Article Text
PDF
Abstract
Objective The present study employed latent profile analysis (LPA) to identify three distinct profiles of subjective well-being (SWB) among Chinese nurses. It further examined the factors influencing these profiles and aimed to provide a foundation for targeted interventions to enhance nurses’SWB.
Design A cross-sectional study was conducted between November 2023 and March 2024.
Setting Data were collected from three Class III Grade A hospitals in China.
Participants A total of 2272 nurses were recruited for this study.
Outcome measures Data collection used a demographic questionnaire, the SWB Scale, the Nurse Job Satisfaction Scale and the Perceived Social Support Scale. LPA identified distinct SWB characteristics, and influencing factors were analysed using χ2 tests and multivariable logistic regression analysis.
Results Nurses’ SWB was classified into three profiles: (1) high health concern–low well-being (27.3%), (2) moderate health concern–moderate well-being (41.1%) and (3) low health concern–high well-being (31.6%). Multivariable regression analysis revealed significant associations of gender, age, years of experience, professional title, position, self-perceived health, social support and job satisfaction with these profiles (p<0.05).
Conclusion Given the heterogeneity of nurses’ SWB identified through LPA, healthcare institutions may design evidence-based interventions tailored to specific profiles (eg, high health concern–low well-being groups) and key predictors (eg, job satisfaction and social support) to promote sustainable well-being and reduce burnout risks.
- Job Satisfaction
- Nurses
- Social Support
Data availability statement
Data may be obtained from a third party and are not publicly available.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
The study included 2272 nurses from three Class III Grade A hospitals in two provinces in China.
We accounted for individual heterogeneity using latent profile analysis.
This cross-sectional design precluded establishing causal inferences between social support, subjective well-being and job satisfaction.
Data were collected via self-administered questionnaires, which may introduce recall and reporting biases.
Introduction
The global shortage of nurses has become a critical issue, profoundly affecting healthcare systems worldwide.1 In the UK, the National Health Service currently reports over 41 000 unfilled nursing positions,2 with more professionals leaving the field than entering it. Similarly, in China, there are just 3.71 registered nurses per 1000 people,3 which falls well below the WHO’s recommended minimum of 4.45 nurses per 1000 people.4 Nursing staff in China face chronic understaffing issues, which necessitates heavier workloads. This leads to reduced quality of care, poorer patient outcomes, lower job satisfaction and higher staff turnover rates. Therefore, understanding the subjective well-being (SWB) of nurses has become a pressing issue that needs attention.
As the core notion of positive psychology, SWB means that people make comprehensive assessments of their whole life.5 This multidimensional construct is theoretically divided into two distinct domains: (1) subjective components, involving self-reported cognitive appraisals (eg, life satisfaction and perceived quality of life) and emotional responses to life events such as happiness,6 and (2) contextual components, comprising objective indicators. The latter comprises three categories: (a) socioeconomic determinants (eg, income level and employment status), (b) relational capital (eg, social network density and community engagement) and (c) environmental determinants (eg, neighbourhood safety, air quality index and accessibility to green spaces). In this study, we used the three-dimensional model of Subjective SWB proposed by Diener—defined as the sum of an individual’s affective experience (positive/negative emotions) and cognitive appraisal (life satisfaction) of their quality of life—to differentiate it from the concept of multidimensional well-being, which includes objective indicators.7 Over the past few years, SWB has gained prominence in the international community, whose measurement is increasingly considered to better symbolise national prosperity compared with fiscal wealth.8 Working life is a significant part of the life of many people, which can promote SWB by providing income, a sense of identity, supportive relationships and the meaning of life.
Social support refers to the material and moral support provided by individuals or organisations, including families, family members, friends, colleagues, partners and associations.9 Social support serves as a critical external resource that substantially contributes to the enhancement of SWB. A qualitative study of nurses found that nurses’ perceived well-being was related to support from family and friends and a sense of belonging to a team.10 The SWB of nurses increases when they feel a strong sense of social support and organisational identity.11 Traditionally, job satisfaction is defined as the perception that an individual has about his or her job.12 However, the extent to which the actual work environment meets individual expectations of the ideal work situation also actively shapes job satisfaction.12 Nurses’ job satisfaction, along with perceived social support, works together to influence nurses’ SWB.
