Registered Nurses’ Well-Being, Michigan, 2022

Abstract

Objectives. To examine nurses’ well-being and identify individual and workplace factors associated with adverse outcomes.

Methods. We administered an e-mail survey to registered nurses in Michigan in March 2022. Outcomes included the Oldenburg Burnout Inventory–Exhaustion scale, self-harm thoughts (yes/no), and overall wellness on a 0 to 10 visual analog scale. Covariates included practice environment, psychological safety, workplace abuse, staffing adequacy, stress coping strategies, and demographics. We examined associations between covariates and exhaustion, thoughts of self-harm (both via logistic regression), and overall wellness (via linear regression).

Results. Among surveyed nurses, 93.63% reported significant exhaustion, 9.88% reported self-harm thoughts, and the mean (SD) overall wellness score was 6.2 (2.3). Factors associated with exhaustion included inadequate staffing, lower psychological safety, and younger age. Factors associated with self-harm thoughts included recent workplace physical abuse and younger age. Factors associated with higher wellness scores included employer support, favorable practice environments, higher job satisfaction, and positive coping strategies.

Conclusions. Negative well-being outcomes were prevalent among registered nurses and were associated with correctable workplace deficits. Nurses’ well-being is a national public health problem that warrants comprehensive interventions at individual, workplace, and community levels. (Am J Public Health. 2024;114(S2):S180–S188. https://doi.org/10.2105/AJPH.2023.307376)

Clinician negative well-being is a major public health problem in the United States. Numerous professional organizations, government officials, and health system leaders have raised collective concerns about the emotional well-being of clinicians.^1,2 The COVID-19 pandemic and associated workplace stressors have exacerbated these concerns.³ Registered nurses are at great risk for poor mental health given their high workloads, long hours, and unfavorable working conditions.⁴ Deaths by suicide are notably higher for female nurses than for both female physicians and the US female population.⁵ The National Academy of Medicine‘s “National Plan for Health Workforce Well-Being” cited a dearth of recent, multisite data to inform interventions and policy strategies.⁶

Despite these concerns and calls for action, surprisingly little recent research has been reported on these phenomena. Specifically, few studies are available in large, representative samples besides single institutional studies about the frequency of mental well-being outcomes.^7,8 From a public health standpoint related to prevention, there are sparse reports to identify individual, workplace, and community factors that place nurses at risk for poor outcomes.⁹ These gaps hinder the development and testing of preventive strategies and interventions specific to this population. Conceptual work on clinician well-being posits that a combination of individual, workplace, and community factors contributes to burnout, moral injury, and adverse mental health outcomes.^6,10

Motivated by these notable knowledge gaps, our team launched the Michigan Nurses Survey in February 2022 to inform subsequent policy changes and interventions. Our research questions were as follows:

• 1.Among practicing registered nurses, what proportion experience emotional exhaustion and thoughts of self-harm?
• 2.How do practicing registered nurses rate their overall wellness?
• 3.What individual and workplace factors are associated with poorer registered nurse well-being, considering emotional exhaustion, thoughts of self-harm, and personal wellness scores?

METHODS

Our survey methods have been published previously.¹¹ Briefly, registered nurses with valid, unrestricted licenses and an e-mail address registered with the state of Michigan as of February 2022 were eligible to participate. We excluded nurses with restricted licenses and those identified in the database as in a disciplinary process.

The study team sent up to 3 e-mail messages to the sample population, 8 days apart, following established methods.¹² To protect identities given the sensitive questions, we used the anonymous feature of the Qualtrics (Provo, UT) online survey platform to blind the study team to e-mail addresses and to prohibit linkage of survey data to identifiers. Because of the large sample size, monetary incentives were not possible, but all those contacted could receive study results regardless of participation.

We asked nurses to report on their primary nursing position. Given the diversity of roles and employment arrangements, survey skip logic omitted certain questions, and nurses could skip any questions that were not pertinent to their role. Hence, some outcomes analyzed have different sample sizes. Individuals who reported high emotional exhaustion, thoughts of self-harm, or low wellness scores received a prompt in the survey with contact information for the state’s 24-hour mental health well-being hotline, as well as the national suicide and crisis hotline. The analyses reported here are of a subset of registered nurses who reported they were currently working in practice.

