Prevalence of and Factors Associated With Nurse Burnout in the US

Key Points Question What were the most recent US national estimates of nurse burnout and associated factors that may put nurses at risk for burnout? Findings This secondary analysis of cross-sectional survey data from more than 3.9 million US registered nurses found that among nurses who reported leaving their current employment (9.5% of sample), 31.5% reported leaving because of burnout in 2018. The hospital setting and working more than 20 hours per week were associated with greater odds of burnout. Meaning With increasing demands placed on frontline nurses during the coronavirus disease 2019 pandemic, these findings suggest an urgent need for solutions to address burnout among nurses.


Introduction
Clinician burnout is a threat to US health and health care. 1 At more than 6 million in 2019, 2 nurses are the largest segment of our health care workforce, making up nearly 30% of hospital employment nationwide. 3 Nurses are a critical group of clinicians with diverse skills, such as health promotion, disease prevention, and direct treatment. As the workloads on health care systems and clinicians have grown, so have the demands placed on nurses, negatively affecting the nursing work environment. When combined with the ever-growing stress associated with the coronavirus disease 2019 (COVID-19) pandemic, this situation could leave the US with an unstable nurse workforce for years to come. Given their far-ranging skill set, importance in the care team, and proportion of the health care workforce, it is imperative that we better understand job-related outcomes and the factors that contribute to burnout in nurses nationwide.
Demanding workloads and aspects of the work environment, such as poor staffing ratios, lack of communication between physicians and nurses, and lack of organizational leadership within working environments for nurses, are known to be associated with burnout in nurses. 4,5 However, few, if any, recent national estimates of nurse burnout and contributing factors exist. We used the most recent nationally representative nurse survey data to characterize burnout in the nurse workforce before COVID-19. Specifically, we examined to what extent aspects of the work environment resulted in nurses leaving the workforce and the factors associated with nurses' intention to leave their jobs and the nursing profession.

Data Source
We used data from the 2018 US Department of Health and Human Services' Health Resources and Service Administration National Sample Survey of Registered Nurses (NSSRN), a nationally representative anonymous sample of registered nurses in the US. The weighted response rate for the 2018 NNRSN is estimated at 49.0%. 6 Details on sampling frame, selection, and noninterview adjustments are described elsewhere. 7 Weighted estimates generalize to state and national nursing populations. 6 The American Association for Public Opinion Research Response Rate 3 method was used to calculate the NSSRN response rate. 6 This study of deidentified publicly available data was determined to be exempt from approval and informed consent by the institutional review board of Emory University. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cross-sectional studies. Data were collected from April 30 to October 12, 2018.

Variables and Definitions
We generated demographic characteristics from questions about years worked in the profession, primary and secondary nursing positions, and work environment. We included the work environment variables of primary employment setting and full-time or part-time status. We grouped responses to a question on dominant nursing tasks as direct patient care, other, and no dominant task. We included 3 categories of educational attainment (diploma/ADN, BSN, or MSN/PhD/DNP degrees) and whether the respondent was internationally educated. Other variables included change in employment setting in the last year, hours worked per week, and reasons for employment change.
We categorized employment setting as (1) hospital (not mental health), (2) other inpatient setting, (3) clinic or ambulatory care, and (4) other types of setting. Workforce stability was defined as the percentage of nurses with less than 5 years of experience in the nursing profession.
We used 2 questions to assess burnout and other reasons for leaving or planning to leave a nursing position. Nurses who had left the position they held on December 31, 2017, were asked to identify the reasons contributing to their decision to leave their prior position. Nurses who were still employed in the position they held on December 31, 2017, and answered yes to the question "Have ). Further, we examined the overlap of the proportions who reported leaving or considered leaving their job owing to burnout and other factors. We then fit 2 separate logistic regression models to estimate the odds that aspects of the work environment, hours, and tasks were associated with the following outcomes related to burnout: (1) left job owing to burnout and (2) considered leaving their job owing to burnout. We controlled for nurse demographic characteristics of age, sex, race, household income, and geographic region and reported odds ratios (ORs) and 95% CIs. Two separate sensitivity analyses were performed: (1) we used a broader theme of burnout defined as a response of burnout, inadequate staffing, or stressful work environment for the regression models; and (2) we stratified the regression models by respondents younger than 45 years and 45 years or older to examine difference by age.

JAMA Network Open | Health Policy
We used SAS, version 9.4 (SAS Institute, Inc), with statistical significance set at 2-sided α = .05.
We used sample weights to account for the differential selection probabilities and nonresponse bias.

