Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 May 23;82(3):2216–2231. doi: 10.1111/jan.17067

Latent Profiles of Nurses' Insomnia, Fatigue, Recovery, Psychological Distress and Burnout During the COVID‐19 Pandemic: Examining the Role of Organisational Support

Hyeonmi Cho 1, Knar Sagherian 2, Linsey M Steege 3,✉
PMCID: PMC12907604  PMID: 40405789

ABSTRACT

Aims

To identify latent profiles of insomnia, fatigue, recovery, psychological distress and burnout among hospital nurses; examine variations in personal and work‐related characteristics across profiles; investigate associations between profiles and outcomes such as patient care quality, nursing work satisfaction and workplace satisfaction; and assess the moderating role of organisational support on these relationships.

Design

A cross‐sectional descriptive study.

Methods

This study analysed survey data from 2488 hospital nurses using latent profile analysis, multinomial logistic regression and hierarchical multiple linear regression.

Results

Four well‐being profiles emerged: low, average, above‐average and high well‐being. Profiles differed significantly in personal and work‐related characteristics. Nurses in the high well‐being profile were associated with better patient safety, care quality and satisfaction. Organisational support moderated the negative associations between low well‐being profiles and nursing work and workplace satisfaction.

Conclusion

Tailored interventions addressing factors associated with low well‐being and enhancing organisational support may be beneficial for improving nurse well‐being, delivering high‐quality care and supporting nurse retention in sustainable healthcare environments.

Implications

Healthcare organisations should prioritise nurse well‐being through targeted interventions, adequate staffing, recovery opportunities and stress management resources to support a resilient and sustainable workforce.

Impact

The findings revealed the diversity of well‐being patterns among hospital nurses and provided insights for identifying subgroups at higher risk of impaired patient safety, reduced care quality and dissatisfaction with nursing work and the workplace. Greater organisational support was associated with weaker negative relationships between poor well‐being and nurse outcomes.

Reporting Method

STROBE guidelines.

Patient or Public Contribution

No direct patient or public contribution.

Keywords: burnout, fatigue, insomnia, nurses, psychological distress, recovery

1. Introduction

Nurses, who represent the largest professional group in the global health workforce, are responsible for maintaining patient safety and delivering high quality care (World Health Organisation 2020). However, demanding workloads, stressful work environments and limited opportunities for rest, often resulting from long shifts and poor scheduling, have negatively impacted nurses' health and well‐being (Alrabae et al. 2021; Brossoit et al. 2020; Cho et al. 2024). These challenges were further magnified during the nearly 3‐year COVID‐19 pandemic (Fernandez et al. 2020). Consequently, there is a growing need to better understand the multifaceted nature of nurses' well‐being and its implications for patient care and satisfaction in the workplace.

2. Background

Subjective well‐being, as defined by Diener (1984), encompasses overall life satisfaction and a sense of happiness. This evaluative process involves both cognitive and affective assessments (positive and negative emotions) of personal experiences, events and life circumstances (Diener et al. 2002). In the context of hospital nursing, poor well‐being is often reflected in burnout, fatigue, poor sleep and psychological distress, factors that are deeply interconnected and mutually influential. For example, hospital nurses' chronic fatigue and burnout are strongly related to each other (Rutledge et al. 2022). Fatigue and sleep problems like insomnia are associated with anxiety and depressive symptoms (Zou et al. 2021) and short sleep durations can impair acute fatigue recovery and increase the risk of burnout (Chin et al. 2015; Sagherian et al. 2021). Despite these interdependencies, many studies have examined individual well‐being factors, such as burnout or fatigue, in isolation, overlooking their complex interplay (Cho and Steege 2021; Chin et al. 2015; Sagherian et al. 2021; Zou et al. 2021). Therefore, clarifying how these factors co‐occur and interact as part of distinct well‐being profiles among hospital nurses is a critical gap that warrants investigation.

Latent Profile Analysis (LPA) offers a person‐centred approach to identifying these distinct well‐being profiles (Berlin et al. 2014; Nylund et al. 2007). By categorising nurses into subgroups that share similar patterns across multiple well‐being indicators, LPA uncovers hidden heterogeneity and highlights how various factors cluster together (Nylund et al. 2007). Understanding these profiles not only provides insight into the complex nature of nurses' well‐being but also enables the development of tailored interventions that address the specific needs of each subgroup.

Moreover, research studies have demonstrated how these aforementioned factors (i.e., insomnia, fatigue, burnout and psychological distress) have negative implications on patient safety, decreased quality of care and decreased job satisfaction among others (Brossoit et al. 2020; Cho and Steege 2021; Jun et al. 2021; Wang et al. 2023). When identified, these well‐being profiles can also be examined in relation to patient‐ and nurse‐related outcomes, as well as aid in the development of strategies that address the specific needs of at‐risk subgroups.

The job demands‐resources (JD‐R) model (Bakker et al. 2005) suggests that workplace resources can lessen the adverse effects of job demands on employee outcomes. One such resource is organisational support, defined as employees' perceptions that their organisation recognises their efforts and is invested in their well‐being (Eisenberger et al. 1986). Many previous studies have demonstrated how organisational support relates to nurses' well‐being and other work‐related outcomes. For example, a recent Chinese study (N = 1058) reported that perceived organisational support, particularly superior and affective organisational support, was associated with lower burnout levels among hospital nurses (Zhang et al. 2024). Another study of 703 Turkish healthcare workers showed that organisational support was associated with lower burnout levels and moderated the association between perceived workload and emotional exhaustion (Doğan et al. 2024). During the COVID‐19 pandemic, higher perceived organisational support was associated with reduced odds of insomnia among Chinese frontline healthcare workers and showed a protective moderating effect between long work hours and fatigue (Zou et al. 2021). Similarly, a study of 474 Chinese hospital nurses identified a moderating effect of organisational support between the practice environment and general well‐being (Sheng et al. 2023). A Swedish study further revealed that a supportive organisational climate predicted increased future job satisfaction and decreased burnout symptoms among hospital nurses after 1 year of follow‐up (Larsman et al. 2024).

Although these findings highlight the importance of organisational support, most prior research has focused on its direct or moderating effects on single well‐being indicators. There has been relatively little exploration of how organisational support might interact with multidimensional well‐being profiles, despite its potential to buffer negative outcomes more comprehensively. Examining this moderating role in the context of nurses' various well‐being factors can provide a more comprehensive perspective on the interplay between workplace resources, individual well‐being and key outcomes in hospital settings.

3. The Study

Based on these considerations, we propose four hypotheses: (1) different profiles of insomnia, fatigue, recovery, psychological distress and burnout exist among hospital nurses, (2) nurses belonging to these distinct profiles differ in personal and work‐related characteristics, (3) nurses in different well‐being profiles exhibit varying outcomes regarding quality of patient care, nursing work satisfaction and workplace satisfaction, and (4) organisational support moderates the relationship between these profiles and their associated outcomes. Through this approach, our study aims to address the gap in understanding how multiple well‐being factors co‐occur, how they differ across nurse subgroups and how organisational support might influence the relationship between these profiles and important outcomes.

The specific aims of this study were as follows: (1) to identify the latent profiles of insomnia, fatigue, recovery, psychological distress and burnout among hospital nurses; (2) to examine variations in personal and work‐related characteristics among the identified profiles; (3) to investigate the associations between these profiles and quality of patient care, nursing work satisfaction and workplace satisfaction; and (4) to assess the moderating effect of organisational support on the relationships between the profiles and their associated outcomes.

4. Methods

4.1. Research Design

A cross‐sectional descriptive design was used. This study analysed online survey data and was reported following the Strengthening the Reporting of Observational Studies in Epidemiology guidelines. A larger parent study aimed to explore the effects of the COVID‐19 pandemic on hospital nurses' health and well‐being and the resulting impact on nursing workforce outcomes, patient care and healthcare organisations (Sagherian et al. 2024).

4.2. Sample and Setting

This study included registered nurses (RNs) working in US hospital settings and providing direct patient care in the United States. No restrictions were placed on hospital characteristics (i.e., hospital size, geographic location and Magnet status) to capture a diverse range of work environments. Nurses on paid and unpaid leave were excluded to ensure actively practising hospital nurses were included. Additionally, nurses in formal leadership or advanced practice roles and those working in outpatient settings were excluded. This allowed the study to focus specifically on RN staff delivering inpatient bedside care on hospital units.

4.3. Data Collection

Data were collected from 21 July to 12 September 2021, via Qualtrics XM (Qualtrics, Provo, UT). To avoid automated or duplicate submissions, we enabled Qualtrics security features such as ‘Prevent multiple submissions’ and ‘Prevent indexing’. Nurses were recruited through nine email listservs from state nursing organisations and government agencies, as well as various nursing‐focused Facebook groups. To ensure eligibility, we included mandatory screening questions that required respondents to confirm if they were RNs delivering direct patient care in a US hospital. Potential participants who answered ‘No’ to any of these criteria (e.g., not actively practicing as an RN, not working in a hospital or not providing direct care as a staff nurse) were deemed ineligible and were not taken to the main survey. Nurses were also encouraged to share the online survey link with their colleagues to enable snowball sampling.

