Gender Differences in the Relationship Between Workplace Civility and Burnout Among VA Primary Care Providers

Abstract

Background

Civility, or politeness, is an important part of the healthcare workplace, and its absence can lead to healthcare provider and staff burnout. Lack of civility is well-documented among mostly female nurses, but is not well-described among the gender-mixed primary care provider (PCP) workforce. Understanding civility and its relationship to burnout among male and female PCPs could help lead to tailored interventions to improve civility and reduce burnout in primary care.

Objective

To analyze gender differences in civility, burnout, and the relationship between civility and burnout among male and female PCPs.

Design

Multi-level logistic regression analysis of a cross-sectional national survey.

Participants

A total of 3216 PCP respondents (1946 women and 1270 men) in 135 medical centers from a 2019 national Veterans Health Administration (VA) survey.

Main Measures

Outcomes: burnout; predictors: workplace civility and gender; controls: race, ethnicity, VA tenure, and supervisory status.

Key Results

Workplace civility was rated higher (p<0.001) among male (mean = 4.07, standard deviation [SD] = 0.36, range 1–5) compared to female (mean = 3.88, SD = 0.33) PCPs. Almost half of the sample reported burnout (47.6%), but this difference was not significant (p = 0.73) between the genders. Higher workplace civility was significantly related to lower burnout among female PCPs (odds ratio [OR] = 0.46, 95% confidence interval [CI] = 0.31 to 0.69), but not among male PCPs (OR = 0.71, 95% CI = 0.42 to 1.22). Interactions between civility and other demographic variables (race, ethnicity, VA tenure, or supervisory status) were not significantly related to burnout.

Conclusion

Female PCPs report lower workplace civility than male PCPs. An inverse relationship between civility and burnout is present for women but not men. More research is needed on this phenomenon. Interventions tailored to gender- and primary care-specific needs should be employed to increase civility and reduce burnout among PCPs.

INTRODUCTION

Civility, which is conduct marked by courtesy, politeness, and respect, is a crucial element of the healthcare workplace. Lack of civility in the workplace, sometimes described as workplace incivility, is defined by rude or discourteous behavior that violates workplace norms. These actions can spiral out of control as workers subject to uncivil behaviors become uncivil themselves.¹ Civility is also an increasingly important concept in primary care, as many primary care providers (PCPs) work in teams in the patient-centered medical home (PCMH) model in the Veterans Health Administration (VA) system,² and in the community.³

Most research on incivility in healthcare has been conducted among nurses.^4–6 Incivility, as well as bullying and rudeness, has been well-documented in the nursing workforce,⁷ and linked to increased nurse burnout,⁸ and decreased nurse-assessed patient safety, and quality of care.⁹ Worldwide, workplace incivility affects an estimated 68–90% of nurses, and over 75% have experienced peer incivility.¹⁰ In the USA, fewer estimates are available, but over 50% of nurses surveyed have experienced incivility in the workplace.¹⁰ Incivility is more common among less experienced nurses, but organizational protective factors, like good leadership, organizational support for nursing, and good communication at work, are associated with less incivility.¹¹

Although the link between workplace incivility and burnout has been well documented in primarily female¹² nursing workforce, incivility has not been well-studied among primary care providers (PCPs), where the gender mix is broader. Incivility has also been linked to burnout,^9,13 a psychological response to workplace stress, which is in turn linked to increased medical errors, lower patient satisfaction, reduced productivity, and higher turnover.^14–16 Burnout is high among providers, affecting between 20 and 40% of those who work in primary care,¹⁷ and over 50% of PCPs in the VA.¹⁸ In this paper, we assessed levels of perceived incivility and burnout among VA PCPs by gender, and explored the relationships between civility and burnout among these PCPs.

METHODS

We conducted retrospective cross-sectional analysis using the 2019 All Employee Survey (AES), which the VA annually fields as a census of employees.¹⁹ We focused on all primary care respondents, for whom there was an estimated 60.9% response rate (N=25,976). To build our sample of PCPs, we identified respondents who indicated that “primary care” was the main type of service they provide and that they worked as a physician, nurse practitioner, or physician assistant.

