Health problems, turnover intention, and actual turnover among shift work female nurses: Analyzing data from a prospective longitudinal study.

AIMS: This study investigated health problems, turnover intention, and actual turnover among shift work nurses. While turnover intention is often used as a proxy variable for turnover, the relationship between these variables requires clarification. This study tested for relevant associations using prospective longitudinal data with a time lag of 12 months. We also tested for associations between health problems (sleep disturbance, fatigue, and depression) and turnover intentions/turnover, with a focus on the mediating role of turnover intention.
METHODS: This study conducted a secondary analysis of data from the Shift Work Nurses' Health and Turnover project, which is a prospective longitudinal cohort study. We analyzed health problems, turnover intention, and actual turnover. The data were analyzed via descriptive statistics, the Pearson's chi-squared test, independent t-test, univariable logistic regression, multiple logistic regression, and causal mediation.
RESULTS: Participants included 491 shift work female nurses. Of these, 112 (22.8%) had turnover intention, while 38 (7.7%) left their jobs within the 12-month period of investigation. Of the 112 with turnover intention, 22 left their jobs (OR 5.68. 95% CI 2.84-11.36). The logistic regression analysis showed that sleep disturbance and fatigue were associated with turnover intentions and actual turnover, while depression was only associated with turnover intention. The causal mediation analysis showed that turnover intention mediated the relationship between health problems (sleep disturbance and fatigue) and actual turnover (sleep disturbance OR 1.31, 95%CI = 1.02-1.60; fatigue OR 2.11, 95%CI = 1.50-2.68); sleep disturbance had a natural direct effect on actual turnover (OR 2.66, 95%CI,2.07-3.21).
CONCLUSION: Turnover intention strongly predicted actual turnover. Sleep disturbances may result in turnover, even in cases without existing turnover intention. These findings highlight the need for early interventions aimed at preventing and alleviating sleep disturbances for shift work female nurses.

Introduction

The high rate of turnover among nurses is a serious health system issue [1-3] that has been exacerbated by the COVID-19 pandemic [4]. Looking at conditions in Korea, the turnover rate has continually increased over the past few years, especially among novice nurses, who leave their jobs at triple the rate shown for all nurses nationwide [5].

The resulting nursing shortages have been associated with decreased patient satisfaction, an increased risk of infection, and longer hospital stays [6]. Meanwhile, nurses who experience high peer turnover are left with heavier workloads that can decrease job satisfaction [7]. This also poses issues related to job safety for nurses who continue to work during shortages, especially due to the increased risk of exposure to blood and other bodily fluids [8]. From the administrative standpoint, hospitals are faced with additional financial burdens due to the loss of skilled nurses and subsequent need to recruit and train new nurses [9].

In previous research, turnover intention has been used as a proxy variable for actual turnover, as it may be the best predictor according to the Theory of Planned Behavior [10, 11]. This is also because actual turnover is difficult to investigate among nurses, particularly due to the need for a longitudinal approach, which requires more time, financing, and effort than cross-sectional research. Further, ethical issues may arise when tracking these behaviors through personal information, which is often required in longitudinal surveys [12]. Amid these concerns, there is a general lack of research on the relationship between turnover intention and actual turnover, with inconsistent results between studies [2, 10]. Finally, reports have shown that health problems such as sleep disturbances, fatigue, and depression can influence turnover intention among nurses, but there is a lack of evidence about the relationships between these health problems and actual turnover [13-15].

Due to these gaps in the literature, this study examined the relationships between health problems, turnover intention, and actual turnover by conducting a secondary analysis of longitudinal data with a time lag of 12 months. To do so, we explored the associations between each health problem and turnover intention/actual turnover, with a focus on the mediating role of turnover intention in the relationship between health problems and actual turnover.

Materials and methods

Study design

This study conducted a secondary analysis of cohort data collected via the Shift Work Nurses’ Health and Turnover (SWNHT) project (2018–2020).