Study Measures

Our team used the National Academy of Medicine framework for clinician well-being to guide variable selection and the analytic approach.¹³ Specifically, the framework articulates how clinician well-being is associated with both external (society and culture, rules and regulations, organizational factors, practice environment, and health care responsibilities) and individual (personal factors and skills and abilities) factors. This study was designed to inform subsequent intervention development and testing. The literature has consistently affirmed the salience of emotional exhaustion as the key burnout component associated with patient and clinician outcomes.^14–16 Therefore, emotional exhaustion was the primary subcomponent of burnout examined.

Well-Being Outcomes

Among currently practicing registered nurses, we examined 3 outcomes: emotional exhaustion, thoughts of self-harm, and overall wellness. We measured emotional exhaustion by the previously validated Oldenburg Burnout Inventory–Exhaustion scale, which is composed of 8 items scored on a 5-point scale and has values at or above 2.25 indicative of significant exhaustion.^17,18 We treated nurse-reported thoughts of self-harm in the past year as a dichotomous (yes/no) outcome. Finally, nurses rated on a 0 to 10 visual analog scale how they were doing overall (overall wellness).

Workplace Factors

Nurses identified their primary practice setting (e.g., inpatient, long-term care) and rated several aspects of their workplace, informed by previous research.^19,20 These aspects included an overall assessment of their practice environment to deliver high-quality nursing care (unfavorable, mixed, or favorable environment), a validated, 5-item measure of psychological safety in the workplace, measured on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree),^21,22 a validated 3-item measure of staffing and resource adequacy (1 = strongly disagree, 5 = strongly agree), assessment of the quality of care delivered (poor, fair, good, or excellent), and overall job satisfaction on a 5-point Likert scale (1 = not at all satisfied, 5 = very satisfied).¹⁹ Nurses rated whether their employer provided them with timely information about COVID-19 (1 = strongly disagree, 5 = strongly agree). We calculated an employer support score by averaging 7 questions rated on a Likert scale (1 = strongly disagree, 5 = strongly agree) regarding how strongly their employer valued their mental health, physical health, and safety and provided workplace flexibility and wellness resources and receipt of recognition from peers.¹¹

Abusive Events in the Workplace

Using the World Health Organization taxonomy,^23,24 we asked nurses about abusive events in their workplace. We asked nurses whether they had experienced verbal, physical, or sexual abuse in the past 12 months. We also asked nurses whether they experienced bullying from their coworkers. We coded each of these 4 events as dichotomous events (yes/no). We asked those who reported these events whether they had reported the event to their managers (yes/no) and, if so, whether they were satisfied with their manager’s response to the event (yes/no).

Individual Factors

Nurses reported the top 3 activities they used to address stress and whether they received mental health services in the past 6 months. We treated age as a categorical variable, with predefined cutpoints that aligned with the state’s nurse licensure reporting system. We also asked nurses to report their gender identity and whether they held an advanced practice nurse role.

Analyses

We used SAS version 9.4 (SAS Institute, Cary, NC) for all analyses. We computed descriptive statistics for outcomes and independent variables. We considered independent variables based on the National Academy of Medicine framework. To reduce potential collinearity and to identify a parsimonious and interpretable set of explanatory variables, we considered covariate selection using backward selection, using an α threshold of 0.05 to select variables for the reduced model. We performed forward selection and best subset analyses with similar results.

Before performing backward selection, the full (and final) models for thoughts of self-harm had 20 variables (final model = 5 variables), emotional exhaustion had 17 variables (final model = 8 variables), and nurses’ well-being had 31 variables (final model = 13 variables). We used logistic regression models to estimate the likelihood of reporting both emotional exhaustion above the scale’s cutpoint and thoughts of self-harm. Linear regression models estimated wellness scale scores as a function of covariates, with higher scores reflecting better overall wellness. To contextualize the relative importance of individual or workplace factors’ effects on outcomes, we used 3 separate models to calculate R² values for each outcome: the final model, as reported in the Results section; a model with only individual characteristics; and a model with only workplace characteristics.