This approach enabled us to gather data from nurses nationwide via an online survey. Several Qualtrics features were used to ensure security, including preventing multiple submissions, disabling indexing and anonymising responses, so that personal information (e.g., IP addresses) was not recorded. The survey link provided study details, and participants who partially completed the survey could return later to finish.

A total of 3987 clicks were recorded on the survey link. After the initial screening, 825 were deemed ineligible based on the eligibility criteria and were excluded, and 429 otherwise eligible individuals did not continue beyond the initial survey pages, leaving 2733 participants who began the survey. Subsequently, 23 were excluded for being on leave, 5 for working outside a hospital and 217 for providing no responses. The final sample included 2488 participants.

Previous guidelines suggested that a minimum sample size of 500 could be sufficient for reliable latent class enumeration under specific conditions (Nylund et al. 2007). More recent simulation studies have indicated that the optimal sample size can vary depending on the complexity of the model (e.g., number of indicators, class separation) and the research context (Tein et al. 2013; Wurpts and Geiser 2014). With a large sample of 2488 nurses, we exceed earlier recommendations and are well positioned to identify underlying profiles accurately while considering model complexity and the need for robust statistical power.

4.4. Ethical Considerations

The institutional review board of the relevant university determined that this study was exempt (2021‐0494). The online survey link provided detailed information about informed consent and the research team's contact details. The survey was conducted anonymously, and no personally identifiable information was collected. Participants who agreed by selecting ‘Yes’ to consent were able to complete the survey. They were also given the opportunity to enter a voluntary drawing for one of 30 $50 Amazon gift cards.

4.5. Bias Assessment and Mitigation

We recognise the potential for common method bias of self‐reported data. To evaluate for bias, we conducted the Harman single‐factor test (Tang and Wen 2020) which identified nine factors with eigenvalues above 1. The first factor accounted for 33.0% of the variance, which is below the commonly used 40% threshold. These results suggest that common method bias does not pose a substantial threat in our study. We attempted to broaden participation by engaging multiple organisations and encouraged snowball sampling; however, self‐selection remains a limitation here. Because of self‐reported data, measurement bias may be a threat if participants under‐ or over‐reported sensitive issues such as fatigue or distress. To mitigate this threat, we used instruments with robust psychometric properties and assured anonymity to reduce potential social desirability bias. We also offered a voluntary drawing for one of 30 $50 Amazon gift cards as a modest incentive. We made participation optional and collected email addresses via a separate link to protect anonymity. However, this incentive may have attracted participants with certain characteristics and potentially introduced some selection bias.

4.6. Measures

4.6.1. Insomnia

The seven‐item Insomnia Severity Index (ISI; Morin et al. 2011) was administered using a 5‐point Likert scale reflecting the past month. Total scores range from 0 to 28, where higher scores indicate more severe insomnia. Scores of 0–7 denote no insomnia, 8–14 mild, 15–21 moderate and 22–28 severe insomnia. The ISI has established strong psychometric properties, including internal consistency and construct and concurrent validity (Morin et al. 2011), and Cronbach's alpha in our sample was 0.87.

4.6.2. Fatigue and Recovery

Nurses' fatigue and recovery levels were measured using the Occupational Fatigue and Exhaustion Recovery (OFER‐15; Winwood et al. 2005, 2006). The OFER‐15 comprises 15 items divided into three subscales—Chronic Fatigue (CF), Acute Fatigue (AF) and Inter‐shift Recovery (IR)—with five items each, rated on a 7‐point Likert scale. Total subscale scores range from 0 to 100, with higher scores reflecting greater fatigue or better inter‐shift recovery. Scores are categorised as follows: 0–25 (low), 26–50 (low to moderate), 51–75 (moderate to high) and 76–100 (high). The OFER‐15 has robust psychometric properties, including internal consistency, test–retest reliability, and construct and discriminant validity (Winwood et al. 2005, 2006). Cronbach's alpha values for the subscales were above 0.86.

4.6.3. Burnout

The Maslach Burnout Inventory‐Human Services Survey (MBI‐HSS) assessed nurses' burnout using 22 items across three subscales: emotional exhaustion (EE), depersonalisation (DP) and personal accomplishment (PA) (Maslach and Jackson 1981). Responses were rated on a 7‐point Likert scale, with higher total scores in each subscale indicating greater levels of EE, DP and PA. Each subscale is classified into low, middle and high categories based on the 33rd and 66th percentiles as cut‐off points (Maslach and Jackson 1986). In our sample, the 33rd and 66th percentiles for EE were 30 and 43, for DP were 7 and 15 and for PA were 30 and 37. The MBI‐HSS has demonstrated reliability and validity in nurses (Loera et al. 2014). In our sample, Cronbach's alpha values for the subscales ranged from 0.78 to 0.93.

4.6.4. Psychological Distress

Psychological distress (PD) was measured with the Patient Health Questionnaire‐4 (PHQ‐4; Kroenke et al. 2009; Löwe et al. 2010), which includes two depression and two anxiety items rated on a 4‐point Likert scale. The total score ranges from 0 to 12, categorised as 0–2 (no distress), 3–5 (mild), 6–8 (moderate) and 9–12 (severe). The PHQ‐4 demonstrates strong reliability and validity (Kroenke et al. 2009; Löwe et al. 2010), and Cronbach's alpha in this study was 0.88.

4.6.5. Overall Patient Safety

Nurses' perceptions of overall patient safety in their units were evaluated with a single question: ‘How would you rate your unit's patient safety?’ This was measured on a 5‐point Likert scale ranging from poor to excellent. The question was adapted from the Hospital Survey on Patient Safety Culture (Alrabae et al. 2021; Sharp et al. 2019).

4.6.6. Nurse‐Reported Quality of Care

The quality of patient care, as perceived by nurses, was assessed using the question: ‘How would you rate the nursing care you personally provided during your most recent shift?’ Responses were recorded on a 4‐point Likert scale, ranging from 1 (poor) to 4 (excellent). This single‐item measure has been widely used in previous nursing studies and has shown a strong correlation with patient health outcomes (McHugh and Stimpfel 2012; Smeds‐Alenius et al. 2016).

4.6.7. Satisfaction With Nursing Work and Workplace

Nurses' satisfaction with their nursing work and workplace was assessed using single‐item questions. For satisfaction with nursing work, nurses rated their agreement with the statement ‘I am satisfied with the nursing work’ on a scale from 1 (strongly disagree) to 7 (strongly agree). Similarly, for workplace satisfaction, they rated their agreement with the statement ‘I am satisfied with this workplace’ on the same scale. Job‐related satisfaction among nurses has often been measured using single‐item questions in previous studies (Lu et al. 2019; Yasin et al. 2020).

4.6.8. Organisational Support

Nurses' perceptions of organisational support were measured using the short version of the Survey of Perceived Organisational Support (Eisenberger et al. 1986; Neves and Eisenberger 2012). This 8‐item scale is rated on a 7‐point Likert scale, with higher total scores indicating greater perceived support. The instrument has demonstrated strong psychometric properties, including robust validity and high internal consistency reliability (Eisenberger et al. 1986; Neves and Eisenberger 2012). In our sample, the Cronbach's alpha for this measure was 0.93.

4.6.9. Personal and Work‐Related Characteristics

Nurses' personal and work‐related characteristics were assessed, including age, sex (male or female), marital status (not married, married or with a partner), nursing education (diploma/associate degree, bachelor's, master's or higher), work status (part‐time or full‐time), night shift (yes or no), shift type (day/night only, rotating or other), weekly work hours in the past 30 days (40 h or less, more than 40 h), hospital Magnet status (Magnet, on the journey towards Magnet or not Magnet/do not know; Magnet status is a designation awarded by the American Nurses Credentialing Center [ANCC] to healthcare institutions that meet rigorous standards of excellence in nursing practice [ANCC, n.d.]), hospital location (rural, sub‐rural or urban), hospital size bed size (hospital size bed size [100 or less, 101–250, 251 or more]), unit type (medical‐surgical, emergency, ICU/step‐down, obstetrics/gynaecology or labour‐delivery, specialty units or float/1‐day services), frequency of caring for COVID‐19 patients (1 = never to 6 = always) and likelihood of taking 30‐min breaks (1 = never to 5 = always). Perceived workload was measured using the Quantitative Workload Inventory, a validated and reliable scale (Spector and Jex 1998), with a Cronbach's alpha of 0.89 in this sample.

4.7. Statistical Analysis

Data were analysed using Mplus version 8.5 (Muthén and Muthén 1998–2020) and IBM SPSS Statistics for Windows Version 28.0 (IBM Corp., 2021). Descriptive statistics for sample characteristics and key variables were presented using means, standard deviations (SDs), frequencies and percentages.

To identify latent profiles for insomnia, fatigue, recovery, psychological distress and burnout, an LPA was performed. The LPA included eight variables—insomnia, CF, AF, IR, PD, EE, DP and PA—which were standardised using z‐scores. Starting with a single profile, additional profiles were added incrementally until an optimal solution was achieved using maximum likelihood estimation.

