2. Using the median scores for each dimension, we found that overall, participants experienced moderate levels of emotional exhaustion, low levels of depersonalization, and high levels of personal accomplishment. In terms of high levels of burnout, more respondents experienced high emotional exhaustion (35.3%) than high depersonalization (18.9%) or low personal accomplishment (6.7%). The majority (95.9%) were satisfied, mostly satisfied, or extremely satisfied with the PA profession as a whole and 87.8% indicated satisfaction with their current position. Additionally, 86.9% believed the flexibility in their profession helps prevent burnout. Chi-square analysis indicated that PAs working in primary care were more likely to experience emotional exhaustion than those in other specialties (P < .002) (Figure 1). Of those who worked in primary care, 47.5% had emotional exhaustion scores in the high range compared with 29.6% of those in nonprimary care settings. Depersonalization and personal accomplishment did not differ between primary care and nonprimary care settings (P = .068 and P = .374, respectively).
We examined the strength of personal and professional characteristics against each of the burnout subscales using Spearman rank correlation coefficient (Table 3). PAs working in primary care reported higher emotional exhaustion than those in nonprimary care settings (P < .01), but the correlation coefficient indicated a weak relationship. Being female was also weakly correlated with emotional exhaustion (P < .01). PAs are more likely to experience depersonalization with increasing age and years of practice (both P < .01).
Satisfaction with one's career and current position were both moderately correlated with an increased sense of personal accomplishment (both P < .01). On the other hand, satisfaction with one's career and current position were both moderately negatively correlated with emotional exhaustion and depersonalization (all P < .01). This suggests that clinicians are less likely to experience emotional exhaustion and depersonalization if they are satisfied with their career and/or current position. We identified no significant correlations with any of the subscales and type of work setting, flexibility to change specialties, and number of times one changed specialties.
This study found that about 1 in 3 PAs (35.3%) in Minnesota experiences burnout as measured by the emotional exhaustion subscale. Emotional exhaustion was found to be higher among female versus male PAs and those working in primary care versus nonprimary care specialties. Age and years of practice were correlated with higher rates of depersonalization. Conversely, satisfaction with one's career and current position was negatively correlated with both emotional exhaustion and depersonalization.
For clinicians who experience it, burnout is most pronounced on the emotional exhaustion subscale. However, despite experiencing higher levels of emotional exhaustion in the workplace, PAs in Minnesota still maintain high levels of personal accomplishment and low levels of depersonalization. When compared with recent studies assessing the prevalence of burnout among US physicians, PAs in Minnesota experience lower rates of burnout on all three subscales than their physician colleagues. For example, in a 2014 study, 46.9% of physicians reported high levels of emotional exhaustion compared with the 35.3% of PAs that our study found.4 Additionally, 34.6% of physicians reported high levels of depersonalization (compared with 18.9% from our study) and 16.3% reported low levels of personal accomplishment (compared with 6.7% from our study).4
Burnout research on PAs is limited, with much of the available literature focusing on unique specialties or populations.9-11 A 2002 study of PAs in emergency medicine showed that fewer (26.2%) experienced emotional exhaustion compared with the overall population of Minnesota PAs (35.3% from our study).9 Conversely, the study showed that PAs in emergency medicine experienced higher levels of depersonalization (35.6% compared with 18.9% from our study). When we examined differences among PAs in Minnesota by specialty, PAs in emergency medicine were grouped with other nonprimary care specialties. We found that nonprimary care PAs (including PAs in emergency medicine) were less likely than those working in primary care to report burnout. A 2016 study of PAs in rural areas saw higher rates of burnout than our sample, with 69% reporting high emotional exhaustion, 46% reporting high depersonalization, and 29% reporting low levels of personal accomplishment.10 However, the low response rate (11.3%) in that study limits its generalizability. One recently published study of PAs in oncology mimicked our results in that there were high levels of career satisfaction (86.4%) alongside moderate levels of burnout.11 For example, 30.4% reported high emotional exhaustion (compared with 35.3% from our study) and 17.6% reported high depersonalization (compared with 18.9% from our study). The most notable difference between the PAs in oncology and our study was in low personal accomplishment (19.6% compared with 6.7%, respectively). Our survey did not have an oncology specialty option, limiting direct comparison.
