The Association between Shift Work and Mental Health Outcomes among Korean Nurses

Abstract

Background: Shift work is a prevalent occupational factor for nurses, yet its impact on mental health remains under-explored in South Korea. We investigated the associations between shift work and generalized anxiety disorder (GAD), depressive symptoms, and fatigue among Korean nurses. Methods: Data from 20,613 participants in the Korean Nurses' Health Study Module 1 (2013-2015) were analyzed. Mental health outcomes were compared between shift and non-shift workers using chi-square tests. Multivariable logistic regression, adjusted for sociodemographic and occupational covariates, was performed to estimate odds ratios. Subgroup analyses were stratified by the presence of sleep problems. Results: Shift workers showed a higher prevalence of GAD, depressive symptoms, and fatigue compared with non-shift workers. In multivariable models, shift work was significantly associated with increased odds of GAD, depressive symptoms, and fatigue. Stratified analyses revealed that, even without sleep problems, shift work was associated with a 1.54-fold higher risk of depressive symptoms. Conclusion: Shift work is a significant risk factor for adverse mental health in nurses. Workplace interventions are essential to mitigate these effects.

연구배경: 교대근무는 간호사의 대표적인 직업적 요인으로 알려져 있으나, 국내 간호사를 대상으로 교대근무가 정신건강에 미치는 영향에 대해서는 충분히 규명되지 않았다. 본 연구는 국내 간호사를 대상으로 교대근무와 범불안장애, 우울증상 및 피로 간의 관련성을 확인하고자 하였다. 방법: 한국간호사건강연구(Korean Nurses' Health Study) Module 1(2013-2015)에 참여한 20,613명을 대상으로 분석하였다. 교대근무 여부에 따른 정신건강 결과의 차이는 카이제곱 검정을 이용하여 비교하였으며, 사회인구학적 및 직업 관련 변수를 보정한 다변량 로지스틱 회귀분석을 수행하여 교차비를 산출하였다. 또한 수면문제 유무에 따른 층화분석을 실시하였다. 결과: 교대근무 간호사는 비교대근무 간호사에 비해 범불안장애, 우울증상 및 피로의 유병률이 높았다. 다변량 분석결과, 교대근무는 범불안장애, 우울증상 및 피로 위험 증가와 유의한 관련성을 보였다. 층화분석 결과, 수면문제가 없는 경우에도 교대근무 간호사는 우울 증상의 위험이 1.54배 높게 나타났다. 결론: 교대근무는 간호사의 정신건강 악화와 관련된 중요한 위험요인으로 확인되었다. 이러한 부정적 영향을 완화하기 위한 직장 차원의 중재가 필요하다.

INTRODUCTION

  Nurses frequently engage in shift work and extended working hours [1]. In previous studies, shift work has been described as employment performed outside conventional daytime hours, typically between 7 or 8 AM and 5 or 6 PM, and may include fixed or rotating schedules [2]. It can cause circadian rhythm disruption by misaligning the internal biological clock with the external sleep–wake schedule, with sleep loss and impaired alertness occurring as secondary consequences [2].
  Mental health, an integral component of overall health and well-being, is influenced by demographic, biological, psychosocial, and genetic factors, as well as cir- cadian rhythms [3]. Accumulating evidence from previous studies suggests that long-term shift work is associated with various adverse health outcomes, including cardiovascular diseases and mental disorders, highlighting the growing need for health protection measures for shift workers [4]. The influence of irregular shift work hours on nurses’ psy- chological health is critical for healthcare systems. Previous studies have demonstrated that fatigue, anxiety, depres- sion, stress, and sleep problems are prevalent among nurses working non-traditional hours [5].
  Some recent studies have focused on the relationship between nurse’s shift work and risk of mental health prob- lems. A meta-analysis of longitudinal studies found that shift workers had a significantly higher risk of poor mental health, particularly depressive symptoms, compared with non-shift workers [6]. Moreover, a recent review reported that shift work is consistently associated with poorer mental health outcomes, including depressed mood and anxiety, particularly among workers experiencing sleep problems such as short sleep duration, excessive daytime sleepiness, and insomnia [7].
  The working conditions of nurses in Korea—such as high workload, staffing shortages—differ from those in many other countries, potentially modifying the impact of shift work on mental health. As a result, findings from inter- national studies may not be fully generalizable to Korean nurses. Moreover, there is still a lack of research focusing on Korean nurses, underscoring the need for studies that spe- cifically examine this association within the Korean health- care context. Therefore, this study aimed to investigate the association between shift work and mental health problems, including generalized anxiety disorder (GAD), depression, and fatigue, among nurses, with particular attention to sleep problems.
  Given the growing policy interest in improving nurses’ working conditions and shift systems, this study also sought to provide foundational evidence for developing institu- tional strategies to mitigate the adverse mental health effects of shift work. In particular, the findings may contribute to future discussions on shift scheduling policies that consider not only the presence of shift work itself, but also the quan- titative and qualitative characteristics of shift patterns.