Data source and sampling

The SWNHT was a longitudinal prospective cohort study that investigated relationships between health and turnover among shift work nurses in Korea. The detailed study methods have been published [13]. Participants included 594 female nurses (i.e., 294 novice nurses with no exposure to rotating shift work and 300 nurses with exposure to eight-hour rotational work, including night shifts, for a period lasting at least one month). Because health problems may vary according to sex [16, 17], the SWNHT was limited to female nurses. For novice nurses, data were collected three times, including before exposure to shift work (novice registered nurse [NRN] T0, n = 294), six months after work (NRN T1, n = 204), and 12 months after T1 (NRN T2, n = 204). For experienced registered nurses, data were collected twice, including at baseline (experienced registered nurse [ERN] T1, n = 300) and 12 months after T1 (ERN T2, n = 269). To enroll nurses, we attached a recruitment notice to the ward bulletin boards and also distributed survey envelope packages after their lecture time in the hospital. Nurses who wished to participate in the study voluntarily contacted the research team. After agreeing to participate in the study, all nurses signed the consent form and completed the baseline questionnaire. All data were collected between March 2018 and April 2020 at two tertiary hospitals in Seoul, South Korea. The number of shift work nurses in the two study hospitals was approximately 1,400 and 1,900, and data from 244 and 247 nurses were analyzed for this study, respectively. Records of nurses leaving the hospital were surveyed from the participants or the nursing department in which they worked during the T2 survey. The SWNHT was approved by the institutional review boards at both tertiary hospitals (IRB No. H-1712-094-907, 2017-12-075-002). To secondary analysis data from the SWNHT, this study was approved by the institutional review boards at Seoul National University (IRB No. E2011/003-012).

In this study, we analyzed data from both NRN T1 (n = 204) and ERN T1 (n = 300), thus spanning a collection period lasting from March 2018 to January 2019. We defined shift work as a combination of day, evening, and night shifts with varying numbers of shifts in a row of 1–5 days and included shift work female nurses with at least 1 month of experience and no omission in major variables. According to these criteria, this subset included a final sample size of 491 participants, excluding 12 with no engagement in eight-hour rotational shift work and one who did not answer items related to the major variables. For these participants, actual turnover was checked at T2, 12 months after the NRN T1, and ERN T1 surveys.

Measures

General characteristics

Each participant provided their information for the following: age (years), education level (bachelor’s degree or lower/master’s degree or higher), marital status (single/married), whether they had children (yes/no), work unit (general ward, intensive care unit, delivery room, or emergency room), and total shift work experience (less than one year/more than one year).

Turnover intention and actual turnover

Turnover intention is defined as an employee’s voluntary resignation intention or attempt [18, 19]. In our study, each participant answered the following question using one of four options (strongly agree, agree, disagree, or strongly disagree): “I plan on staying for the next year” [20]. Prior to the analysis, turnover intention was converted into a dichotomous mediating variable, in which 0 = intent to stay (strongly agree or agree) and 1 = intent to leave (disagree or strongly disagree). Based on turnover data from the T2 survey, we determined whether actual turnover had occurred within a 12-month period following the completion of the T1 survey.

Sleep disturbance

Sleep disturbance was measured using the Korean version of the Insomnia Severity Index (ISI), which was developed by Morin [21] and translated by the Korean Sleep Research Society [22]. The Korean ISI comprised of seven items that are rated in a 5-point scale (0–4 points), and the score ranges from 0 to 28. Higher scores indicate lower sleep quality and scores above 10 indicate sleep disturbance during the past 2 weeks [21]. In the study context, the Korean ISI received a Cronbach’s alpha of 0.92.

Fatigue

Fatigue was measured using the Fatigue Severity Scale (FSS), which comprised of nine items concerning the degree of fatigue over the previous week. Each item is rated in a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree), and average scores obtained by dividing the total score (range 9–63) by the number of items indicates the degree of fatigue; the cutoff point for fatigue is more than 4 points on average [23]. In the study context, the FSS received a Cronbach’s alpha of 0.91.