To the best of our knowledge, the only nationwide study on PA burnout not focusing on a specialty was published in 2018 using data from the 2016 American Academy of PAs (AAPA) salary survey.12 Similar to our findings, this study found that PAs generally report enthusiasm for work and happiness with their careers despite experiencing symptoms of burnout. This study also identified an association between burnout and female sex, as more female PAs were found to have quit a job due to stress than their male counterparts (32.2% compared with 25.6%).12 However, a comparative limitation is that this was a retrospective survey that did not use the MBI, nor did it use the terminology of satisfaction or assess for correlations. Therefore, although these findings parallel ours, we cannot directly compare data.
Our results suggesting that PAs in primary care have higher rates of burnout than those in other specialties parallel data on primary care physicians.8,22 Multiple studies have shown that the high rate of burnout among primary care physicians is associated with an excessive workload, clerical burden, loss of control over work, and problems with work-life balance.8,15,22,23 A recent review of literature related to PA burnout also suggests that the increased likelihood of developing burnout in primary care may be related to the fact that this is a specialty in which PAs are helping to fill the void of expected physician shortages.15 Similar to our study, Rabatin and colleagues found that female physicians working in primary care were more likely to experience burnout than their male colleagues.22
Historically, career and job satisfaction among PAs have been high, as demonstrated by multiple studies throughout the evolution of the profession.13 This point has been echoed in more recent studies, both national and specialty-specific.11,12 Not only are our findings consistent with these trends, we also found that career and job satisfaction among PAs in Minnesota are higher than in a previous national study.21 Minnesota PAs indicated their satisfaction with their career was 95.9% compared with 92.4% nationally, and satisfaction with their current job was 87.8% compared with 81.8% nationally.18 A 2016 Minnesota PA workforce report asked a similar question: “How satisfied have you been with your career in the last 12 months? And overall?”17 In this report, satisfaction with career was higher in the overall category than in the past 12 months, suggesting similar findings to our study. Our study found that whether clinicians were satisfied with their current position, they were still likely to be satisfied with the PA profession as a whole. Our results also showed that burnout was negatively correlated with job satisfaction, suggesting that satisfaction with one's current position may be a protective factor against burnout.
Career flexibility in the PA profession has long been embraced by practicing PAs.14 This trend is echoed among PAs practicing in Minnesota, where we found 61.2% identified working in two or more specialties throughout their career. Additionally, we found that Minnesota PAs overwhelmingly (86.9%) believe that the flexibility to move between specialties is an important factor in preventing career burnout. Our data, however, revealed no statistically significant correlation in the burnout subscales between respondents who have worked in multiple specialties. This may be due to differing interpretations of career flexibility, such as work hours, amount of time off, or flexibility in the job. In addition, asking the question in a yes/no structure rather than as a scaled variable limited analysis and interpretations for nuance.
Although we used a tool with validity evidence to measure burnout and had a response rate consistent with or stronger than recently published studies about burnout among PAs, our data have some limitations.11,12 For example, the Minnesota Board of Medical Practice database had email addresses for only 1,166 of the 2,242 PAs in the state. Our response rate of 31.4% ends up being only 13.9% of PAs in Minnesota, which may contribute to unmeasured selection bias.17 However, despite this limited sample size, our study closely reflects statewide and national demographics.17,18 Also, the self-report nature of this survey introduces nonresponse bias. For example, those who chose to respond to the survey may have stronger feelings about the survey content than those who chose not to respond. Additionally, we set up a skip pattern at the start of the survey, asking respondents if they were practicing; if they said “no,” their survey ended without completing the MBI questions. Unsolicited comments from several of these 23 respondents indicated that they had retired early or changed careers due to burnout, suggesting that our burnout data may underestimate the presence of burnout. Similarly, our study may underrepresent the younger population of PAs, as only 24% of survey respondents were under age 35 years; that same year, 40% of PAs practicing in Minnesota were under age 35 years.17
A limitation of our study is that given study population constraints (such as inclusion criteria and possible sample size) and large number of specialties, we were unable to fully assess burnout across all individual specialties. The small sample size compared with number of specialties is the main reason we chose to group specialties into primary care versus nonprimary care. A national study with a larger sample size may allow for more analysis of individual specialties.