The final number of profiles was determined by evaluating statistical criteria, interpretability and parsimony, comparing models with different numbers of profiles. The Lo–Mendell–Rubin likelihood ratio test (LMRT) was used to compare models, with a p‐value below 0.05 indicating that the more complex model (k‐class) provided a better fit than the simpler one (k‐1 class). Model selection was guided by the Akaike information criterion (AIC), Bayesian information criterion (BIC), sample size‐adjusted BIC (ssBIC) and entropy (Burnham and Anderson 2004). Lower values for AIC, BIC and ssBIC indicated better model fit, while entropy values above 0.80 were considered acceptable for class assignment accuracy (Berlin et al. 2014). As an additional measure of classification quality, the mean probabilities for each assigned class were recommended to exceed 0.80 (Nylund et al. 2007). To address missing data in LPA, Full Information Maximum Likelihood (FIML) estimation was used to ensure unbiased parameter estimation, even with incomplete data.

Multinomial logistic regression used the latent profiles as the categorical dependent variable, with nurses' personal and work‐related characteristics as independent variables. The independent variables included in the multinomial logistic regression were selected based on univariate analyses (p < 0.3) and prior empirical evidence indicating their associations with nurses' well‐being (Gifkins et al. 2020; Farag et al. 2022, 2024; Lindqvist et al. 2015; Rodríguez‐García et al. 2020; Sagherian et al. 2023).

To examine the relationships between latent profiles and patient care and job outcomes and the moderating role of organisational support, hierarchical multiple linear regression was performed separately for overall patient safety, nurse‐reported quality of patient care, nursing work satisfaction and workplace satisfaction. Two models were created for each outcome variable: Model 1 included latent profiles and organisational support, while Model 2 added two‐way interactions between latent profiles and organisational support to assess the moderating effect. All models were adjusted for nurses' personal and work‐related characteristics, such as age, gender, marital status, nursing education, employment status, shift type, night shift status, hospital's Magnet status, hospital location, unit type, frequency of caring for COVID‐19 patients, workload and frequency of taking 30‐min breaks during work. In both the multinomial logistic and hierarchical multiple linear regression models, continuous variables were standardised using z‐scores. Missingness across the variables used in the multinomial logistic and hierarchical multiple linear regression models ranged from under 1% to 26%. For variables with 5% or more missing data, missing values were imputed using the expectation–maximisation (EM) algorithm (Schafer 1999). In contrast, variables with under 5% missing data, primarily demographic measures such as age, marital status and Magnet status of the hospital, were handled via listwise deletion. This approach was adopted to reduce potential bias associated with more extensive missingness while minimising unnecessary imputation.

When significant interactions were found in the hierarchical multiple linear regression, simple slope analyses were conducted at one SD below the mean (low), at the mean and one SD above the mean (high) of organisational support to probe its moderating effect on the relationship between profiles and patient care and job outcomes.

All continuous variables used in the analyses were evaluated for normality by examining skewness and kurtosis values, which were within ±2. Visual inspection of histograms and Q–Q plots indicated no major deviations from normality, suggesting the distributions were acceptable. For the linear regression assumptions, scatterplots of standardised residuals versus predicted values, Durbin–Watson statistics and normal probability plots demonstrated no serious violations of linearity, independence or homoscedasticity. Cook's distance values were also below 0.5, indicating an absence of highly influential outliers.

5. Results

5.1. Participant Characteristics

The study sample included nurses with an average age of 41.50 years (SD = 12.32, range = 20–76), of whom 89.7% were female. Most nurses were married or living with a partner (65.2%) and held a Bachelor's degree as their highest nursing qualification (65.2%). The majority worked full‐time (84.6%), and 70.3% of nurses worked 40 h or less per week. About half of the nurses worked in Magnet or Magnet‐journey hospitals (51.3%), in urban settings (46.8%) and in hospitals with 251 or more beds (51.1%). Regarding unit types, 26.4% were in medical/surgical units and 34.6% were in intensive care or step‐down units. The average perceived workload was 21.44 (SD = 4.27, range = 5–25). Approximately 32.1% of nurses reported caring for patients with COVID‐19 frequently or always over the past few months, and 40.6% reported rarely or never taking rest breaks during shifts.

5.2. Latent Profile Analysis

The model fit parameters for the LPA are summarised in Table 1. Although the five‐profile model had a significant LMRT, the test became non‐significant for the six‐profile model (p = 0.145), suggesting no further improvement in fit beyond five profiles. Subsequently, we examined additional indicators (entropy, AIC, BIC and ssBIC) and visually inspected an elbow plot. The entropy value for the five‐profile solution fell below 0.80 (0.793), indicating diminished classification clarity (Berlin et al. 2014). Moreover, the elbow plot showed that improvements in AIC, BIC and ssBIC between the four‐ and five‐profile solutions were minimal, indicating limited incremental benefit relative to added complexity (Figure S1). Conceptually, the five‐profile solution did not offer greater interpretability. Taken together, considerations of statistical fit, entropy, parsimony and interpretability led to the conclusion that the four‐profile solution provided the most appropriate representation of the data.

TABLE 1.

Model fit parameters for latent profile analysis.

No. profiles AIC BIC ssBIC Entropy LMRT p‐value Smallest size, n (%) Mean class assignment probabilities
1 53,614 53,708 53,657
2 47,619 47,764 47,685 0.852 < 0.001 845 (34%) 0.95, 0.97
3 45,571 45,769 45,661 0.852 0.001 295 (11.9%) 0.94, 0.92, 0.94
4 44,510 44,760 44,623 0.828 0.002 165 (6.6%) 0.95, 0.91, 0.86, 0.92
5 44,126 44,429 44,264 0.793 0.002 132 (5.3%) 0.94, 0.84, 0.89, 0.83, 0.88
6 43,902 44,257 44,063 0.760 0.145 132 (5.3%) 0.89, 0.93, 0.81, 0.76, 0.86, 0.75
Open in a new tab

Note: This table compares one‐ through six‐profile solutions for insomnia, fatigue, recovery, psychological distress and burnout among hospital nurses. Bold indicates the model selected.

Abbreviations: AIC, Akaike information criterion; BIC, Bayesian information criterion; LMRT, Lo–Mendell–Rubin adjusted likelihood ratio test; ssBIC, sample‐size adjusted Bayesian Information Criterion.

Figure 1 presents the characteristics of four profile groups, showing the z‐score distribution for insomnia, CF, AF, IR, PD, EE, DP and PA. The table below provides the mean and SD for each variable. Each profile group was labelled based on comparisons of average values with the overall sample mean.

FIGURE 1.

FIGURE 1

Open in a new tab

Latent profiles for insomnia, fatigue, recovery, psychological distress and burnout. The bar graphs illustrate the standardised (z‐score) mean values for each variable in the four identified well‐being profiles. The table below provides the corresponding unstandardised means, with standard deviations in parentheses, for each measure within each profile.

The high well‐being profile (n = 165, 6.6%) had the lowest averages for insomnia, CF, AF, PD, EE and DP, and the highest for IR and PA among the profile groups. The above‐average well‐being profile (n = 605, 24.0%) had below‐average values for insomnia, CF, AF, PD, EE and DP, while IR and PA were above the average among the profile groups. All variables were close to the sample mean in the average well‐being profile (n = 1000, 40.2%). The low well‐being profile (n = 718, 28.9%) had the highest averages for insomnia, CF, AF, PD, EE and DP, and the lowest for IR and PA among the profile groups.

5.3. Multinomial Logistic Regression of the Identified Latent Profiles Concerning the Personal and Work‐Related Characteristics of Nurses

Table 2 shows the multinomial logistic regression results for factors associated with profile membership, using the high well‐being profile as the reference.

TABLE 2.

Multinomial logistic regression of the identified latent profiles concerning the personal and work‐related characteristics of nurses.