METHODS

1. Data source and study population

  This population-based study utilized the database of the Korean Nurses’ Health Study Module 1, conducted from March 15, 2013, to December 15, 2015, consisted of 20,613 female registered nurses of childbearing age (20–45 years) residing in Korea [8]. The Korea Nurses’ Health Study data- base contains comprehensive information collected primar- ily through web-based surveys, including sociodemographic characteristics, medical history, medication use, lifestyle factors, and occupational exposures and work conditions [8]. The database employs de-identified identifiers to en- sure individuals’ anonymity and to allow linkage between datasets. This study employed a cross-sectional design to investigate the associations between shift work and various mental health outcomes among nurses. The study protocol was reviewed and approved by the Institutional Review Board (IRB) of Eulji University (IRB number: EUIR2024- 051). The requirement for informed consent was waived, as the database did not include identifiable information.

2. Measures of generalized anxiety disorder, depressive symptoms, and fatigue

  This study examined three outcome variables. First, State- Trait Anxiety Inventory-6 was used to assess the severity of self-reported anxiety [9]. Item scores were summed and rescaled to the full 20-item range (20–80), with a cut-off of ≥12 (raw), equivalent to ≥40 (rescaled), indicating clinically significant anxiety [9]. Second, depressive symptoms was assessed using the 9-item Patient Health Questionnaire, with total scores ranging from 0 to 27 and a cut-off of ≥10 indicating clinically significant depressive symptoms [10]. Third, fatigue was evaluated using the 11-item Chalder Fa- tigue Scale, with responses rated on a 4-point Likert scale (0–3) and summed to yield a total score of 0–33; a cut-off of ≥15 was applied to define the presence of fatigue. Higher scores indicated greater symptom severity [11].

3. Independent variable (shift work status)

  The main independent variable was shift work status, which was assessed using a self-reported questionnaire. Participants were asked the following question: “During the past year, have you engaged in shift work?” To ensure com- parability between groups, responses were dichotomized into ‘Yes’ (exposed group: nurses who performed rotating or fixed shift work) and ‘No’ (unexposed group: fixed day workers).

4. Covariates

  Several potential confounding factors were included in this study: age (20–29 years, 30–39 years, ≥40 years); income level (quartiles); hospital size (measured by the number of beds, <30 beds, 30–99 beds, 100–299 beds, ≥300 beds); marital status (married or unmarried); number of children (0, 1, 2, ≥3); job position (staff nurse or nurse man- ager); sleep problem (no or yes); employment type (perma- nent type or contract type); and medical history (no or yes).

5. Statistical analyses

  Descriptive analyses were performed to summarize the distributions of baseline characteristics according to shift work status. Differences in categorical variables were ex- amined using chi-square tests, whereas for ordinal vari- ables (e.g., income level, number of hospital beds), p  for trend values were calculated using the Cochran-Armitage trend test and reported in the tables. To further evaluate the associations between shift work and GAD, depressive symptoms, and fatigue, we conducted multivariable logistic regression analyses adjusting for relevant sociodemographic and occupational covariates. In addition, stratified analyses by the presence or absence of sleep problems were per- formed to explore potential effect modifications. Multicol- linearity among independent variables was assessed using variance inflation factors, and no significant multicollinear- ity was identified. Results are presented as odds ratios (ORs) with 95% confidence intervals (CIs). All statistical analyses were performed using SAS ver. 9.4 (SAS Institute, Cary, NC, USA). The significance level was set at p<0.05.