The cross-sectional nature of this study limits our ability to determine causality but suggests opportunities for future prospective studies to identify specific causes of burnout. Following PAs from the beginning of their careers through changes in specialties or employers would allow us to better understand trends in burnout over time. Additionally, follow-up studies could explore specific factors of the work environment that may affect levels of burnout, such as the number of hours worked per week, work-life balance, administrative or clerical duties, and workplace or supervisor characteristics.
Comparing the findings of burnout studies is difficult because of the lack of a standardized definition of burnout. A recent review article found that researchers used five predominant approaches to the definition of burnout in healthcare.24 Future studies should strive to employ a standardized definition of burnout.
This study is the first to evaluate the prevalence of burnout among practicing PAs in a single state. The research shows that PAs in Minnesota display moderate levels of burnout while maintaining high levels of career satisfaction. This study also found that PAs perceive career flexibility to be negatively associated with burnout. A troubling parallel was found: burnout is experienced by female physicians and PAs working in primary care. This is particularly interesting, as the high amount of burnout among PAs in primary care could indicate a reason for fewer providers choosing a primary care position. As we look to address the deficit of providers in primary care shortage areas, administrators should consider interventions to reduce this trend.
The clinical implication of our study is to create awareness of PA career satisfaction, rates of burnout, and the importance of career flexibility. This research is important for the NCCPA and the AAPA to consider when making decisions that may hinder the flexibility of the profession. Further research into the topic of burnout among PAs should examine a national population, compare burnout rates between PAs and other medical providers such as physicians and NPs, consider additional factors as potential causes of burnout, and examine the influence of job satisfaction and burnout on patient care.
3. Shanafelt T, Sloan J, Satele D, Balch C. Why do surgeons consider leaving practice. J Am Coll Surg
4. Shanafelt TD, Hasan O, Dyrbye LN, et al Changes in burnout
and satisfaction with work-life balance in physicians and the general US working population between 2011 and 2014. Mayo Clin Proc
5. Shanafelt T, Dyrbye L. Oncologist burnout
: causes, consequences, and responses. J Clin Oncol
6. Jiang H, Li C, Gu Y, Lu H. Nurse satisfaction and burnout
in shanghai neurology wards. Rehabil Nurs
7. Maslach C, Jackson SE. The measurement of experienced burnout
. J Organ Behav
8. 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
9. Bell RB, Davidson M, Sefcik D. A first survey: measuring burnout
in emergency medicine physician assistants. JAAPA
10. Benson MA, Peterson T, Salazar L, et al Burnout
in rural physician assistants: an initial study. J Physician Assist Educ
11. Tetzlaff ED, Hylton HM, DeMora L, et al National study of burnout
and career satisfaction
among physician assistants in oncology: implications for team-based care. J Oncol Pract
12. Coplan B, McCall TC, Smith N, et al Burnout
, job satisfaction, and stress levels of PAs. JAAPA
13. Hooker RS, Kuilman L, Everett CM. Physician assistant
job satisfaction: a narrative review of empirical research. J Physician Assist Educ
14. Warner ML, Maio C, Hudmon KS. Career patterns of physician assistants: a retrospective longitudinal study. JAAPA
16. Shanafelt TD, Mungo M, Schmitgen J, et al Longitudinal study evaluating the association between physician burnout
and changes in professional work effort. Mayo Clin Proc
18. National Commission on Certification of Physician Assistants. 2016 statistical profile of certified physician assistants. http://www.nccpa.net/research
. Accessed March 12, 2019.
19. Maslach C, Jackson SE, Leiter M, et al Maslach Burnout Inventory Manual
. 3rd ed. Palo Alto, CA: Consulting Psychologists Press; 1996.
20. Adwan JZ. Pediatric nurses' grief experience, burnout
and job satisfaction. J Pediatr Nurs
21. LaBarbera DM. Physician assistant
vocational satisfaction. JAAPA
22. Rabatin J, Williams E, Baier Manwell L, et al Predictors and outcomes of burnout
in primary care
physicians. J Prim Care Community Health
23. Linzer M, Poplau S, Grossman E, et al A cluster randomized trial of interventions to improve work conditions and clinician burnout
in primary care
: results from the healthy work place (HWP) study. J Gen Intern Med
24. Doulougeri K, Georganta K, Montgomery A. “Diagnosing” burnout
among healthcare professionals: can we find consensus. Cogent Med
Keywords:Copyright © 2019 American Academy of Physician Assistants
physician assistant; PA; burnout; career satisfaction; job flexibility; primary care