RESULTS

Of 165 185 e-mail addresses associated with a registered nurse license in Michigan, 17 936 (11%) recipients opened the survey invitation, and of these, 13 687 (76%) completed the survey (8.3% of the surveyed population). From these, 7316 (53% of the respondents) were practicing registered nurses and provided data on the outcomes of interest. For the 3 outcomes of emotional exhaustion, thoughts of self-harm, and overall wellness, 6886 (94% of the eligible respondents), 6889 (94%) and 6582 (90%) had requisite data on outcomes and covariates for analysis, respectively. Table 1 shows participant characteristics and the distribution of workplace factors across the sample.

TABLE 1—

Individual and Workplace Characteristics of Study Participants: Michigan, February–March 2022

	No. (%) or Mean (SD), Median (IQR), Range
Individual characteristics
Age, y
≤ 34	1385 (18.9)
35–44	1611 (22.0)
45–54	1838 (25.1)
55–64	1914 (26.2)
≥ 65	568 (7.8)
Advanced practice nurse
No	6267 (85.7)
Yes	1017 (13.9)
Missing	32 (0.4)
Practice setting
Acute care	3866 (52.9)
Long-term care	331 (4.5)
Community	742 (10.1)
School nursing	112 (1.5)
Education	183 (2.5)
Other/missing	2082 (28.5)
Gender identity
Male	735 (10.1)
Female	6485 (88.6)
Another choicea	96 (1.3)
Stress reduction strategies
Engage in hobbies	3675 (50.2)
Sleep	5081 (69.5)
Spend time with family and friends	5594 (76.5)
Reported emotional exhaustion	6850 (93.6)
Indicated top concern outside of work is the economy	4254 (58.2)
Received mental health counseling in the past 6 months	1607 (22.0)
Workplace characteristics
Employer provided technology and training to do job during COVID-19
Strongly disagree	746 (10.2)
Disagree	1266 (17.3)
Neither agree nor disagree	1593 (21.8)
Agree	2147 (29.4)
Strongly agree	1528 (20.9)
Current practice environment delivers high-quality care
Unfavorable	1289 (17.6)
Mixed	3937 (53.8)
Favorable	2055 (28.1)
Job satisfaction
Very dissatisfied	462 (6.3)
Dissatisfied	1598 (21.8)
Neither dissatisfied or satisfied	1569 (21.5)
Satisfied	2752 (37.6)
Very satisfied	912 (12.5)
Workplace abuse
Physical abuse	1866 (25.5)
Emotional abuse	3600 (49.2)
What factors would better prepare nurses for a future public health crisis?
Increased staffing	5462 (74.7)
Consistent, better executed state public health policies	5002 (68.4)
Psychological safetyb	3.2 (0.8), 3.2 (2.6–3.8), 1–5
Staffing and resource adequacyb	2.5 (1.3), 2.0 (1.0–3.5), 1–5
Employer supportb	3.0 (1.0), 3.0 (2.3–3.7), 1–5

Note. IQR = interquartile range. The sample size was n = 7316.

^a Category was collapsed to protect participant privacy. Choices included male, female, transgender, nonbinary, gender nonconforming, or a write-in option, or it was left blank.

^b Measure is an average of a series of 5-point Likert scale questions with 1 reflecting less favorable views on psychological safety, staffing adequacy, and employer support, and 5 reflecting the most favorable views.

Sparse data are available from Michigan on nursing characteristics. However, the study sample resembles the distribution of age and advanced practice status of the state population. Specifically, 18.9% of the sample were aged 34 years or younger, compared with 23% of the statewide population. Slightly fewer (7.8%) survey participants were aged 65 years or older, compared with the population (13.5%). Compared with the statewide population of 9%, 13.9% of the study sample had advanced practice preparation.

Among 7249 respondents who completed the Oldenburg Burnout Inventory, 6850 (93.6%) practicing nurses reported exhaustion that exceeded the scale’s established cutpoint of more than 2.25. Of 7293 participants who answered the question, 723 (9.8%) reported thoughts about self-harm. Of 7173 participants, the mean (SD) overall wellness rating of nurses was 6.2 (2.3; 0 = not good at all, 10 = very good).