Variables Above‐average well‐being vs. high well‐being (Rf.) Average well‐being vs. high well‐being (Rf.) Low well‐being vs. high well‐being (Rf.)
Odds ratio 95% CI Odds ratio 95% CI Odds ratio 95% CI
Age 0.81 * 0.67–0.98 0.77 ** 0.63–0.93 0.63 ** 0.51–0.78
Sex
Male 0.73 0.43–1.24 0.57 * 0.33–0.99 0.33 ** 0.18–0.61
Female (ref.)
Marital status
Not married 0.75 0.50–1.12 0.98 0.65–1.47 0.98 0.64–1.51
Married or with partner (ref.)
Education level in nursing
Diploma or associate degree 1.10 0.59–2.04 1.35 0.71–2.56 1.29 0.65–2.59
Bachelor's degree 1.02 0.59–1.76 1.09 0.62–1.93 1.09 0.59–2.03
Master's or doctoral degree (ref.)
Work status
Part‐time 0.89 0.54–1.48 0.80 0.48–1.34 0.57 * 0.32–0.995
Full‐time (ref.)
Night shift
No 0.56 0.27–1.17 0.75 0.36–1.59 0.79 0.36–1.76
Yes (ref.)
Shift type
Day or night only 1.03 0.50–2.13 0.99 0.47–2.06 0.87 0.39–1.92
Rotating 0.78 0.40–1.51 1.05 0.53–2.07 1.21 0.59–2.48
Others (ref.)
Weekly work hours in the past 30 days
40 h or less 0.96 0.61–1.49 1.30 0.82–2.04 1.29 0.80–2.08
More than 40 h (ref.)
Magnet status of the hospital
Magnet or on the journey towards Magnet 0.89 0.60–1.32 0.75 0.50–1.13 0.58 * 0.38–0.89
Not a Magnet hospital or do not know (ref.)
Hospital location
Rural 0.71 0.43–1.18 0.78 0.46–1.31 0.68 0.39–1.18
Sub rural 1.34 0.87–2.06 1.03 0.66–1.60 1.07 0.67–1.71
Urban (ref.)
Unit
Medical surgical 0.85 0.41–1.77 1.27 0.59–2.72 1.13 0.49–2.62
Emergency department 0.80 0.31–2.03 1.02 0.39–2.67 0.90 0.32–2.52
Intensive care unit or step down 0.99 0.48–2.05 1.23 0.57–2.63 1.22 0.53–2.80
OBGYN or labour‐delivery 1.35 0.58–3.13 1.77 0.73–4.28 1.30 0.49–3.43
Specialty units 0.82 0.38–1.78 1.46 0.64–3.35 1.64 0.65–4.12
Float/1‐day services and procedures (ref.)
Workload 1.54 ** 1.30–1.82 3.83 ** 3.16–4.63 9.61 ** 7.46–12.37
Frequency of caring for patients with COVID‐19 1.06 0.87–1.28 1.11 0.91–1.35 1.29 * 1.05–1.60
Likelihood of 30‐min breaks during work time 0.68 ** 0.56–0.83 0.65 ** 0.54–0.80 0.51 ** 0.43–0.67
Open in a new tab

Note: Odds ratios indicate how each independent variable is associated with being in a specific latent profile compared with the high well‐being profile. The reference category for each categorical variable in parentheses. Bold indicates statistically significant results.

Abbreviation: CI, confidence interval.

*

p < 0.05.

**

p < 0.01.

Compared to nurses in the high well‐being profile, older nurses demonstrated lower odds of belonging to the above‐average well‐being (odds ratio [OR] = 0.81, 95% confidence interval [CI] = 0.67–0.98), average well‐being (OR = 0.77, 95% CI = 0.63–0.93) and low well‐being (OR = 0.63, 95% CI = 0.51–0.78) profiles. Similarly, male nurses were less likely than female nurses to be in either the average (OR = 0.57, 95% CI = 0.33–0.99) or low well‐being (OR = 0.33, 95% CI = 0.18–0.61) profiles, relative to the high well‐being reference. Part‐time nurses also had reduced odds of belonging to the low well‐being profile (OR = 0.57, 95% CI = 0.32–0.995) compared to full‐time nurses.

With respect to work environment, nurses working in Magnet‐designated or Magnet‐journey hospitals were less likely to be in the low well‐being profile (OR = 0.58, 95% CI = 0.38–0.89) than those working in non‐Magnet facilities. Higher reported workload showed a strong positive association with being in any profile other than high well‐being, including the above‐average (OR = 1.54, 95% CI = 1.30–1.82), average (OR = 3.83, 95% CI = 3.16–4.63) and low well‐being (OR = 9.61, 95% CI = 7.46–12.37) profiles. Moreover, more frequent care for patients with COVID‐19 increased the likelihood of belonging to the low well‐being profile (OR = 1.29, 95% CI = 1.05–1.60). In contrast, nurses who took 30‐min breaks more often had significantly lower odds of being in the above‐average (OR = 0.68, 95% CI = 0.56–0.83), average (OR = 0.65, 95% CI = 0.54–0.80) and low (OR = 0.51, 95% CI = 0.43–0.67) well‐being profiles, relative to the high well‐being group.

5.4. Associations Between Latent Profiles and Patient Care and Job Outcomes and the Moderating Role of Organisational Support

In the first set of hierarchical multiple linear regressions (Model 1), the four latent profiles were entered to examine their associations with overall patient safety, nurse‐reported quality of care, nursing work satisfaction and workplace satisfaction (Table 3).

TABLE 3.

Hierarchical multiple linear regression analysis of the identified latent profiles with patient care and nurse job outcomes.

Overall patient safety Nurse‐reported quality of care Nursing work satisfaction Workplace satisfaction
Model 1 Model2 Model 1 Model2 Model 1 Model2 Model 1 Model2
B (95% CI) B (95% CI) B (95% CI) B (95% CI) B (95% CI) B (95% CI) B (95% CI) B (95% CI)
Profiles (Ref. high well‐being)
Above‐average well‐being −0.13 (−0.28–0.03) −0.12 (−0.37–0.13) −0.17* (−0.34– −0.01) −0.12 (−0.39–0.15) −0.37 ** (−0.52– −0.22) −0.57 ** (−0.81– −0.33) −0.21 ** (−0.33–‐0.09) −0.54 ** (−0.73– −0.35)
Average well‐being −0.31 ** (−0.48– −0.15) −0.34 * (−0.59– −0.06) −0.41 ** (−0.59– −0.24) −0.37 ** (−0.63– −0.10) −0.75 ** (−0.91– −0.59) −0.95 ** (−1.19– −0.71) −0.51 ** (−0.63– −0.38) −0.83 ** (−1.02– −0.64)
Low well‐being −0.58 ** (−0.76–‐0.40) −0.51 ** (−0.77–‐0.26) −0.59 ** (−0.78–‐0.39) −0.49 ** (−0.77–‐0.22) −1.27 ** (−1.44– −1.10) −1.39 ** (−1.64– −1.14) −0.80 ** (−0.94– −0.66) −1.10 ** (−1.30– −0.90)
Organisational support (OS) 0.28 ** (0.24–0.33) 0.27 ** (0.12–0.41) 0.14 ** (0.09–0.19) 0.21 * (0.05–0.36) 0.23 ** (0.19–0.27) 0.08 (−0.06–0.22) 0.54 ** (0.51–0.58) 0.30 ** (0.18–0.41)
Profiles and OS interactions
Above‐average well‐being × OS −0.03 (−0.19–0.14) −0.04 (−0.22–0.14) 0.16 (−0.01–0.32) 0.26 ** (0.18–0.41)
Average well‐being × OS −0.07 (−0.20–0.12) −0.07 (−0.25–0.09) 0.10 (−0.05–0.26) 0.26 ** (0.14–0.38)
Low well‐being × OS 0.03 (−0.01–0.33) 0.03 (−0.15–0.21) 0.27 ** (0.11–0.44) 0.28 ** (0.16–0.41)
Open in a new tab

Note: Model 1 includes the four latent profiles (using ‘high well‐being’ as reference) and organisational support (OS) as predictors and Model 2 adds interaction terms (profiles × OS) to test for moderation effects. Models were adjusted for age, sex, marital status, education level in nursing, work status, shift type, night shift, Magnet status of the hospital, hospital location, unit type, frequency of caring for patients with COVID‐19, workload, likelihood of 30‐min breaks during work time. Bold indicates statistically significant results.

Abbreviation: CI, confidence interval.

*

p < 0.05.

**

p < 0.01.

Compared to the high well‐being profile, nurses in the average and low well‐being profiles reported significantly lower patient safety (B = −0.31, 95% CI = −0.48 to −0.15 for average well‐being; B = −0.58, 95% CI = −0.76 to −0.40 for low well‐being) and lower quality of care (B = −0.41, 95% CI = −0.59 to −0.24 for average well‐being; B = −0.59, 95% CI = −0.78 to −0.39 for low well‐being). Organisational support was positively associated with patient safety (B = 0.28, 95% CI = 0.24–0.33) and quality of care (B = 0.14, 95% CI = 0.09–0.19).

For nursing work satisfaction and workplace satisfaction, all three non–high‐well‐being profiles showed significantly lower scores compared to the high well‐being profile, with the low well‐being profile having the lowest outcomes for both variables (B = −1.27, 95% CI = −1.44 to −1.10 for nursing work satisfaction; B = −0.80, 95% CI = −0.94 to −0.66 for workplace satisfaction). Organisational support was positively associated with both work satisfaction (B = 0.23, 95% CI = 0.19–0.27) and workplace satisfaction (B = 0.54, 95% CI = 0.51–0.58).

When interaction terms were added (Model 2), organisational support did not significantly moderate the relationships between well‐being profiles and either overall patient safety or quality of care. However, significant interactions were observed for nursing work satisfaction in the low well‐being profile (B = 0.27, 95% CI = 0.11–0.44) and for workplace satisfaction across the above‐average (B = 0.26, 95% CI = 0.18–0.41), average (B = 0.26, 95% CI = 0.14–0.38) and low well‐being (B = 0.28, 95% CI = 0.16–0.41) profiles. These interactions indicate that organisational support moderated the negative associations between those profiles and both work and workplace satisfaction.