Depression

Depression was assessed using the shortened Center for Epidemiological Studies Depression Scale (CES-D), which is comprised of 10 items concerning depressive feelings and thoughts during the previous week. Each item is rated in a 4-point scale (0 = less than one day, 3 = about five to seven days), with higher total scores (range 0–30) indicating more depressive symptoms; total scores of 10 or above indicate depression [24]. In the study context, the shortened CES-D received a Cronbach’s alpha of 0.87.

Statistical analyses

All analyses were conducted using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). Descriptive statistics (frequencies, percentages, means, and standard deviations) were analyzed for the general characteristics, while Pearson’s chi-squared test and independent t-test were used to identify the differences between turnover intention and actual turnover according to the general characteristics and health problems.

Multiple logistic regressions were used to investigate the associations between health problems and turnover intention, including all general characteristics except for age used as covariates, as age was highly correlated with total shift work experience (r = 0.92, p < 0.001). In the multiple logistic regressions for actual turnover, we added turnover intention to the models to confirm the weakening associations between health problems and actual turnover.

We conducted a causal mediation analysis (CAUSALMED procedure in SAS) to explore the mediating effects of turnover intention. Specifically, the causal mediation analysis is a statistical method based on the counterfactual framework, and which enables researchers to estimate exactly even when the mediator and outcome are dichotomous or if there is an interaction between the exposure and the mediator [25]. Three estimates were obtained through the analysis, including the total effect (TE), natural direct effect (NDE), and natural indirect effect (NIE). TE is the sum of the NDE and NIE, while NDE is the effect of a shift in the outcome based on exposure, assuming that the mediator is fixed, and NIE is the effect of a shift in the outcome based on the mediator, assuming that the exposure is fixed. We also obtained the mediated percentage, which is the proportion of NIE of TE [26]. Fig 1 shows a directed acyclic graph of this causal mediation analysis.

Fig 1

Directed acyclic graph for the mediation analysis.

Results

Participant characteristics

As mentioned, the analyzed participants included 491 female nurses with exposure to shift work, including night shifts. Their mean age was 26.1 years (standard deviation [SD] = 4.30); 87.8% (n = 431) were single. Most had less than five years of experience as nurses (n = 384, 78.2%). Of all participants, 112 (22.8%) had turnover intention and 38 (7.7%) left their jobs. Of the 112 (22.8%) with turnover intention, 22 left their jobs (OR, 5.54. 95% CI 2.79–10.99). Notably, the prevalence of both sleep disturbance and fatigue was significantly higher among the participants with turnover intention and those who left their jobs. Meanwhile, the prevalence of depression was significantly higher among those with turnover intention (χ2 = 15.34, p<0.001) (Table 1).

Table 1

General characteristics and health problems by turnover intention and actual turnover.

Variables	Categories	Total	Intend to stay	Intend to leave	χ2 or t	Ρ	Stayer	Leaver	χ2 or t	Ρ
		(n = 491, 100.0%)	(n = 379, 77.2%)	(n = 112, 22.8%)			(n = 453, 92.3%)	(n = 38, 7.7%)
		n(%) or M±SD	n(%) or M±SD	n(%) or M±SD			n(%) or M±SD	n(%) or M±SD
Age (years)		26.1±4.30	26.00±4.06	26.44±5.02	-0.85	0.399	26.15±4.43	25.52±2.12	1.56	0.123
Education	≤ BSN	449(91.5)	348(91.8)	101(90.2)	0.30	0.585	412(91.0)	37(97.4)	1.85	0.235
	≥ MSN	42(8.5)	31(8.2)	11(9.8)			41(9.0)	1(2.6)
Marital status	Single	431(87.8)	335(88.4)	96(85.7)	0.58	0.447	395(87.2)	36(94.7)	1.86	0.173
	Married	60(12.2)	44(11.6)	16(14.3)			58(12.8)	2(5.3)
Has children	Yes	31(6.3)	23(6.1)	8(7.1)	0.17	0.681	31(6.8)	0(0.0)	2.78	0.096
	No	460(93.7)	356(93.9)	104(92.9)			422(93.2)	38(100.0)
Work unit	Ward	364(74.1)	278(73.4)	86(76.8)	0.62	0.733	333(73.5)	31(81.6)	1.19	0.550
	ICU	110(22.4)	87(22.9)	23(20.5)			104(23.0)	6(15.8)
	DR, ER	17(3.5)	14(3.7)	3(2.7)			16(3.5)	1(2.6)
Total shift work experience	Less than one year	233(47.5)	186(49.1)	47(42.0)	1.75	0.185	215(47.5)	18(47.4)	0.00	0.991
	One year or more	258(52.5)	193(50.9)	65(58.0)			238(52.5)	20(52.6)
Sleep disturbance	Yes	307(62.5)	220(58.1)	87(77.7)	14.22	<0.001**	275(60.7)	32(84.2)	8.27	0.004**
	No	184(37.5)	159(41.9)	25(22.3)			178(39.3)	6(15.8)
Fatigue	Yes	321(65.4)	227(59.9)	94(83.9)	22.06	<0.001**	289(63.8)	32(84.2)	6.45	0.011*
	No	170(34.6)	152(40.1)	18(16.1)			164(36.2)	6(15.8)
Depression	Yes	202(41.1)	138(36.4)	64(57.1)	15.34	<0.001**	183(40.4)	19(50.0)	1.34	0.248
	No	289(58.9)	241(63.6)	48(42.9)			270(59.6)	19(50.0)