We examined a main outcome of burnout, main predictors of workplace civility and gender, and demographic controls of race, ethnicity, VA tenure, and supervisory status. Respondents were coded as burned out, as in similar studies,²⁰ if they reported emotional exhaustion or depersonalization “once a week” or more from a response set ranging from “never” to “every day.” Perceived workplace civility was constructed based on responses to five items asking about cooperation, accepting differences, conflict resolution, and psychological safety (Cronbach’s α = 0.91). We aggregated workplace civility to the medical center level and retained sites with at least five respondents (intraclass correlation [ICC] 1 = 0.08; ICC 2 = 0.67). Respondents also provided responses to demographic characteristics of gender, race, ethnicity, VA tenure, and supervisory status. We omitted cases when gender was missing and used an indicator of “unknown” for missing responses. The burnout and workplace civility survey items used in this study are presented in Table 1.

Table 1.

Burnout and Workplace Civility Survey Items

Burnout

1. I feel burned out from my work 2. I worry that this job is hardening me emotionally Items rated on seven-point ordinal scale ranging from “Never” to “Every day.”

Workplace civility

1. People treat each other with respect in my workgroup. 2. Disputes or conflicts are resolved fairly in my workgroup. 3. The people I work with cooperate to get the job done. 4. Members in my workgroup are able to bring up problems and tough issues. 5. Discrimination is not tolerated at my workplace. Items rated on a five-point Likert scale ranging from Strongly Disagree to Strongly Agree.

In two separate multi-level logistic models for men and women, we assessed associations between civility and burnout. We regressed individual-level burnout on medical center-level workplace civility and individual-level demographic characteristics. Additionally, we explored additional gender-specific models with a series of interactions between workplace civility and demographics to understand if other characteristics interacted with civility to explain variance in burnout.

RESULTS

Our final sample size was 3216 PCP respondents (response rate 60.9%) from 135 medical centers. The sample (Table 2) consisted of 65.9% physicians (MD/DO), 27.3% nurse practitioners (NP), and 6.8% physician assistants (PA). Approximately 60% of respondents were female, 65.4% were white, and 10.9% held a supervisory position. Medical center-level civility was rated higher by men (mean = 4.07, standard deviation [SD] = 0.36) compared to women (mean = 3.88, SD = 0.33; p < 0.001). Nearly half of PCPs reported burnout (47.6%). The difference in burnout was non-significant between men (47.2%) and women (47.8%; p = 0.73).

Table 2.

Descriptive Characteristics of the Sample (N=3216)

Measure	Overall	Male (N=1270)	Female (N=1946)
	%	%	%
Burnout	47.6	47.2	47.8
Occupation
MD/DO	65.9	83.9	54.1
NP	27.3	9.0	39.3
PA	6.8	7.1	6.6
Race
White	65.4	62.9	69.2
Black	7.2	4.9	8.6
Asian	18.8	17.7	19.4
Other	3.4	3.4	3.5
Unknown	5.2	4.7	5.6
Ethnicity
Non-Hispanic	84.3	82.4	85.6
Hispanic	6.5	8.1	5.4
Unknown	9.2	9.0	9.5
VA tenure
<1 year	11.4	9.9	12.4
1–2 years	8.9	7.6	9.8
2–5 years	23.1	21.9	23.8
5–10 years	20.2	20.0	20.3
10–20 years	26.2	27.6	25.3
20+ years	8.9	11.7	7.1
Unknown	1.4	1.4	1.3
Supervisor status
Yes	10.9	15.6	7.9
No	87.6	83.5	90.2
Unknown	1.5	0.9	1.9

Abbreviations: DO, Doctor of Osteopathy; MD, Doctor of Medicine; NP, nurse practitioner; PA, physician assistant; SD, standard deviation; VA, Veterans Health Administration