The results of the simple slope analyses (Figure 2) indicate that, among nurses in the low well‐being profile, the negative association with nurse work satisfaction was smaller at higher levels of organisational support (B = −1.11, 95% CI = −1.32 to −0.91) than at lower levels of organisational support (B = −1.66, 95% CI = −2.03 to −1.30). Similarly, the negative associations between the above‐average, average and low well‐being profiles and workplace satisfaction, relative to the high well‐being profile, were weaker at higher organisational support (above‐average well‐being: B = −0.28, 95% CI = −0.40 to −0.15; average well‐being: B = −0.57, 95% CI = −0.70 to −0.43; low well‐being: B = −0.81, 95% CI = −0.98 to −0.65) than at lower organisational support (above‐average well‐being: B = −0.80, 95% CI = −1.10 to −0.50; average well‐being: B = −1.09, 95% CI = −1.38 to −0.80; low well‐being: B = −1.38, 95% CI = −1.67 to −1.09).

FIGURE 2.

FIGURE 2

Open in a new tab

Simple slope analyses for interactions between latent well‐being profiles and organisational support. Plots depict predicted outcomes (e.g., nursing work satisfaction, workplace satisfaction) at low (–1 SD) and high (+1 SD) levels of OS in each latent profile.

6. Discussion

The primary aim of this study was to identify latent profiles of hospital nurses based on patterns of insomnia, fatigue, recovery, psychological distress and burnout using a person‐centered approach. In addition, the study examined variations in personal and work‐related characteristics across these profiles and assessed their associations with patient care and nurse job outcomes, exploring the moderating role of organisational support. The findings revealed the diversity of well‐being patterns among hospital nurses and provided insights for identifying subgroups at higher risk of impaired patient safety, reduced care quality and dissatisfaction with nursing work and the workplace. Greater organisational support showed the potential to mitigate the adverse effects of poor well‐being on nurse outcomes.

6.1. Identification of Distinct Well‐Being Profiles Among Hospital Nurses (Aim 1)

Four distinct profiles were identified. Interpreted based on the previously established cut‐off criteria (Winwood et al. 2006; Maslach and Jackson 1986; Morin et al. 2011; Kroenke et al. 2009), the high well‐being profile (6.6%) included nurses with no insomnia, low CF, low to moderate AF, high IR, no PD, low EE and DP, and high PA; the above‐average well‐being profile (24.0%) included nurses with mild insomnia, low to moderate CF and IR, moderate to high AF, mild PD, low EE and moderate DP and PA; the average well‐being profile (40.2%) included nurses with mild insomnia, moderate to high CF and AF, low to moderate IR, mild PD, and middle‐range EE, DP and PA; and the low well‐being profile (28.9%) included nurses with moderate insomnia, high CF and AF, low IR, moderate PD, high EE and DP, and low PA.

Interestingly, despite their overall well‐being variables being close to the sample mean, nurses in the average well‐being profile reported moderate to high levels of AF and CF and low to moderate levels of IR. The inherently demanding nature of the nursing profession, particularly for hospital nurses, may have contributed to these higher fatigue levels, as their roles often involve physically and mentally strenuous tasks with limited opportunities for rest (Chappel et al. 2017; Yuan et al. 2023). Lower IR in this profile may also indicate nurses' vulnerability to accumulating fatigue over time (Farag et al. 2024). Because data collection occurred in the middle of the COVID‐19 pandemic, nursing workloads, patient acuity and nurses' psychological demands were likely amplified (Turale et al. 2020), which may have further contributed to fatigue, insomnia and distress across the well‐being profiles.

6.2. Differences in Personal and Work‐Related Characteristics Across Profiles (Aim 2)

Our study found significant personal and work‐related characteristics variations across the identified profiles. Specifically, older nurses were more likely to belong to profiles with higher levels of well‐being. For older nurses, this trend might reflect a ‘healthy survival effect’, where those who successfully adapt to and manage various workplace stressors over time are more likely to remain in the profession. Another possibility is that, with experience, older nurses may develop effective strategies or skills to cope with stressors (Roeder et al. 2020), reducing its impact and decreasing the likelihood of belonging to the low well‐being profile. A recent study on hospital nurses supports this finding, which reported that older nurses were better able to detach from work‐related thoughts after their shifts (Cho et al. 2024), potentially enhancing their psychological well‐being. A similar observation was made in service and sales workers, where younger workers perceived higher emotional demands compared to their older colleagues (Han et al. 2018). These findings suggest the need to closely monitor and support the health and well‐being of younger and early‐career nurses, as they may be more vulnerable to workplace stressors and their associated negative effects.

Although male nurses accounted for only 10.3% of our study sample, their higher likelihood of belonging to the high well‐being profile may result from differences in how they respond to workplace stressors and adopt coping mechanisms. This finding aligns with a study on healthcare workers, which reported that women scored lower than men in emotional flourishing and recovery at work (Rink et al. 2022). However, to better understand how gender influences responses to workplace stressors and associated well‐being outcomes, further research with a sufficient sample size of male nurses is needed, ensuring that comparisons with female nurses are matched for personal and work‐related characteristics.

Nurses working in Magnet‐designated or Magnet‐journey hospitals had reduced odds of being in the low well‐being profile. Magnet hospitals, recognised by the ANCC, are known for their emphasis on nursing excellence and supportive work settings that enhance both nurse satisfaction and patient care quality (ANCC, n.d.). These institutions prioritise professional development, empower nursing staff and implement evidence‐based practices, which lead to better patient outcomes and lower rates of turnover, burnout and job dissatisfaction compared to non‐Magnet hospitals (Rodríguez‐García et al. 2020). Our study adds to existing evidence on the protective effects of supportive nursing work environments, such as those in Magnet‐designated hospitals, in mitigating factors that negatively impact nurses' well‐being.

Conversely, higher quantitative workloads and frequent care for patients with COVID‐19 were associated with an increased likelihood of nurses being in the low well‐being profile. High quantitative workloads are both physically and mentally taxing, requiring sustained energy and effort over long hours, which can lead to numerous negative health and well‐being outcomes for nurses (Xiao et al. 2022). These workloads' combined physical and mental demands often leave little room for recovery (Gifkins et al. 2020; Farag et al. 2024). Caring for patients with COVID‐19 can further compound this burden by introducing unique psychological challenges. Nurses frequently report heightened anxiety and fear of infection when caring for COVID‐19 patients (Fernandez et al. 2020), which amplifies the emotional toll of their work.

In addition to these stressors, the lack of adequate rest during shifts emerged as another significant factor in our study. Nurses who rarely took 30‐min breaks were more likely to belong to the low well‐being profile. Rest periods during work are essential for reducing fatigue, maintaining focus and preventing the accumulation of physical and emotional exhaustion (Sagherian et al. 2021; Lyubykh et al. 2022). Breaks can allow nurses to step away from their duties, engage in stress‐relieving activities such as mindfulness exercises, or connect with colleagues for social support, providing much‐needed recovery during demanding workdays.

6.3. Associations With Patient Care Quality and Nursing Job Outcomes (Aim 3)

In the current study, nurses in the average and low well‐being profiles reported significantly lower overall patient safety and quality of care compared to those in the high well‐being profile. This aligns with previous studies demonstrating the adverse impact of nurse fatigue, insomnia, burnout and psychological distress on patient safety and the quality of care (Cho and Steege 2021; Jun et al. 2021; Wang et al. 2023; Weaver et al. 2018). These well‐being challenges can compromise nurses' ability to provide safe and effective care, underscoring the importance of addressing nurse well‐being to maintain high standards of patient safety and care quality.

The significantly lower work satisfaction and workplace satisfaction observed in the above‐average, average and low well‐being profiles compared to the high well‐being profile further reinforces existing evidence that compromised well‐being negatively affects professional fulfilment and engagement (Brossoit et al. 2020; Cho and Steege 2021; Jun et al. 2021). Given the established links between nurse job satisfaction and turnover intention (Lu et al. 2019), the associations between well‐being profiles and these outcomes suggest an urgent need to identify, monitor and intervene for nurses at risk of belonging to negative well‐being profiles. Early identification and targeted interventions for these groups could prevent further declines in well‐being and improve nurse retention and job performance.

6.4. Moderating Role of Organisational Support (Aim 4)

Another interesting finding was the moderating role of organisational support on the negative associations of lower well‐being profiles with nurse work and workplace satisfaction. Organisational support, as defined in this study, includes recognising nurses' contributions, addressing their concerns and demonstrating genuine care for their well‐being (Eisenberger et al. 1986; Neves and Eisenberger 2014). Such support can strengthen nurses' sense of value and belonging (Eisenberger et al. 1986), which can be particularly important for nurses during periods of heightened stress, such as the COVID‐19 pandemic (Cho et al. 2023). Nurses who perceive their organisations as supportive are more likely to remain committed to their roles, even when confronted with high workloads or emotional challenges (Ekmekci et al. 2021; Naseer et al. 2021). Tangible efforts by organisations to acknowledge and address nurses' needs could serve as a psychological buffer, helping nurses maintain a positive attitude towards their work and workplace.

