Twenty-five facilities enrolled sufficient staff for inclusion (≥10 respondents), allowing 928 NHCP respondents to be included (49% CNAs, 29% nurses, 8% therapists, and 13% nonclinical roles) (Table 1). The overall percentage of staff who reported an ITL in their current job (ITL rate) was 36%, ranging from 25% to 51% among the 7 academic clusters. Thirty-five percent of NHCPs were classified as being burned out (cluster range, 20%−56%). Most (70%) NHCPs reported experiencing at least 1 type of incivility within the previous 6 months, with 23% reporting some incivility and 47% reporting high incivility rates. Frequency of incivility and source of incivility (eg, other staff, leadership, family, resident) varied by each of the 10 incivility items (Supplementary Figure 1). The most common types of incivility included the deliberate withholding of information (eg, best practices, resident preferences) necessary for the participant’s job (46%) and anger or rage directed at the participant (33%), with 9% reporting at least monthly occurrences.

Lower rates of ITL were reported by older (aged 50–59 years) NHCPs compared with NHCPs aged 18 to 29 years (Table 2). In unadjusted analyses, rates of ITL were found to significantly vary by race, dependent status, and job type, but not by sex or marital status. Increased burnout, incivility, and lower resilience were all significantly associated with ITL, as was a facility 4-star overall quality rating. There was a trend for facility staffing rating to be associated with ITL; however, the median values were similar (median 4-star facilities among both participants with ITL and those without ITL, P = .046) (Table 2).

Multivariate adjusted models incorporating independent predictors of ITL included burnout (adjusted odds ratio [aOR], 3.82; 95% CI, 2.61–5.59) and experiencing high rates of incivility (aOR, 2.05; 95% CI, 1.32–3.19). Job type was also a significant predictor in the model, with all nurses (registered nurses and licensed practical nurses grouped together) (aOR, 1.53; 95% CI, 1.01–2.33) and therapists (occupational, physical, and speech/language) (aOR, 2.12; 95% CI, 1.13–3.97) more likely to report ITL than CNAs. Facility staffing rating was also significantly associated with ITL, with study participants working in facilities with higher staffing star ratings reporting a higher ITL with an aOR of 1.31 for each incremental increase in star rating from a baseline of 1 (95% CI, 1.08–1.60). Race, dependent status, and resilience were no longer significantly associated with ITL in multivariate analysis. However, they are retained in the best subset model listed in Table 3.

Health care workforce shortages are particularly pronounced and intractable in long-term care, where 72% of nursing homes report that their workforce remains lower than prepandemic levels despite most facilities’ attempts to recruit personnel by increasing wages, bonuses, benefits, and promotion and growth opportunities.¹ By using validated measures of both positive and negative aspects of workplace well-being, and qualitative responses, this study contributes to a growing body of literature on the postpandemic health care environment. Our results highlight the importance of approaches to improving the working environment, particularly for nurses and therapists, and for younger employees.

We found that 36% of NHCPs reported that they would leave their job within a year if they could. Therapists and phlebotomists were nearly twice as likely to report ITL compared with CNAs, even after controlling for demographics, burnout, incivility, and facility characteristics. This finding is consistent with a national study showing that therapists of mixed licensure (eg, speech therapists, occupational therapists) reported among the highest rates of burnout of US health care workers during the pandemic.³⁶ Importantly, therapists comprise only 8% of our sample and inclusion of these staff in our analysis does not affect the associations measured between ITL and other job types. Although CNAs were less likely to report an ITL than nurses or therapists, 34% of CNAs surveyed in our study reported that they would leave within the next year if they could. Previous research has documented the high levels of stress and low perceived value experienced by CNAs,³⁶ especially those working in long-term care.³⁷

Burnout had the largest association with ITL in our study, consistent with and adding to a corpus of research that consistently finds this association.^14,15 Thirty-five percent of respondents in our study reported burnout, which is alarming given that burnout among health care workers in long-term care may be associated with higher rates of medical errors, poorer patient/resident outcomes, and lower care quality and safety.^38,39