Most research on nurses’ well-being has relied on variable-centred approaches, focusing on isolated dimensions and their relationships with other variables. This approach, however, often overlooks the complexity of SWB characteristics. Latent profile analysis (LPA)13 offers a more comprehensive, individual-centred method by categorising individuals into distinct latent profiles, enabling the exploration of profile characteristics and their associations with other factors. While LPA has been applied to SWB in various professions, studies specifically focusing on nurses remain scarce. Nursing, as a caring profession, entails heightened stress from both social expectations and self-imposed demands. Consequently, nurses may exhibit distinct SWB characteristics compared with other fields. Research suggests that nurses generally have lower levels of well-being than other occupational groups, which is associated with high levels of emotional labour, prolonged exposure to traumatic events (eg, patient deaths) and high-intensity work environments.14 However, some studies have also pointed to a unique ‘high-stress-high-meaning’ well-being profile among nurses, in which a sense of professional purpose may buffer negative emotions despite high levels of stress,15 rendering them a particularly pertinent population for LPA research. This study aims to address the following research questions: (1) Do nurses with varying demographic and occupational characteristics exhibit significant differences in SWB? (2) Are there correlations between SWB, job satisfaction and perceived social support? (3) Do different SWB profiles display distinct levels of job satisfaction and perceived social support? By investigating these questions, this study seeks to inform targeted interventions to enhance nurses’ SWB, potentially addressing some aspects of the global nursing shortage and contributing to a more resilient healthcare system.
Methods
Study design and sample
Using a convenience sampling approach, nurse survey data were collected from three Class III Grade A hospitals in southeastern China between November 2023 and March 2024. To enhance the geographical representativeness of the sample, participating hospitals were purposively selected based on their geographical distribution, including both Guangzhou (the provincial capital) and other provincial regions. Among the three institutions that agreed to participate, data collection encompassed diverse clinical settings: general care units (comprising medical, surgical and medical-surgical units) and specialised care units (including intensive care units, perioperative departments and emergency departments). Registered nurses actively performing clinical duties were recruited for participation, provided they met the following inclusion criteria: (1) employment in a comprehensive hospital, (2) registration as a practising nurse, (3) current engagement in clinical duties and (4) willingness to participate in the study. Nurses who were interns, enrolled in academic programmes at other institutions, or involved in other similar studies were excluded to maintain sample homogeneity.
Patient and public involvement
Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Data collection
This study used a convenience sampling approach, distributing electronic questionnaires via the online platform ‘Wen Juanxing (https://www.wjx.cn/) from November 2023 to March 2024. Data were collected anonymously through targeted snowball sampling, allowing participants who met the inclusion criteria to access the survey link on their mobile devices or computers. On accessing the link, participants were presented with an electronic informed consent form; those who agreed proceeded to the questionnaire interface. To ensure data quality, surveys completed in less than 200 s were excluded from the analysis.
Instruments
Demographic questionnaire
The questionnaire was specifically designed for this study, capturing data on participants’ gender, age, education level, length of service, professional title, position, hospital grade and self-perceived health status.
General Well-Being Schedule
The Chinese version of the General Well-Being Schedule (GWBS), translated from the original GWB Schedule16 to measure an individual’s subjective sense of well-being,17 is widely used to assess the SWB of nurses.18 The instrument consists of 18 items in six dimensions, including satisfaction and interest in life, health concerns, energy, mood (depressed or joyful), control of emotions and behaviours and the degree of relaxation and tension. GWBS contains 14 items that are rated on a 6-point scale and four items rated on a 10-point scale. The overall GWB score is the totality of six dimensions with higher scores representing better overall well-being. In this study, the scale demonstrated good internal consistency, with a Cronbach’s alpha coefficient of 0.867.
Nurses’ Job Satisfaction Scale
Nurses’ Job Satisfaction Scale (NJSS) developed by Warr et al is a 5-point Likert type scale containing 15 items to measure job satisfaction and is translated by Lu et al.19 NJSS contains two subscales: interpersonal relationship at work and working status. The overall NJSS score is the sum of the two subscales, with higher scores reflecting higher job satisfaction among nurses. In this study, the scale demonstrated excellent internal consistency, with a Cronbach’s alpha coefficient of 0.964.