However, while organisational support was significantly associated with improved overall patient safety and quality of care, it did not significantly moderate the relationship between well‐being profiles and these outcomes. A possible explanation could be that the mechanisms through which poor well‐being profiles influence patient safety and care quality could not be addressed by organisational support alone. Nurses in the low well‐being profile may experience direct impairments in their capacity to deliver safe and effective care, limiting the buffering potential of organisational support. Consistent with our findings, previous studies have shown that physical and mental exhaustion can directly impair cognitive functioning, decision‐making and task performance (Harrison and Horne 2000; Lemonaki et al. 2021), which are critical for safe patient care.

Healthcare organisations should go beyond merely acknowledging and addressing nurses' needs to mitigate the negative impact of poor well‐being on patient care outcomes. Efforts should focus on proactively preventing and managing the root causes of poor well‐being, such as reducing excessive workloads and ensuring sufficient rest periods during work shifts. By addressing these systemic drivers of poor well‐being profiles and complementing the benefits of organisational support, healthcare organisations can more effectively enhance both nurse well‐being and patient safety and care quality.

6.5. Implications

Healthcare organisations should prioritise nurse well‐being as a fundamental element of patient safety, high‐quality care and nurse engagement and retention. Tailored interventions for nurses in the low well‐being profile are essential to address key stressors such as excessive workloads, caring for high‐acuity patients and insufficient rest breaks during shifts. Policies should strengthen work system resilience by ensuring adequate staffing, supporting recovery opportunities and fostering work–life balance.

In addition, systemic changes should be supported by strong organisational initiatives that go beyond simply acknowledging nurses' needs. Effective strategies can include recognising nurses' contributions, providing resources for stress management and proactively addressing workplace challenges that compromise well‐being. By creating supportive environments and tackling systemic factors contributing to poor well‐being, healthcare organisations can enable nurses to perform at their best, improve patient outcomes and enhance nurse satisfaction and retention, ultimately supporting a more sustainable healthcare workforce.

6.6. Limitations

This study has a few limitations to consider. First, all variables were measured through self‐reports from nurses, which may have introduced social desirability bias. Future studies should aim to replicate this study using objectively measured data. Second, because this study used a cross‐sectional design, it does not allow for establishing causal inferences regarding the relationships among variables. Future research could consider using a longitudinal design to better understand the directionality of these relationships.

Third, although we recruited a large number of nurses from various regions in the United States, the use of listserv‐based convenience sampling and nursing networking groups on Facebook may have introduced selection bias. Membership in these Facebook groups was not strictly limited to hospital nurses, and despite our mandatory screening question, some reliance on participants' good faith was necessary to verify professional status. As evidenced in our participant characteristics, the sample had a higher proportion of nurses working in critical care, step‐down or medical‐surgical units, many of whom were caring for COVID‐19 patients. This skewed representation may limit the generalizability of our findings to nurses in other specialties or less affected units. Future studies using more diversified sampling strategies (e.g., randomised sampling, multiple recruitment channels) could offer greater representativeness and broader applicability of results. Moreover, although both full‐time and part‐time nurses were included in our sample, we did not exclude or conduct separate analyses for part‐time nurses due to our focus on hospital nurses broadly and the relatively small proportion of part‐time participants. Nonetheless, differences in work schedules or staffing patterns may lead to distinct well‐being outcomes for part‐time nurses, suggesting that future studies could benefit from a more targeted analysis or exclusion criteria to better isolate these effects. In addition, comparisons with the 2022 National Sample Survey of Registered Nurses (NSSRN; Health Resources and Services Administration, n.d.) suggest that this study sample differs from the broader RN workforce in the United States. While national estimates for RNs across all practice settings report an average age of 45.8, 13% male nurses, and 21% holding a master's degree or higher, the average age of our sample was 41.5, with 10.3% male nurses and 10.8% holding a master's degree or higher. These discrepancies likely reflect our exclusive focus on actively practicing, hospital‐based staff nurses rather than RNs in more diverse care settings. Consequently, caution is warranted when generalising these findings to the wider nursing population.

Additionally, although we used Qualtrics security features (e.g., ‘Prevent multiple submissions’, ‘Prevent indexing’) to discourage automated or repeated entries, there remains a possibility that bots or imposters may have completed the survey. Future studies might incorporate additional verification methods to further reduce the risk of inauthentic responses.

Finally, despite recruiting nurses from settings that varied in Magnet status, location (rural, sub‐rural or urban) and hospital bed size, we did not collect information on certain organisational factors (e.g., for‐profit vs. not‐for‐profit status), which could further influence nurse well‐being and patient care. Data collection occurred during the COVID‐19 pandemic, a period of increased workloads and unique challenges for nurses and healthcare systems worldwide (Turale et al. 2020). As such, pandemic‐related stressors may have amplified certain aspects of nurse well‐being. Our results should therefore be interpreted with caution when applied to non‐pandemic contexts. Replication studies in post‐pandemic or more typical healthcare environments are warranted to confirm the stability and generalisability of these findings.

7. Conclusion

This study identified four distinct well‐being profiles among hospital nurses based on their levels of insomnia, fatigue, recovery, psychological distress and burnout: low well‐being, above‐average well‐being, average well‐being and high well‐being profile. These profiles differed significantly in personal and work‐related characteristics, including age, sex, work status, Magnet status of the hospital, workload, frequency of caring for patients with COVID‐19 and frequency of taking rest breaks during work hours. Nurses in the high well‐being profile were associated with higher overall patient safety, quality of care and nursing work and workplace satisfaction relative to those in other profiles. Organisational support was linked to moderating the negative associations between lower well‐being profiles and nursing work and workplace satisfaction. However, organisational support was not found to moderate the relationship between well‐being profiles and patient safety or quality of care, suggesting the need to address systemic factors such as excessive workloads and insufficient recovery opportunities. These findings underscored the importance of implementing tailored interventions to address nurses' risk factors for low well‐being and enhancing organisational support to create sustainable healthcare environments that improve nurse well‐being, promote high‐quality patient care and support nurse retention.

Disclosure

The statistics were checked prior to submission by an expert statistician, and state their name and email address: Chanmoo Park, chanmoo13@snu.ac.kr.

Ethics Statement

The institutional review board of the relevant university determined that this study was exempt (2021‐0494).

Consent

Voluntary informed consent was obtained online from the study participants.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1.

JAN-82-2216-s001.docx (18.3KB, docx)

Funding: This work was supported by the American Nurses Foundation. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of the American Nurses Foundation.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request. The data are not publicly available due to privacy or ethical restrictions.