Of the 7302 nurses who reported their strategies to deal with work stressors, 5594 (76.5%) reported spending time with family and friends, 5081 (69.5%) reported sleeping, and 3675 (50.2%) engaged in hobbies.

Factors Associated With Emotional Exhaustion

In multivariable analyses using the sample of practicing nurses with complete data (n = 6886), factors significantly associated with a higher likelihood of emotional exhaustion (scores above the cutpoint of 2.25) included being aged 34 years or younger versus aged 45 to 54 years (odds ratio [OR] = 1.81; 95% confidence interval [CI] = 1.16, 2.83; Table 2), mixed assessment of practice environment (vs unfavorable assessment; OR = 1.76; 95% CI = 1.00, 3.12), agreeing that employer provided technology and training to do their job during COVID-19 (vs strongly disagreeing; OR = 2.14; 95% CI = 1.29, 3.57), and use of sleep as a stress coping strategy (OR = 1.85; 95% CI = 1.47, 2.33; Table 2).

TABLE 2—

Factors Associated With Nurses’ Emotional Exhaustion: Michigan, February–March 2022

	OR (95% CI)
Workplace factors
Employer provided technology and training to do job during COVID-19 (Ref: strongly disagree)
Disagree	1.70 (0.97, 2.98)
Neither agree nor disagree	1.29 (0.88, 2.16)
Agree	2.14 (1.29, 3.57)
Strongly agree	1.03 (0.63, 1.70)
Psychological Safety Scale	0.75 (0.63, 0.89)
Staffing and resource adequacy	0.82 (0.74, 0.92)
Current practice environment to deliver high-quality care (Ref: unfavorable)
Mixed practice environment	1.76 (1.00, 3.12)
Favorable practice environment	0.99 (0.53, 1.82)
Job satisfaction (Ref: very dissatisfied)
Dissatisfied	1.03 (0.39, 2.72)
Neither dissatisfied or satisfied	0.52 (0.20, 1.36)
Satisfied	0.32 (0.13, 0.83)
Very satisfied	0.16 (0.06, 0.42)
Individual factors
Age, y (Ref: 45–54)
≤ 34	1.81 (1.16, 2.83)
35–44	1.10 (0.79, 1.55)
55–64	1.03 (0.76, 1.38)
≥ 65	0.68 (0.47, 0.98)
Stress reduction strategies
Engage in hobbies	0.57 (0.45, 0.72)
Sleep	1.85 (1.47, 2.33)
R2 for final model	0.23

Note. CI = confidence interval; OR = odds ratio. The sample size was n = 6886. Individual’s Oldenburg Burnout Inventory Emotional Exhaustion subscale score was > 2.25.

Factors associated with a lower likelihood of emotional exhaustion included being aged 65 years and older versus 45 to 54 years (OR = 0.68; 95% CI = 0.47, 0.98), higher perceived workplace psychological safety (OR = 0.75; 95% CI = 0.63, 0.89), being very satisfied (OR = 0.16; 95% CI = 0.06, 0.42) or satisfied (OR = 0.32; 95% CI = 0.13, 0.83) in their current position (vs very dissatisfied), increased staffing and resource adequacy (OR = 0.82; 95% CI = 0.74, 0.92), and engaging in hobbies as a stress coping strategy versus not engaging in this strategy (OR = 0.57; 95% CI = 0.45, 0.72). The R² for the final model was 0.23, with workplace factors accounting for approximately 87% of the variance explained.

Factors Associated With Self-Harm Thoughts

The denominator for this outcome was 6889 (Table 3). Physical abuse in the workplace over the past year was significantly associated with increased likelihood of self-harm thoughts (OR = 1.30; 95% CI = 1.09, 1.56). Individual factors associated with increased self-harm thoughts included receipt of counseling in the past 6 months (OR = 3.99; 95% CI = 3.38, 4.72), use of sleep as a stress coping strategy (OR = 1.30; 95% CI = 1.06, 1.59), being aged 34 years or younger versus aged 45 to 54 years (OR = 1.44; 95% CI = 1.41, 1.83). Spending time with family and friends as a stress coping strategy was associated with a lower likelihood of self-harm thoughts (OR = 0.63; 95% CI = 0.52, 0.75). Nurses aged 55 years and older were less likely to report self-harm thoughts (Table 3). The R² for the final model was 0.12, with individual factors accounting for approximately 90% of the variance explained.