RESULTS

  Among 20,613 nurses, 12,252 nurses (59.4%) had GAD; 5,846 nurses (28.4%) had depressive symptoms, and 15,099 nurses (73.3%) had fatigue (Table 1). Mental health prob- lems were more prevalent among individuals engaged in shift work, with prevalence rates of GAD 62.7%, depressive symptoms 32.2%, and fatigue 76.7% (Table 1). In addition, mental health problems were more prevalent among indi- viduals with sleep problems, including GAD 82.5%, depres- sive symptoms 68.1%, and fatigue 94.5% (Table 1).
  Table 2 presents the association between shiftwork and mental health problems, including GAD, depressive symp- toms, fatigue after adjusting for all covariates. Nurses engaged in shift work had approximately 1.2, 1.4, and 1.4 higher risk of GAD, depressive symptoms, and fatigue, respectively (GAD: OR, 1.18; 95% CI, 1.10–1.27; depres- sive symptom: OR, 1.45; 95% CI, 1.32–1.59; fatigue: OR, 1.41; 95% CI, 1.30–1.53). In addition, nurses with sleep problems had approximately 3.3, 6.7, and 6.5 higher risk of GAD, depressive symptoms and fatigue (GAD: OR, 3.27; 95% CI, 2.96–3.62; depressive symptom: OR, 6.74; 95% CI, 6.17–7.36; fatigue: OR, 6.53; 95% CI, 5.54–7.71).
  Table 3 presents the associations between shift work and mental health problems stratified by sleep problem status, after adjusting for all covariates. Among nurses without sleep problems, shiftwork was associated with a 1.5-fold risk of depressive symptoms, with a statistically significant interaction. (OR, 1.54; 95% CI, 1.39–1.70). Although the interactions were not statistically significant, higher risks of GAD and fatigue were observed among nurses engaged in shift work without sleep problems (GAD: OR, 1.19; 95% CI, 1.10–1.28; fatigue: OR, 1.42; 95% CI, 1.31–1.54).