References

  1. Alrabae, Y. M. A. , Aboshaiqah A. E., and Tumala R. B.. 2021. “The Association Between Self‐Reported Workload and Perceptions of Patient Safety Culture: A Study of Intensive Care Unit Nurses.” Journal of Clinical Nursing 30, no. 7–8: 1003–1017. 10.1111/jocn.15646. [DOI] [PubMed] [Google Scholar]
  2. American Nurses Credentialing Center . n.d. “ANCC Magnet Recognition Program.” Accessed December 30, 2024. https://www.nursingworld.org/organizational‐programs/magnet/.
  3. Bakker, A. B. , Demerouti E., and Euwema M. C.. 2005. “Job Resources Buffer the Impact of Job Demands on Burnout.” Journal of Occupational Health Psychology 10, no. 2: 170–180. 10.1037/1076-8998.10.2.170. [DOI] [PubMed] [Google Scholar]
  4. Berlin, K. S. , Williams N. A., and Parra G. R.. 2014. “An Introduction to Latent Variable Mixture Modeling (Part 1): Overview and Cross‐Sectional Latent Class and Latent Profile Analyses.” Journal of Pediatric Psychology 39, no. 2: 174–187. 10.1093/jpepsy/jst084. [DOI] [PubMed] [Google Scholar]
  5. Brossoit, R. M. , Crain T. L., Hammer L. B., Lee S., Bodner T. E., and Buxton O. M.. 2020. “Associations Among Patient Care Workers' Schedule Control, Sleep, Job Satisfaction, and Turnover Intentions.” Stress and Health 36, no. 4: 442–456. 10.1002/smi.2941. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Burnham, K. P. , and Anderson D. R.. 2004. “Multimodel Inference: Understanding AIC and BIC in Model Selection.” Sociological Methods & Research 33, no. 2: 261–304. 10.1177/0049124104268644. [DOI] [Google Scholar]
  7. Chappel, S. E. , Verswijveren S. J., Aisbett B., Considine J., and Ridgers N. D.. 2017. “Nurses' Occupational Physical Activity Levels: A Systematic Review.” International Journal of Nursing Studies 73: 52–62. 10.1016/j.ijnurstu.2017.05.006. [DOI] [PubMed] [Google Scholar]
  8. Chin, W. , Guo Y. L., Hung Y. J., Yang C. Y., and Shiao J. S. C.. 2015. “Short Sleep Duration Is Dose‐Dependently Related to Job Strain and Burnout in Nurses: A Cross Sectional Survey.” International Journal of Nursing Studies 52, no. 1: 297–306. 10.1016/j.ijnurstu.2014.09.003. [DOI] [PubMed] [Google Scholar]
  9. Cho, H. , Sagherian K., and Steege L. M.. 2023. “Hospital Staff Nurse Perceptions of Resources and Resource Needs During the COVID‐19 Pandemic.” Nursing Outlook 71, no. 3: 101984. 10.1016/j.outlook.2023.101984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cho, H. , and Steege L. M.. 2021. “Nurse Fatigue and Nurse, Patient Safety, and Organizational Outcomes: A Systematic Review.” Western Journal of Nursing Research 43, no. 12: 1157–1168. 10.1177/019394592199089. [DOI] [PubMed] [Google Scholar]
  11. Cho, H. , Steege L. M., and Pavek K. U.. 2024. “Psychological Detachment From Work During Nonwork Time as a Moderator and Mediator of the Relationship of Workload With Fatigue and Sleep in Hospital Nurses.” Sleep Health 10, no. 5: 558–566. 10.1016/j.sleh.2024.05.005. [DOI] [PubMed] [Google Scholar]
  12. Diener, E. 1984. “Subjective Well‐Being.” Psychological Bulletin 95, no. 3: 542–575. [PubMed] [Google Scholar]
  13. Diener, E. , Lucas R. E., and Oishi S.. 2002. “Subjective Well‐Being: The Science of Happiness and Life Satisfaction.” In Handbook of Positive Psychology, edited by Snyder C. R. and Lopez S. J., 463–473. Oxford University Press. [Google Scholar]
  14. Doğan, A. , Ertuğrul B., and Akin K.. 2024. “Examination of Workload Perception, Burnout, and Perceived Organizational Support in Emergency Healthcare Professionals: A Structural Equation Model.” Nursing & Health Sciences 26, no. 1: e13092. 10.1111/nhs.13092. [DOI] [PubMed] [Google Scholar]
  15. Eisenberger, R. , Huntington R., Hutchison S., and Sowa D.. 1986. “Perceived Organizational Support.” Journal of Applied Psychology 71, no. 3: 500–507. [Google Scholar]
  16. Ekmekci, O. T. , Xhako D., and Camgoz S. M.. 2021. “The Buffering Effect of Perceived Organizational Support on the Relationships Among Workload, Work–Family Interference, and Affective Commitment: A Study on Nurses.” Journal of Nursing Research 29, no. 2: e140. 10.1097/JNR.0000000000000419. [DOI] [PubMed] [Google Scholar]
  17. Farag, A. , Scott L., Perkhounkova Y., Abad P. J., and Hein M.. 2024. “Exploring Factors Influencing Nurses' Intershift Recovery and Its Effects on Progression From Acute Fatigue to Chronic Fatigue.” Journal of Nursing Regulation 15, no. 3: 23–32. 10.1016/S2155-8256(24)00072-3. [DOI] [Google Scholar]
  18. Farag, A. , Scott L. D., Perkhounkova Y., Saeidzadeh S., and Hein M.. 2022. “A Human Factors Approach to Evaluate Predictors of Acute Care Nurse Occupational Fatigue.” Applied Ergonomics 100: 103647. 10.1016/j.apergo.2021.103647. [DOI] [PubMed] [Google Scholar]
  19. Fernandez, R. , Lord H., Halcomb E., et al. 2020. “Implications for COVID‐19: A Systematic Review of Nurses' Experiences of Working in Acute Care Hospital Settings During a Respiratory Pandemic.” International Journal of Nursing Studies 111: 1–8. 10.1016/j.ijnurstu.2020.103637. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gifkins, J. , Johnston A., Loudoun R., and Troth A.. 2020. “Fatigue and Recovery in Shiftworking Nurses: A Scoping Literature Review.” International Journal of Nursing Studies 112: 1–15. 10.1016/j.ijnurstu.2020.103710. [DOI] [PubMed] [Google Scholar]
  21. Han, K. M. , Shin C., Yoon H. K., Ko Y. H., Kim Y. K., and Han C.. 2018. “Emotional Labor and Depressive Mood in Service and Sales Workers: Interactions With Gender and Job Autonomy.” Psychiatry Research 267: 490–498. 10.1016/j.psychres.2018.06.044. [DOI] [PubMed] [Google Scholar]
  22. Harrison, Y. , and Horne J. A.. 2000. “The Impact of Sleep Deprivation on Decision Making: A Review.” Journal of Experimental Psychology: Applied 6, no. 3: 236–249. 10.1037/1076-898X.6.3.236. [DOI] [PubMed] [Google Scholar]
  23. Health Resources and Services Administration . n.d. “NCHWA Nursing Workforce Dashboard.” Accessed March 31, 2025. https://data.hrsa.gov/topics/health‐workforce/nursing‐workforce‐dashboards.
  24. Jun, J. , Ojemeni M. M., Kalamani R., Tong J., and Crecelius M. L.. 2021. “Relationship Between Nurse Burnout, Patient and Organizational Outcomes: Systematic Review.” International Journal of Nursing Studies 119: 1–11. 10.1016/j.ijnurstu.2021.103933. [DOI] [PubMed] [Google Scholar]
  25. Kroenke, K. , Spitzer R. L., Williams J. B. W., and Löwe B.. 2009. “An Ultra‐Brief Screening Scale for Anxiety and Depression: The PHQ–4.” Psychosomatics 50, no. 6: 613–621. 10.1016/S0033-3182(09)70864-3. [DOI] [PubMed] [Google Scholar]
  26. Larsman, P. , Pousette A., and Törner M.. 2024. “The Impact of a Climate of Perceived Organizational Support on Nurses' Well‐Being and Healthcare‐Unit Performance: A Longitudinal Questionnaire Study.” Journal of Advanced Nursing 80, no. 12: 4921–4932. 10.1111/jan.16188. [DOI] [PubMed] [Google Scholar]
  27. Lemonaki, R. , Xanthopoulou D., Bardos A. N., Karademas E. C., and Simos P. G.. 2021. “Burnout and Job Performance: A Two‐Wave Study on the Mediating Role of Employee Cognitive Functioning.” European Journal of Work and Organizational Psychology 30, no. 5: 692–704. 10.1080/1359432X.2021.1892818. [DOI] [Google Scholar]
  28. Lindqvist, R. , Smeds Alenius L., Griffiths P., Runesdotter S., and Tishelman C.. 2015. “Structural Characteristics of Hospitals and Nurse‐Reported Care Quality, Work Environment, Burnout and Leaving Intentions.” Journal of Nursing Management 23, no. 2: 263–274. 10.1111/jonm.12123. [DOI] [PubMed] [Google Scholar]
  29. Loera, B. , Converso D., and Viotti S.. 2014. “Evaluating the Psychometric Properties of the Maslach Burnout Inventory‐Human Services Survey (MBI‐HSS) Among Italian Nurses.” PLoS One 9, no. 12: e114987. 10.1371/journal.pone.0114987. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Löwe, B. , Wahl I., Rose M., et al. 2010. “A 4‐Item Measure of Depression and Anxiety: Validation and Standardization of the Patient Health Questionnaire‐4 (PHQ‐4) in the General Population.” Journal of Affective Disorders 122, no. 1–2: 86–95. 10.1176/10.1016/j.jad.2009.06.019. [DOI] [PubMed] [Google Scholar]
  31. Lu, H. , Zhao Y., and While A.. 2019. “Job Satisfaction Among Hospital Nurses: A Literature Review.” International Journal of Nursing Studies 94: 21–31. 10.1016/j.ijnurstu.2019.01.011. [DOI] [PubMed] [Google Scholar]
  32. Lyubykh, Z. , Gulseren D., Premji Z., et al. 2022. “Role of Work Breaks in Well‐Being and Performance: A Systematic Review and Future Research Agenda.” Journal of Occupational Health Psychology 27, no. 5: 470–487. 10.1037/ocp0000337. [DOI] [PubMed] [Google Scholar]
  33. Maslach, C. , and Jackson S. E.. 1981. “The Measurement of Experienced Burnout.” Journal of Organizational Behavior 2, no. 2: 99–113. [Google Scholar]