TABLE 3—

Factors Associated With Nurses’ Thoughts of Self-Harm: Michigan, February–March 2022

	OR (95% CI)
Workplace factors
Physical abuse in workplace	1.30 (1.09, 1.56)
Individual factors
Age, y (Ref: 45–54)
≤ 34	1.44 (1.41, 1.83)
35–44	1.13 (0.89, 1.42)
55–64	0.76 (0.59, 0.98)
≥ 65	0.58 (0.36, 0.92)
Received mental health counseling in past 6 months	3.99 (3.38, 4.72)
Stress reduction strategies
Spend time with family and friends	0.63 (0.52, 0.75)
Sleep	1.30 (1.06, 1.59)
R2 for final model	0.12

Note. CI = confidence interval; OR = odds ratio. The sample size was n = 6889.

Factors Associated With Overall Wellness

We analyzed this outcome for 6582 participants (Table 4). Factors associated with higher overall wellness scores on the single-item visual analog scale (0 = poor, 10 = excellent) included positive employer support (B = 0.18; 95% CI = 0.12, 0.24), favorable (B = 0.63; 95% CI = 0.46, 0.85) or mixed (B = 0.32; 95% CI = 0.20, 0.50) versus unfavorable work environments, and being satisfied (B = 1.98; 95% CI = 1.73, 2.21) or very satisfied (B = 2.49; 95% CI = 2.19, 2.77) versus very dissatisfied in their job. Strategies associated with higher wellness scores included spending time with family and friends (B = 0.38; 95% CI = 0.25, 0.47) and engaging in hobbies (B = 0.26; 95% CI = 0.17, 0.36). Advanced practice registered nurses (vs registered nurses; B = 0.22; 95% CI = 0.06, 0.33) and those who reported the economy as a top concern outside of work (B = 0.17; 95% CI = 0.07, 0.26) also reported higher overall wellness.

TABLE 4—

Factors Associated With Nurses’ Overall Wellness Scores: Michigan, February–March 2022

	B (SE; 95% CI)
Workplace factors
Employer support	0.18 (0.03; 0.12, 0.24)
Factors that would better prepare nurses for future public health crisis
Increased staffing	−0.35 (0.06; −0.46, −0.23)
Consistent, better executed state public health policies	−0.16 (0.05; −0.26, −0.06)
Current practice environment to deliver high-quality care (Ref: unfavorable)
Mixed practice environment	0.32 (0.08; 0.20, 0.50)
Favorable practice environment	0.63 (0.10; 0.46, 0.85)
Job satisfaction (Ref: very dissatisfied)
Dissatisfied	0.80 (0.11; 0.56, 0.99)
Neither dissatisfied nor satisfied	1.42 (0.12; 1.17, 1.64)
Satisfied	1.98 (0.12; 1.73, 2.21)
Very satisfied	2.49 (0.15; 2.19, 2.77)
Workplace abusive events in the past year
Physical abuse	−0.27 (0.06; −0.40, −0.15)
Emotional abuse	−0.42 (0.06; −0.51, −0.29)
Individual factors
Indicated top concern outside of work is the economy	0.17 (0.05; 0.07, 0.26)
Spend time with family and friends to reduce stress	0.38 (0.06; 0.25, 0.47)
Engage in hobbies to reduce stress	0.26 (0.05; 0.17, 0.36)
Received mental health counseling in past 6 mo	−0.55 (0.06; −0.66, −0.43)
Reported emotional exhaustion	−0.87 (0.11; −1.10, −0.67)
Advanced practice nurse: yes (Ref: no)	0.22 (0.07; 0.06, 0.33)
Age, y (Ref: 45–54)
≤ 34	−0.36 (0.07; −0.50, −0.21)
35–44	−0.15 (0.07; −0.28, −0.01)
55–64	0.17 (0.07; 0.04, 0.30)
≥ 65	0.52 (0.10; 0.33, 0.72)
R2 for final model	0.34

Note. CI = confidence interval. The sample size was n = 6582.