In stratified, multi-level models, we found that greater workplace civility was significantly associated with lower odds of burnout for women (odds ratio [OR] = 0.46, 95% confidence interval [CI] = 0.31 to 0.69; p < 0.001; Table 3), reflecting that in workplaces with higher civility, women were less impacted by burnout. Workplace civility, however, was not significantly related to burnout among men (OR = 0.71, 95% CI = 0.42 to 1.22; p = 0.19). Across occupational groups, we observed that female nurse practitioners reported lower burnout compared to female physicians (OR = 0.65, 95% CI = 0.52 to 0.80). Female supervisors also reported lower burnout rates (OR = 0.60, 95% CI = 0.42 to 0.86). Among other demographic characteristics, we observed Black male (OR = 0.54, 95% CI = 0.30 to 0.95) and Asian female (OR = 0.61, 95% CI = 0.47 to 0.79) providers reporting lower burnout, compared to white male and female providers, respectively. For both male and female providers, burnout was lowest among employees with less than 1 year of tenure. We did not observe significant differences in burnout by other demographic variables when examining interactions between workplace civility and these variables in additional exploratory models.

Table 3.

Hierarchical regression model for burnout in male and female primary care providers

Measure	PCPs-Male		PCPs-Female
	OR	95% CI	OR	95% CI
Workgroup civility score	0.71	(0.42- 1.22)	0.46*	(0.31 - 0.69)
Occupation: (Ref=MD/DO)
Nurse practitioner	0.85	(0.56 - 1.29)	0.65*	(0.52 - 0.80)
Physician assistant	0.98	(0.62 - 1.56)	1.00	(0.67 - 1.49)
Spanish, Hispanic, or Latino: (Ref=No)
Yes	0.90	(0.58 -1.08)	0.92	(0.60 - 1.40)
Unknown	1.42	(0.94 - 2.17)	1.27	(0.90 - 1.79)
Race: (Ref=White)
Asian	0.79	(0.57 - 1.08)	0.61*	(0.47 - 0.79)
Black	0.54*	(0.30 - 0.95)	0.71	(0.50 - 1.01)
Other	1.06	(0.55 - 2.03)	0.99	(0.58 - 1.69)
Unknown	1.55	(0.85 - 2.83)	1.30	(0.82 - 2.06)
VA Tenure (Ref=<1 year)
One to 2 years	1.59	(0.88 - 2.86)	1.32	(0.87 - 1.99)
2 to 5 years	2.65*	(1.65 - 4.21)	2.74*	(1.95 - 3.86)
5 - 10 years	2.53*	(1.57 - 4.07)	2.30*	(1.61 - 3.27)
10 – 20 years	2.49*	(1.57 - 3.94)	2.23*	(1.59 - 3.13)
More than 20 years	1.90*	(1.12 - 3.21)	2.18*	(1.39 - 3.43)
Unknown	1.82	(0.63 - 5.21)	3.60*	(1.39 - 9.35)
Supervisor: (Ref=None)
Yes	0.85	(0.61 -1.18)	0.60*	(0.42 - 0.86)
Unknown	1.16	(0.31 - 4.32)	0.60	(0.27 - 1.32)

Abbreviations: DO, Doctor of Osteopathy; MD, Doctor of Medicine; NP, nurse practitioner; PA, physician assistant; PCP, primary care provider; SD, standard deviation; VA, Veterans Health Administration

*p<.0.5

DISCUSSION

Overall, our findings showed that female PCPs reported similar rates of burnout and lower civility scores compared to male PCPs. Female PCPs in workplaces with higher aggregate civility scores were less likely to be burned out, while no such relationship existed among male PCPs or in other demographic groups. Burnout also differed by occupation, race, VA tenure, and supervisor status.