Workplace incivility and mistreatment are consistently associated with burnout across licensure and professional statuses.^21–23 It is striking, then, that even after controlling for job type, duration at facility, burnout levels, and sociodemographic variables, providers in our study who experienced high rates of incivility were more than twice as likely to report an intention to leave. In the current study, the most common types of incivility reported by NHCPs were less severe experiences, such as having information deliberately withheld, spreading rumors, or facing hostile reactions from others, but these experiences, when frequent, were associated with increased ITL. These data highlight the importance of interventions and policies that promote psychological safety, the shared belief within a team that it is safe to take interpersonal risks such as asking questions, sharing ideas, and reporting mistakes or problems.⁴⁰ Workplace incivility is also associated with nursing home cultures of abuse and neglect and is increasingly considered an indicator of quality and safety.⁴¹ Taken together, one of the most actionable and consequential findings from our study adds to accumulating data on the importance of addressing incivility and mistreatment to address provider retention.⁴²

Although self-reported resilience was associated with ITL in the univariate model, it was no longer significant in the multivariate model. This was unexpected because previous studies have found that resilience is associated with job satisfaction and is protective against burnout and ITL among health care workers.^43–45 In our sample, self-reported resilience was less important than factors related to workplace atmosphere and well-being, such as incivility and emotional exhaustion. Of note, our qualitative findings indicate that NHCPs who reported an intent to stay in their role overwhelmingly reported positive affect in their reasoning, using language such as love and joy, and adding to a consistent body of research showing that positive emotional responses to the workplace lead to an intention to stay in health care roles.^2,46

The most common reasons participants reported for an ITL were to pursue other opportunities and/or to improve compensation or benefits. Notably, fewer participants reported an ITL due to moral distress or poor climate, somewhat inconsistent with our quantitative finding that incivility was highly associated with ITL. This may reflect reporting differences, with NHCPs being less likely to report or reflect on incivility or related experiences in an open-ended format without explicit prompting. Other factors mentioned in the open-ended responses, such as the working environment, seeking better compensation, and the physical toll of the job, were cited as reasons for intending to leave, and are consistent with other studies examining attrition among nurses and CNAs in nursing homes.^47,48 Although few participants cited the COVID-19 pandemic as a reason for staying or leaving the profession, those who did stated that their prepandemic workplace atmosphere had not returned, and cited lingering exhaustion, stress, and workforce shortages as a result. Surprisingly, the best subset regression model found that facility staffing rates were positively associated with ITL. We are unsure how to interpret this unexpected association, but caution is warranted. It is noteworthy that the median quality and staffing ratings were nearly identical between those reporting an ITL and those indicating an intention to stay.

Older NHCPs surveyed in our study were less likely to report ITL. Although this is consistent with a wealth of studies showing that younger providers are more likely to report an intention or desire to leave their job,^49,50 caution is warranted in interpreting these findings. Open-ended responses indicate that a less common but explicitly named reason for reporting ITL was retirement, just as a common reason for reporting an intent to stay in the job was having invested in their current job and not wanting to start anew. Taken together, these responses suggest that lower rates of ITL among older NHCPs may not necessarily reflect high levels of job satisfaction. Still, they may instead reflect situational or employment factors influencing an intention to stay. In fact, there is evidence that age is not a predictor of job satisfaction among providers working in long-term care.⁵¹

This study had notable strengths, including a large sample size from diverse geographic areas, quantitative and qualitative data that enabled a richer and more comprehensive understanding, and multilevel analyses that quantified the role of incivility on ITL after controlling for burnout. However, several limitations warrant consideration. First, we measured self-reported ITL without tracking actual professional changes or attrition. Second, workplace incivility was assessed retrospectively, which may introduce bias (eg, social desirability, recall). Finally, the voluntary nature of the survey may have led to self-selection bias, and we do not have access to payroll-based staffing to report response rates that likely varied across sites and by role.

In this mixed-methods, cross-sectional study, over one-third of NHCPs reported an ITL their job, which was highest among nurses and therapists, and lowest among older NHCPs. Burnout and incivility were common and increased ITL risk, regardless of role, tenure, or demographics. Reducing burnout and incivility may promote well-being and help lower NHCP attrition.

Supplementary Data

Supplementary data related to this article can be found online at https://doi.org/10.1016/j.jamda.2025.105962.