Factors associated with lower overall wellness scores included receipt of mental health services in the past 6 months (B = −0.55; 95% CI = −0.66, −0.43), report of physical (B = −0.27; 95% CI = −0.40, −0.15) and emotional (B = −0.42; 95% CI = −0.51, −0.29) abuse in the workplace in the past year, being aged 34 years or younger (B = −0.36; 95% CI = −0.50, −0.21) or 35 to 44 years (B = −0.15; 95% CI = −0.28, −0.01) versus 45 to 54 years, and higher scores on the emotional exhaustion scale (B = −0.87; 95% CI = −1.10, −0.67). The R² for the final model was 0.34, with workplace factors accounting for approximately 89% of the variance explained.

DISCUSSION

Using a statewide sample of registered nurses surveyed in 2022, we identified a substantial number of practicing nurses who reported adverse well-being, specifically, high rates of emotional exhaustion. Importantly, several individual and workplace factors were associated with overall wellness and with increased likelihood of emotional exhaustion or self-harm thoughts.

The findings highlight the importance of ongoing, systematic evaluation of the well-being of registered nurses. The Breen Act, passed by Congress and signed by the president, would fund new awareness campaigns, assessment and treatment programs, and new investigations into health care worker well-being.²⁵ Many of these initiatives—which include additional descriptive and interventional research specific to health care workers—have yet to issue new funding announcements or report results from existing efforts at the time of this article’s publication. The importance of these existing initiatives and expanding on them cannot be overstated because fundamental epidemiological research on mental health outcomes among registered nurses is scant, as is an evidence base of interventions that target nurses and health care workers.²⁶ Differential patterns of methods used during deaths by suicide vary notably between health care workers and the general population.⁵ Toxic substances are used more frequently in health care workers’ suicides, whereas the general population more frequently uses firearms. Risk reduction strategies that are targeted to health care workers are promising strategies for this higher-risk population.

The association of recent receipt of mental health services with several outcomes likely reflects an encouraging finding that nurses with poorer mental health outcomes are appropriately seeking professional services. The access, equity, and effectiveness of these interventions remain understudied. Nor is it clear whether mental health professionals who treat health care workers recognize the unique combination of individual and workplace characteristics that place them at risk for poor outcomes.²⁷

Across all analyses, younger nurses reported poorer well-being. This suggests that specific strategies to support newer entrants into nursing may be warranted, including peer support, stress reduction, comprehensive workplace orientation, and transition programs. Poorer well-being among younger nurses threatens the stability of the US nursing workforce, as nurses with poorer well-being are more likely to leave their positions.¹¹

Nurses reported emotional well-being was poorer in the setting of workplace abuse, including physical and emotional events. Amid increased frequency and severity of violence in health care settings,²⁴ having institutional, state, and federal policies to reduce violent workplace events is an important strategy to optimize clinician well-being. Our results showed that individual factors appear to explain more of the variance for self-harm thoughts; conversely, workplace factors explained more of the variance in emotional exhaustion and overall wellness. These preliminary findings warrant confirmation in studies that have larger samples of nurses and are designed to test multilevel research questions.

Interestingly, some coping strategies were associated with improved outcomes (i.e., hobbies, spending time with family and friends), whereas use of increased sleep was associated consistently with poorer outcomes. The mechanistic underpinnings for these findings are not yet clear. These findings may be integrated into multicomponent interventions designed to support nurses’ well-being.

Limitations and Strengths

The key limitations to this work are a relatively low response rate, consistent with the e-mail–based study design. The response rate is comparable to other large-scale nurse surveys,²⁸ and survey participants resemble the nurse population in Michigan.^11,29 The statewide sampling approach overcomes important limitations of single-institution studies or convenience samples, which may be subject to bias. Outcomes are participant reported, without correlation with clinical records or formal diagnoses. Data were not available on employers or geographic location, which could be important contextual factors. The cross-sectional design precludes temporal relationships between risk factors and outcomes. Strengths include a large statewide sample, use of validated measures, and timely data collection and analysis.