Results
Of proportions of nurses reported burnout in the West (16.6%), and higher proportions reported burnout in the Southeast (30.0%) (Figure 1 and Figure 2). Figure 3 shows the overlap between leaving or considering leaving their position owing to burnout and other reasons. For both outcomes, the highest overlap response with burnout was for stressful work environment (68.6% of those who left their job and 63.0% of those who considered leaving their job due to burnout).
The adjusted regression models estimating the odds of nurses indicating burnout as a reason for leaving their positions or considering leaving their position revealed statistically significant associations between workplace settings and hours worked per week, but not for tasks performed, and burnout (  who worked more than 40 h/wk had an OR of 3.28 (95% CI, 1.61-6.67) for identifying burnout as a reason they left their position.
For nurses who reported ever considering leaving their job, working in a hospital setting was associated with 80% higher odds of burnout as the reason than for nurses working in a clinic setting   The sensitivity analysis results in which a broader classification of burnout was used showed a similar relationship between odds of burnout and working more than 40 h/wk (OR, 3.86; 95% CI, 2.27-6.59) for those who left their job (OR, 2.66; 95% CI, 2.13-3.31). Stratification by those younger than 45 years and 45 years or older did not significantly change the findings. Figure 3 shows the overlap in nurses who reported burnout and other reasons for leaving their current position or considering leaving their current positions. The greatest overlap occurred in responses of burnout and stressful work environment (68.6% of those who reported leaving and 59.5% of those who considered leaving) and inadequate staffing (63.0% of those who reported leaving and 60.9% of those who considered leaving).

Discussion
Our findings from the 2018 NSSRN show that among those nurses who reported leaving their jobs in 2017, high proportions of US nurses reported leaving owing to burnout. Hospital setting was associated with greater odds of identifying burnout in decisions to leave or to consider leaving a nursing position, and there was no difference by dominant work function.
Health care professionals are generally considered to be in one of the highest-risk groups for experience of burnout, given the emotional strain and stressful work environment of providing care to sick or dying patients. 8,9 Previous studies demonstrate that 35% to 54% of clinicians in the US   experience burnout symptoms. [10][11][12][13] The recent National Academy of Medicine report, "Taking Action Against Clinician Burnout: A Systems Approach to Professional Well-Being," recommended health care organizations routinely measure and monitor clinician burnout and hold leaders accountable for the health of their organization's work environment in an effort to reduce burnout and promote well-being. 1 Moreover, it appears the numbers have increased over time. Self-reported regional variation in burnout deserves attention. The lower reported rates of nurse burnout in California and Massachusetts could be attributed to legislation in these states regulating nurse staffing ratios; California has the most extensive nurse staffing legislation in the US. 20 The high rates of reported burnout in the Southeast and the overlap of burnout and inadequate staffing in our findings could be driven by shortages of nurses in the states in this area, particularly South Carolina and Georgia. 15 Geographic distribution, nurse staffing, and its association with selfreported burnout warrant further exploration.
Our data show that the number of hours worked per week by nurses, but not the dominant function at work, was positively associated with identifying burnout as a reason for leaving their position or considering leaving their position. also focus on aspects of the work environment known to improve job satisfaction, including staffing ratios, continued nursing education, and support for interdisciplinary teamwork.

Limitations
Our study has some limitations. First, our findings are from cross-sectional data and limit causal inference; however, these data represent the most recent and, to our knowledge, the only national survey with data on nurse burnout. Second, our burnout measure is crude, and more extensive measures of burnout are needed. Third, 4 states did not have enough respondents to release data (Montana, Wyoming, North Dakota, and South Dakota). However, these data were weighted, and they represent the most comprehensive data available on the registered nurse workforce. Fourth, nonresponse analyses of these data reveal underestimation of certain races/ethnicities, specifically Hispanic nurses, and small sample sizes limited analyses of burnout by race/ethnicity. Fifth, the public use file of the NSSRN does not disaggregate the MSN, PhD, and DNP degrees in nursing practice categories. Given that these job tasks can vary, we addressed this limitation by examining dominant function at work. Last, the response rate was modest at 49.0% (weighted). Despite these limitations, this analysis is most likely the first to provide an updated overview of registered nurse burnout across the US.

Conclusions
Burnout continues to be reported by registered nurses across a variety of practice settings nationwide. How the COVID-19 pandemic will affect burnout rates owing to unprecedented demands on the workforce is yet to be determined. Legislation that supports adequate staffing ratios is a key part of a multitiered solution. Solutions must come through system-level efforts in which we reimagine and innovate workflow, human resources, and workplace wellness to reduce or eliminate burnout among frontline nurses and work toward healthier clinicians, better health, better care, and lower costs. 32