  34. Maslach, C. , and Jackson S. E.. 1986. Maslach Burnout Inventory Manual. 2nd ed. Consulting Psychologists Press. [Google Scholar]
  35. McHugh, M. D. , and Stimpfel A. W.. 2012. “Nurse‐Reported Quality of Care: A Measure of Hospital Quality.” Research in Nursing & Health 35, no. 6: 566–575. 10.1002/nur.21503. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Morin, C. M. , Belleville G., Bélanger L., and Ivers H.. 2011. “The Insomnia Severity Index: Psychometric Indicators to Detect Insomnia Cases and Evaluate Treatment Response.” Sleep 34, no. 5: 601–608. 10.1093/sleep/34.5.601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Muthén, L. K. , and Muthén B. O.. 1998. –2020. Mplus User's Guide. 8th ed. Muthén & Muthén. [Google Scholar]
  38. Naseer, T. , Ahmad F., Bano A., Kausar N., Mehmood S., and Ali F.. 2021. “Impact of Role Conflict and Workload on Job Satisfaction: Moderating Role of Perceived Organizational Support.” Psychology and Education 58, no. 5: 4656–4677. [Google Scholar]
  39. Neves, P. , and Eisenberger R.. 2012. “Management Communication and Employee Performance: The Contribution of Perceived Organizational Support.” Human Performance 25, no. 5: 452–464. 10.1080/08959285.2012.721834. [DOI] [Google Scholar]
  40. Nylund, K. L. , Asparoutiov T., and Muthén B. O.. 2007. “Deciding on the Number of Classes in Latent Class Analysis and Growth Mixture Modeling: A Monte Carlo Simulation Study.” Structural Equation Modeling 14, no. 4: 535–569. 10.1080/10705510701575396. [DOI] [Google Scholar]
  41. Rink, L. C. , Silva S. G., Adair K. C., Oyesanya T. O., Humphreys J. C., and Sexton J. B.. 2022. “The Association Between Well‐Being Behaviors and Resilience in Health Care Workers.” Western Journal of Nursing Research 44, no. 8: 743–754. 10.1177/01939459211017515. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Rodríguez‐García, M. C. , Márquez‐Hernández V. V., Belmonte‐García T., Gutiérrez‐Puertas L., and Granados‐Gámez G.. 2020. “How Magnet Hospital Status Affects Nurses, Patients, and Organizations: A Systematic Review.” AJN: American Journal of Nursing 120, no. 7: 28–38. 10.1097/01.NAJ.0000681648.48249.16. [DOI] [PubMed] [Google Scholar]
  43. Roeder, A. C. , Garner J. T., and Carr K.. 2020. “Workplace Relationships, Stress, and Verbal Rumination in Organizations.” Southern Communication Journal 85, no. 2: 63–72. 10.1080/1041794X.2019.1697893. [DOI] [Google Scholar]
  44. Rutledge, D. N. , Douville S., and Winokur E. J.. 2022. “Chronic Fatigue Predicts Hospital Nurse Turnover Intentions.” JONA: The Journal of Nursing Administration 52, no. 4: 241–247. 10.1097/NNA.0000000000001139. [DOI] [PubMed] [Google Scholar]
  45. Sagherian, K. , Cho H., and Steege L. M.. 2024. “The Insomnia, Fatigue, and Psychological Well‐Being of Hospital Nurses 18 Months After the COVID‐19 Pandemic Began: A Cross‐Sectional Study.” Journal of Clinical Nursing 33, no. 1: 273–287. 10.1111/jocn.16451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sagherian, K. , McNeely C. A., and Steege L. M.. 2021. “Did Rest Breaks Help With Acute Fatigue Among Nursing Staff on 12‐h Shifts During the COVID‐19 Pandemic?” Journal of Advanced Nursing 77, no. 12: 4711–4721. 10.1111/jan.14944. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Sagherian, K. , Steege L. M., Cobb S. J., and Cho H.. 2023. “Insomnia, Fatigue and Psychosocial Well‐Being During COVID‐19 Pandemic: A Cross‐Sectional Survey of Hospital Nursing Staff in the United States.” Journal of Clinical Nursing 32, no. 15–16: 5382–5395. 10.1111/jocn.15566. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Sharp, L. , Rannus K., Olofsson A., Kelly D., Oldenmenger W. H., and EONS RECaN Group . 2019. “Patient Safety Culture Among European Cancer Nurses—An Exploratory, Cross‐Sectional Survey.” Journal of Advanced Nursing 75, no. 12: 3535–3543. 10.1111/jan.14177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Sheng, H. , Tian D., Sun L., Hou Y., and Liu X.. 2023. “Nurse Practice Environment, Perceived Organizational Support, General Well‐Being, Occupational Burnout and Turnover Intention: A Moderated Multi‐Mediation Model.” Nursing Open 10, no. 6: 3828–3839. 10.1002/nop2.1641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Smeds‐Alenius, L. , Tishelman C., Lindqvist R., Runesdotter S., and McHugh M. D.. 2016. “RN Assessments of Excellent Quality of Care and Patient Safety.” International Journal of Nursing Studies 61: 117–124. 10.1016/j.ijnurstu.2016.06.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Spector, P. E. , and Jex S. M.. 1998. “Development of Four Self‐Report Measures of Job Stressors and Strain.” Journal of Occupational Health Psychology 3, no. 4: 356–367. 10.1037/1076-8998.3.4.356. [DOI] [PubMed] [Google Scholar]
  52. Tang, D. D. , and Wen Z. L.. 2020. “Common Method Bias Test: Problems and Suggestions.” Psychological Science 1: 215–223. 10.16719/j.cnki.1671-6981.20200130. [DOI] [Google Scholar]
  53. Tein, J. Y. , Coxe S., and Cham H.. 2013. “Statistical Power to Detect the Correct Number of Classes in Latent Profile Analysis.” Structural Equation Modeling: A Multidisciplinary Journal 20, no. 4: 640–657. 10.1080/10705511.2013.824781. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Turale, S. , Meechamnan C., and Kunaviktikul W.. 2020. “Challenging Times: Ethics, Nursing, and the COVID‐19 Pandemic.” International Nursing Review 67, no. 2: 164–167. 10.1111/inr.12598. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Wang, M. , Wei Z., Wang Y., and Sun L.. 2023. “Mediating Role of Psychological Distress in Associations Between Medical Errors, Adverse Events, Suicidal Ideation, and Plans Among Operating Room Nurses.” BMJ Open 13, no. 6: e069576. 10.1136/bmjopen-2022-069576. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Weaver, M. D. , Vetter C., Rajaratnam S. M., et al. 2018. “Sleep Disorders, Depression, and Anxiety Associated With Adverse Safety Outcomes.” Journal of Sleep Research 27, no. 6: e12722. 10.1111/jsr.12722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Winwood, P. C. , Lushington K., and Winefield A. H.. 2006. “Further Development of the Occupational Fatigue Exhaustion Recovery (OFER) Scale.” Journal of Occupational and Environmental Medicine 48, no. 4: 381–389. 10.1097/01.jom.0000194164.14081.06. [DOI] [PubMed] [Google Scholar]
  58. Winwood, P. C. , Winefield A. H., Dawson D., and Lushington K.. 2005. “Development of a Scale to Measure Work‐Related Fatigue and Recovery: The OFER Scale.” Journal of Occupational and Environmental Medicine 47, no. 6: 594–606. 10.1097/01.jom.0000161740.71049.c4. [DOI] [PubMed] [Google Scholar]
  59. World Health Organization . 2020. State of the World's Nursing 2020: Investing in Education, Jobs and Leadership. World Health Organization. Accessed March 24, 2025. https://www.who.int/publications/i/item/9789240003279. [Google Scholar]
  60. Wurpts, I. C. , and Geiser C.. 2014. “Is Adding More Indicators to a Latent Class Analysis Beneficial or Detrimental? Results of a Monte‐Carlo Study.” Frontiers in Psychology 5: 920. 10.3389/fpsyg.2014.00920. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Xiao, Q. , Cooke F. L., and Chen L.. 2022. “Nurses' Well‐Being and Implications for Human Resource Management: A Systematic Review.” International Journal of Management Reviews 24, no. 4: 599–624. 10.1111/ijmr.12295. [DOI] [Google Scholar]
  62. Yasin, Y. M. , Kerr M. S., Wong C. A., and Bélanger C. H.. 2020. “Factors Affecting Nurses' Job Satisfaction in Rural and Urban Settings: A PRISMA Review.” Journal of Advanced Nursing 76, no. 4: 963–979. 10.1111/jan.14293. [DOI] [PubMed] [Google Scholar]
  63. Yuan, Z. , Wang J., Feng F., et al. 2023. “The Levels and Related Factors of Mental Workload Among Nurses: A Systematic Review.” International Journal of Nursing Practice 29, no. 5: e13148. 10.1111/ijn.13148. [DOI] [PubMed] [Google Scholar]
  64. Zhang, H. , Hoe V. C. W., and Wong L. P.. 2024. “The Association Between Burnout, Perceived Organizational Support, and Perceived Professional Benefits Among Nurses in China.” Heliyon 10, no. 20: e39371. 10.1016/j.heliyon.2024.e39371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Zou, X. , Liu S., Li J., et al. 2021. “Factors Associated With Healthcare Workers' Insomnia Symptoms and Fatigue in the Fight Against COVID‐19, and the Role of Organizational Support.” Frontiers in Psychiatry 12: 652717. 10.3389/fpsyt.2021.652717. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1.

JAN-82-2216-s001.docx (18.3KB, docx)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request. The data are not publicly available due to privacy or ethical restrictions.

Articles from Journal of Advanced Nursing are provided here courtesy of Wiley

RESOURCES