Perceived Social Support Scale
This instrument was designed by Zimet.20 This scale comprises three dimensions: support from friends, support from family and support from others. It is measured using 7-point scores (7=‘extreme consent’ to 1=‘extreme disgust’). The overall Perceived Social Support Scale score is the sum of the three dimensions; the higher the total score, the higher the level of social support. In this study, the scale demonstrated excellent internal consistency, with a Cronbach’s alpha coefficient of 0.976.
Statistical analysis
Advantages of the person-centred approach
In our attempt to look for the profiles of happiness, person-centred analyses would be a category of statistical techniques that best suit the purpose.21 Statistical analyses, including cluster analysis, latent class analysis and latent profile analyses, are typical techniques that fall into this category. Individuals are regarded as the functioning whole instead of the sum of the parts in a person-centred analysis.22 and it serves several advantages in its usage. First, it enables us to identify the profiles among individuals, thus providing distinctive insight into the heterogeneity of the target population. Second, it simplifies the otherwise complex higher-order interactions among variables in the variable-centred analyses into a brief and simple representation.23 Lastly, tailored-made interventions for subpopulations become possible to suit their needs better.
Latent profile analysis
We used LPA in examining the number of unobserved classes (ie, the categorical latent profiles of happiness), describing the characteristics of the classes and estimating the probabilities of class memberships for each individual. Data for each item in the six dimensions were entered into the LPA, and in this study, one to six potential profile models were explored sequentially from the initial model (one profile) until the most appropriate model was determined with a log-likelihood test.
The data analysis protocol was executed through a two-stage analytical framework. Primary statistical analyses, including descriptive statistics, logistic regression and one-way analysis of variance (ANOVA) with post hoc least significant difference tests, were performed using SPSS 26.0 (IBM Corp.). LPA was conducted with Mplus V.8.3. Model fit indices for LPA included the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC) and Adjusted Bayesian Information Criterion (aBIC), with lower values indicating better model fit. Classification accuracy was assessed through entropy, with values ranging from 0 to 1, where higher values reflect more accurate classifications.24 Model significance was tested using the Lo-Mendell-Rubin (LMR) likelihood ratio test and bootstrap likelihood ratio test, with p<0.05 indicating a statistically better fit than the previous model.25 The optimal model was selected based on a comprehensive evaluation of these fit indices and significance tests. Previous studies have confirmed that the minimum sample size of LPA is 500.26 In this study, 2272 eligible nurses participated, well surpassing this threshold and reinforcing the study’s robustness and statistical power.
Ethical procedures
Participation in the study was voluntary and anonymous. The study adhered to the ethical standards of the Declaration of Helsinki (64th WMA General Assembly, Fortaleza, Brazil, October 2013) and complied with relevant Polish regulations. Ethical approval was granted by the Ethics Review Committee of the First Hospital of Guangzhou Medical University (Ethics No. ES-2024-K119-01) prior to study commencement.
Results
Descriptive statistics and correlation analysis of major variables
A total of 2490 electronic questionnaires were distributed, with 2272 completed and returned, yielding a response rate of 91.2%. Of the respondents, 148 were male nurses and 2124 were female nurses, aged 20 to 55 years (mean age=32.89±8.35 years). Specifically, 1135 participants were aged 20–30 years, 704 were 31–40 years, 343 were 41–50 years and 90 were 51 years or older. Self-perceived health status was reported as very healthy by 427 nurses, relatively healthy by 940, fair by 737, not so good by 154 and poor by 14. Additional demographic information is presented in table 1. Table 2 presents the means, SD and correlations for the study variables. Notably, associations between control over behaviour and emotions, friend-based support and other forms of support did not reach statistical significance.
General information of survey respondents and results of univariate analysis of three potential profiles of overall well-being (n=2272, (percentage, %))
Means, SD and correlations of variables
Latent profiles of subjective well-being
To determine the optimal number of latent profiles, the single-profile baseline model served as an initial reference. Table 3 presents model fit metrics for competing latent profile models. As the number of profiles increased, the AIC, BIC and aBIC values decreased monotonically. All six models demonstrated entropy values>0.80; however, profiles 5 and 6 contained subgroups with probabilities<5%. Both the LMR and LMR-adjusted likelihood ratio test indices for the three- and four-profile models reached statistical significance (p<0.05), indicating improved model fit with additional profiles. A scree plot of BIC values revealed a distinct inflection point between the three- and four-profile models, suggesting diminishing returns in fit improvement. Based on parsimony and interpretability, the three-profile model was selected as optimal. Each profile demonstrated mean posterior classification probabilities>90%, indicating high reliability and precision.