DISCUSSION

  This study demonstrated that shift work was significantly associated with increased risks of anxiety, depressive symp-toms, and fatigue among nurses. Importantly, even among nurses without sleep problems, those working shifts had a 1.5-fold higher risk of depressive symptoms compared with non-shift nurses, suggesting a close link between shift work and sleep-related mental health issues. Previous studies have shown similar results for the association between shift work and mental health. In a multi-center study in China, Li et al. [3] reported that 62.1% of shift-working nurses had depression and 58.8% had anxiety, and these outcomes were significantly associated with fatigue, psychological stress around night shifts, long working hours, and poor sleep quality. Chen et al. [12] highlighted occupational burnout as a critical predictor of poor mental health, and notably, night-shift workers reported worse health outcomes than their day-shift counterparts.
  The mechanism underlying the association between shift work and poor mental health outcomes may involve mul- tiple occupational and physiological pathways. Nurses en- gaged in shift work are more likely to experience circadian rhythm disruption, poor sleep quality, and cumulative fa- tigue, all of which have been strongly linked to higher risks of depression and anxiety [13,14]. Furthermore, organiza- tional factors such as heavy workload, long working hours, and job insecurity may exacerbate these adverse effects, sug- gesting that both biological and psychosocial mechanisms contribute to the increased vulnerability of shift-working nurses to mental health problems [15,16].
  Xu et al. [4] reported that lifestyle factors explained 31.3% of the association between shift work and depression, with sleep duration alone accounting for 7.7% of the mediated effect. In this context, sleep problems may represent an important pathway linking shift work to depression. There- fore, the absence of an additional effect of shift work among individuals with pre-existing sleep problems may reflect attenuation of the sleep-related mechanism rather than the absence of an association [17,18].
  This study has several limitations. First, this study em- ployed a cross-sectional survey design, which limits the ability to infer causality between variables. Although cohort data were available, a cross-sectional approach was adopted because mental health variables were measured using iden- tical items only in one survey module, limiting longitudinal comparisons across survey cycles. Consequently, temporal relationships and causal inference could not be assessed. Future research should adopt longitudinal study designs to clarify the temporal and causal relationships between shift work, sleep problems, and mental health outcomes. In particular, more detailed assessments of sleep quality, sleep regularity, and circadian disruption are warranted to better capture the multifaceted role of sleep in this associa- tion. Second, our study did not collect detailed information regarding the quantitative characteristics of shift work, such as shift duration, monthly shift hours, and the proportion of night shifts. Future research should further examine these characteristics—including the frequency and continuity of night shifts—to comprehensively understand their impact on physical and mental health.
  Third, although we adjusted for as many confounders as possible, the effects of residual confounders, such as work- place environments, sleep quality, and psychological inter- ventions, could not be completely eliminated. Furthermore, incorporating work environment–related factors—such as staffing levels, workload intensity, unit characteristics, and organizational support—into analytical models may provide a more comprehensive understanding of how occupational contexts interact with shift work to influence nurses’ mental health. Such evidence will be essential for developing tar- geted and effective interventions to protect the mental well-being of nurses. Fourth, reliance on self-reported data may have resulted in recall bias and inaccuracies in the measure- ment of key variables. Fifth, this study had limited ability to reflect differences in the severity of mental health symptoms across outcomes, as clear severity classification criteria were not available for some measures used in the analysis.
  Finally, this study is limited by the use of data collected between 2013 and 2015, approximately a decade before the current study period. Since then, the nursing work environment in Korea has undergone substantial changes, including the expansion of integrated nursing care services, implementation of shift system improvement programs, and the COVID-19 (coronavirus disease 2019) pandemic. Therefore, caution is needed when generalizing the find- ings to the current nursing environment. Nevertheless, the association between shift work and mental health has been consistently reported in previous studies, and our findings provide meaningful insights into nurses’ mental health and the improvement of working conditions.
  Despite these limitations, this study is significant in that it identified differences in mental health according to shift work status among nurses. Furthermore, it highlights the need for institutional support and policy development re- garding shift work to safeguard nurses’ health rights. The strengths of this study are as follows. First, this study pro- vides evidence on the association between shift work, sleep problems, and mental health outcomes among Korean nurs- es, a population that has been underrepresented in previ- ous research. Second, by jointly considering shift work and sleep problems and conducting subgroup and interaction analyses, this study offers a more nuanced understanding of mental health risks. Third, multiple mental health out- comes, including GAD, depressive symptoms, and fatigue, were examined simultaneously, allowing for a comprehen- sive assessment of nurses’ mental health.
  Shift work should not be viewed as a homogeneous expo- sure, as different scheduling characteristics such as rotating shifts, consecutive night shifts, insufficient recovery time, and irregular schedules may differentially affect nurses’ mental health. Although the present study could not ex-amine detailed shift patterns due to data limitations, future workforce policies and intervention programs should con- sider both the quantitative and qualitative characteristics of shift schedules.
  In conclusion, this study found that nurses engaged in shift work showed higher levels of GAD, depressive symp- toms, and fatigue compared with non–shift-working nurses. Notably, an increased risk of depressive symptoms was ob- served among shift-working nurses even in the absence of sleep problems, suggesting that shift work itself may exert a detrimental effect on mental health beyond sleep-related factors. These findings underscore the need for policy-level and organizational interventions that explicitly consider shift work as a key occupational risk factor for nurses’ men- tal health.
  Further research is needed to examine the severity of mental health symptoms, including depression, anxiety, and fatigue, in order to better understand the differential impact of shift work according to symptom severity. In addition, longitudinal studies using more recent datasets are warrant- ed to evaluate how changes in nursing policies and working environments have influenced nurses’ mental health over time.

CONFLICTS OF INTEREST

  No potential conflict of interest relevant to this article was reported.

ORCID

Soo Jeong Shim: https://orcid.org/0009-0005-8319-4078
Jin Young Nam: https://orcid.org/0000-0003-0540-8753