To inform public health interventions, subsequent investigations should include detailed measures on nurses’ use of mental health services and workplace supports (e.g., employee assistance programs) and the perceived effectiveness of these services. Longitudinal designs may help clarify the temporal sequence of factors that contribute to poor well-being.

Public Health Implications

Deficits in registered nurses’ well-being constitute a public health emergency and warrant a comprehensive, multilevel response. Given the high prevalence of negative well-being outcomes among practicing registered nurses in this study, we propose opportunities for individual-, workplace-, and community-level strategies.

Individual-level interventions

As some outcomes were associated with individual factors, such as female gender identity and younger age, targeted interventions can be developed to support individuals at risk for poor outcomes. Although participants identified positive coping strategies, workplace exhaustion makes it difficult to engage in hobbies and visit friends and family. Clinicians who have nurses as clients should recognize the intersection of individual and workplace factors that may contribute to poorer outcomes and assess for workplace stressors and workplace abuse as part of their therapeutic work. Given female nurses’ increased risk of suicide and self-harm, clinicians could incorporate routine assessment of self-harm thoughts.⁵

Workplace interventions

Workplace leaders need to recognize the high prevalence of emotional exhaustion and adopt mitigation strategies.³ Nurses’ well-being requires that foundational, or basic, needs be met before addressing higher-level needs. From an organizational view, leaders can eliminate mandatory overtime, reduce requests for nurses to work extra shifts, enforce planned breaks, decrease the pressure to take on missed care from other staff, and reduce administrative tasks that are not essential to patient care.³⁰ Approaches that improve reporting and follow-up actions on abusive events, coupled with using abusive events reporting to inform subsequent antiviolence strategies, should be considered.

The National Academy of Medicine’s “National Plan for Health Workforce Well-Being”⁶ and the “Surgeon General’s³¹ Framework for Workplace Mental Health & Well-Being” set forth priorities to assist health care organizations develop, institutionalize, and update policies, processes, and practices to support the well-being of health care workers. For example, our data demonstrated the significant relationship between psychological safety and emotional exhaustion. One of the 5 essentials in the Surgeon General’s framework, “protection from harm,” directly addresses psychological safety—creating conditions for physical and psychological safety through actions such as ensuring adequate rest, making it safe to speak up in the workplace, and normalizing a focus on mental health. As another example, we found that positive employer support—typified by employer support for nurses’ safety and well-being—was associated with higher overall well-being.

A second essential from the Surgeon General’s framework is “mattering at work”—people want to know that their work matters, which is associated with lower stress. Actions to demonstrate mattering at work include engaging workers in workplace decisions, salaries commensurate with responsibilities and risks, and showing gratitude and recognition that is meaningful to workers.

Proactive leadership approaches include recognizing that generic self-help, wellness, and resilience programs will not solve nurses’ physical and emotional exhaustion. Workplace wellness programs need to be engaging, easily accessible, and offered confidentially and to include staff perspectives in their design and scope.

Community interventions

The National Academy of Medicine’s plan identifies the societal implications of health care worker well-being: “It is our ethical obligation to take action to protect those who care for all [of] us.”^32(p71) Although few studies have examined community-based approaches to improve nurse well-being, societal interventions are a promising strategy to help nurses maintain their health and promote their well-being. Nurses have routinely cited challenges in family caregiving and waning community support, especially during the pandemic.³ Akin to parallel programs available to first responders and veterans, local governments and social service agencies could offer wraparound supportive services (e.g., child or dependent care programs) to nurses who serve their communities.³³

In conclusion, the registered nurse workforce is an essential component of the US health care system. Presently, nurses face multiple threats to well-being, which in turn can disrupt health care and service delivery across the nation. Multilevel interventions are necessary to address these complex phenomena, with the goals of reduced burnout, improved well-being, and retention of this vital component of the health care workforce.