Fit statistics for profile structure
Figure 1 shows the scores of the three-profile model on 18 items. C1 is the largest group, accounting for 41% (n=934) of all nurses. Nurses in this group demonstrate the lowest level of well-being. Notably, while C1 scored highest on item 16 (subdimensions representing health concerns), they exhibited lower scores on other items. This pattern indicates a combination of elevated health concerns and diminished perceived well-being, justifying the label ‘high health concern–low well-being group’. Group C2 displays moderate scores across all items, earning the designation ‘moderate health concern–moderate well-being group’. In contrast, group C3 shows low scores on health concerns and high scores on all other items. This profile reflects minimal health-related worries coupled with robust well-being, thus being labelled ‘the low health concern–high well-being group’.
Latent profiles for well-being.
Predictor of the latent profile membership
To examine the demographic and other variables affecting SWB in nurses across different profiles, a multinomial logistic regression analysis was conducted using class 1 as the reference group. Predictors included gender, age, length of service, title, position, self-perceived health status, job satisfaction scores and social support scores. Table 1 summarises the influencing factors of latent profile membership. The results demonstrated that gender, age, length of service, title, position, self-perceived health status, job satisfaction scores and social support scores significantly predicted profile membership. Specifically, (1) when comparing C1 (high health concern–low well-being) with C2 (moderate health concern–moderate well-being), nurses holding lower positions and reporting higher social support levels showed greater likelihood of belonging to the moderate well-being group. (2) In the comparison between C1 and C3 (low health concern–high well-being), nurses with longer service duration, lower positions, higher social support and elevated job satisfaction were more frequently classified. into the high well-being group. (3) Contrasting C2 with C3, nurses exhibiting lower job satisfaction and reduced social support tended to remain in the moderate well-being group. Collectively, these findings indicate that social support, job satisfaction and position serve as critical determinants of nurses’ well-being profiles, increasing the probability of transitioning toward the moderate well-being group (see table 4).
Results on well-being variables and latent profiles
Discussion
The present study employed LPA to delineate heterogeneous patterns of SWB among clinical nurses and examined demographic correlates of profile membership. This person-centred approach reveals previously unrecognised SWB diversity across distinct subgroups, challenging the conventional homogeneity assumption prevalent in prior research. By identifying unique intervention entry points for each profile, our findings provide a nuanced framework for developing tailored strategies to enhance nurses’ psychological well-being in high-stress clinical environments.
LPA identified three distinct SWB profiles with optimal model fit: (1) high health concern–low well-being group (31.6%): Characterised by significantly compromised SWB scores, particularly in energy levels, life engagement and mindfulness, this subgroup exhibited the lowest health self-evaluations. The observed deficits may stem from inadequate familial and social support systems. (2) Medium health concern–moderate well-being group (41%): Nurses in this profile demonstrated balanced yet suboptimal functioning across all SWB dimensions. (3) Low health concern–high well-being group (27.3%): Marked by robust psychological resilience, elevated vitality and proactive life engagement, this group likely benefits from strong interpersonal support networks—a critical external resource for mitigating occupational stressors.27 Notably, 72.6% of participants fell into the first two profiles, indicating that a substantial majority of Chinese clinical nurses experience subpar SWB, underscoring an urgent need for systemic interventions. To address this disparity, we propose a dual-pronged strategy. First, organisational-level reforms implement transparent decision-making protocols for nurse-related policies, including equitable compensation structures, merit-based promotion systems and trauma-informed support frameworks. Second, workplace ecology optimisation redesign shift scheduling algorithms to align with circadian rhythms and establish peer-mentored resilience training programmes. Such measures could enhance nurses’ psychological ownership of their well-being while reducing attrition rates in high-intensity clinical settings.