The connection between civility and burnout is especially concerning among female PCPs, as they are at higher risk of burnout overall.¹⁵ It is unclear why burnout and civility are linked in women but not men, but it is possible that women in healthcare place more value on teamwork and collaboration. Female students in the health sciences hold more positive attitudes about teamwork,²¹ as well engaging in education about teamwork.²² The PCMH model of primary care is implemented across the VA as patient-aligned care teams (PACTs), sets of four-person interdisciplinary teams that care for a single panel of patients. Among PACTs, team communication²³ and task delegation²⁴ were associated with lower burnout among PCPs, while participatory decision-making²⁵ was associated with lower burnout among PCPs and staff. Outside of the VA, workplace cohesiveness,²⁶ teamwork,²⁷ and working effectively with colleagues²⁸ are all related to lower burnout among PCPs. Further research is needed to more fully understand the relationship between civility and burnout in women.

Burnout in primary care can lead to many negative healthcare outcomes at the patient, provider, and organizational level. PCP burnout is associated with less favorable patient-provider relationships,²⁹ and an increased likelihood of medical errors and suboptimal patient care.³⁰ Burned out PCPs are also less likely to be satisfied and report a greater intent to leave their practice.³¹ In addition, burned out PCPs often do leave their practice, as they exhibit a nearly 60% increase in the likelihood of turnover after 2–3 years.³²

Improving civility and reducing burnout in the healthcare workplace is achievable, however. The CREW (Civility, Respect, and Engagement at Work) intervention has been shown to increase civility^13,33 and reduce burnout¹³ among groups of mixed-profession healthcare workers in the VA and in Canadian hospitals. CREW is 6-month process consultation intervention of weekly meetings within a local workgroup run by local facilitators, and aided by a national research team of clinical and counseling psychologists with training in organizational development. The research team trains facilitators at individual sites on the CREW intervention and provides them with a toolkit of specific intervention activities. The research team fields a baseline survey with the workgroup to measure civility, burnout, and other organizational attitudes/behaviors. Facilitators receive a summary of their workgroup’s responses, and can use that data to construct their local intervention. Weekly meetings involving toolkit-suggested activities like civility-related education, discussions, and interactive vignettes are run by facilitators with local workgroups. After 6 months, the follow-up assessments are conducted and shared with facilitators. This research team/client relationship allows the research team to impart expert knowledge about civility to local facilitators, and allows local facilitators the freedom to shape the intervention to meet local needs. CREW interventions tailored to meet gender- and primary care-specific needs could increase civility, reduce burnout, and close the civility/burnout gender gap in primary care.

This study has several important strengths: (1) it analyzes a national survey of all VA PCPs; (2) the survey’s response rate of 60.9% is high for a survey of providers; and (3) the survey contains questions on civility and burnout in the same instrument. However, it also has several limitations: (1) the analysis is cross-sectional, which precludes any causal inferences about the associations reported; (2) the survey only asked respondents about workplace civility overall, not about civility between leadership or supervisors and workers, or by gender or any other demographic characteristic; and (3) the study was conducted inside the VA, and findings may not be generalizable to other healthcare systems.

Civility was lower among female VA PCPs compared to male PCPs, and only female PCPs reporting increased civility were less likely to be burned out. Women in healthcare may value teamwork and collaboration more than men, and the combination of lower civility and high teamwork in primary care could lead to more burnout among female PCPs. Additional research is needed to understand the relationship between civility and burnout in women. Tailoring an organizational intervention like CREW to meet gender- and primary care-specific needs could help increase civility and reduce burnout among all PCPs. Teamwork is the present and future of primary care. In this environment, maintaining civility among PCPs is crucial step in reducing primary care burnout.

Funding

Dr. Apaydin was supported by the VA Office of Academic Affiliations through the Advanced Fellowship in Health Services Research & Development. Dr. Yano was supported by a VA Health Services Research & Development Senior Research Career Scientist Award (Project #RCS 05-195).

Declarations

Conflict of Interest

The authors have no conflicts of interest to declare.

Disclaimer

The contents of this article do not represent the views of the US Department of Veterans Affairs or the United States Government.