Nurses with 11 to 20 years of service are more likely to be in the high health concern–low well-being group. This may be attributed to their critical roles in clinical settings and family responsibilities, which contribute to heightened work pressure. Middle-aged nurses were also found to report relatively low levels of SWB, likely due to the demands of balancing family obligations with career growth. Additionally, their comparatively limited research skills and job performance relative to senior nurses may hinder career advancement, further impacting their well-being.28 Studies have shown that excessive work hours can lead to nurse fatigue and burnout, which can decrease work engagement and concentration. Long work hours can adversely affect nurses’ physical and mental health and increase the likelihood of emotional exhaustion.29 To address these issues, nurse managers play a pivotal role in fostering job satisfaction by addressing both systemic organisational factors and individualised needs. Drawing on the Job Demands-Resources model30 and empirical evidence, the following evidence-based strategies are recommended: implement predictive staffing algorithms to align nurse-patient ratios with real-time acuity levels, reducing unsustainable workloads, establish shared governance councils where nurses co-design unit policies and quality improvement initiatives, provide clinical decision-making authority in patient care plans for senior nurses, create dual-track career ladders allowing nurses to advance as clinical experts or administrative leaders, offer compensated time (eg, 4 hours/week) for pursuing certifications or advanced degrees and implement real-time feedback systems (eg, monthly ‘listening rounds’) to address concerns proactively.
This study contributes theoretically in terms of being the first person-oriented research on the China nurse population. Differences in SWB among clinical nurses were explored. The profile structure revealed can serve as a reference for future studies, and we can integrate the future data to revise the profile structure.
The adoption of person-centred analyses using LPA or similar statistical strategies is increasingly popular in the nursing field in the recent decade. Studied outcomes included commitment,31 work demands and32 moral sensitivity.33 Many stopped at mere classification without further predicting distal outcomes. We believe that person-centred analyses would be important to the nursing field for the benefits of devising tailored interventions as mentioned above and would advocate this approach in future nursing studies.
Limitation
This study has several limitations. First, as with most cross-sectional studies, establishing clear causal relationships between variables was challenging. Additionally, the use of self-reported questionnaires, with key variables measured simultaneously, may introduce common method biases. Future research should consider alternative methodologies to mitigate these limitations. Second, further qualitative research is warranted to provide a more nuanced understanding of individual experiences, complementing the quantitative findings. Third, this study’s data were gathered exclusively from nurses in university hospitals, which may limit the generalisability to the broader nursing community, including those in smaller hospitals. Expanding data collection to include nurses from various hospital types is recommended for future studies. Lastly, this study focused solely on a Chinese sample; replication in diverse international contexts is necessary to enhance the generalisability of these findings globally.
Conclusion
This research innovatively identified the subgroup characteristics and predictors of nurses’ SWB through LPA. We found three distinct profiles of SWB among nurses: the high health concern–low well-being group, the moderate health concern–moderate well-being group and the low health concern–high well-being group. Furthermore, we revealed that potential predictors of profile membership include gender, age, length of service, title, position, self-perceived health status, job satisfaction and social support. This study suggests that nursing administrators can design targeted interventions and training programmes based on the heterogeneity of nurses’ SWB in future practice. For example, administrators could provide tailored incentives (eg, peer support and narrative medicine) to nurses already engaged in their work, aligned with the unique characteristics and needs of each subgroup, thereby enhancing their SWB. In summary, improving nurses’ SWB is crucial to meeting demands for quality care, reducing nursing turnover and alleviating workforce shortages.
Data availability statement
Data may be obtained from a third party and are not publicly available.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants. Approval was obtained from the Ethics Review Committee of the First Hospital of Guangzhou Medical University (Ethics No. ES-2024-K119-01) prior to the investigation. Participants gave informed consent to participate in the study before taking part.
References
Footnotes
Contributors P-PS, SMC and XW made significant contributions to conception and design, acquisition of data or analysis and interpretation of data and final approval of the version to be published. Each author is expected to participate fully in the work by assuming public responsibility for appropriate portions of the content. FC, HF, XF, LW, HH and JO contributed to the drafting of the manuscript or critical revision of important intellectual content. XW agreed to be responsible for all aspects of the work to ensure that issues related to the accuracy or completeness of any portion of the work are appropriately addressed. investigated and resolved. Guarantor: XW.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.