BSN = Bachelor of Science in Nursing; MSN = Master of Science in Nursing; ICU = intensive care unit; DR = delivery room; ER = emergency room

*p<0.05

**P<0.01

Association between health problems and turnover intention/actual turnover

According to the multiple logistic regression analysis with covariates, sleep disturbance, fatigue, and depression were each associated with turnover intention (Table 2). Moreover, turnover intention was significantly associated with actual turnover (OR 5.68, 95%CI = 2.84–11.36). While sleep disturbance was associated with actual turnover regardless of whether we adjusted for turnover intention (OR 3.48, 95%CI = 1.41–8.57), this association was weakened by the inclusion of turnover intention (OR 2.73, 95%CI = 1.08–6.83). Finally, fatigue was positively associated with actual turnover (OR 3.03, 95%CI = 1.23–7.45), but this was no longer significant after adjusting for turnover intention (Table 3).

Table 2

Associations between health problems and turnover intention according to the multiple logistic model.

Variable	Between sleep disturbance and turnover intention		Between fatigue and turnover intention		Between depression and turnover intention
	OR	95% CI	OR	95% CI	OR	95% CI
Sleep disturbance	2.50**	1.52–4.11	NA		NA
Fatigue	NA		3.52**	2.03–6.10	NA
Depression	NA		NA		2.70**	1.72–4.24

Adjusted for education, marital status, children (yes/no), work unit, and total shift work experience

*p<0.05

**P<0.01

Table 3

Associations between health problems and actual turnover according to the multiple logistic model.

Variable	Between turnover intention and actual turnover		Between sleep disturbance and actual turnover		Among sleep disturbance, turnover intention, and actual turnover		Between fatigue and actual turnover		Among fatigue, turnover intention, and actual turnover		Between depression and actual turnover		Among depression, turnover intention, and actual turnover
	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI
Turnover intention	5.68**	2.84–11.36	NA		5.02**	2.48–10.13	NA		4.94**	2.43–10.03	NA		5.72**	2.79–11.68
Sleep disturbance	NA		3.48**	1.41–8.57	2.73*	1.08–6.83	NA		NA		NA		NA
Fatigue	NA		NA		NA		3.03*	1.23–7.45	2.11	0.83–5.37	NA		NA
Depression	NA		NA		NA		NA		NA		1.42	0.71–2.81	0.97	0.46–2.02

Adjusted for education, marital status, children (yes/no), work unit, and total shift work experience

*p<0.05

**P<0.01

Mediating effects of turnover intention

Sleep disturbance and fatigue had significant effects on actual turnover (sleep disturbance OR 3.50, 95%CI = 2.15–4.80; fatigue OR 3.23, 95%CI = 1.02–5.01), while turnover intention mediated sleep disturbance, fatigue, and depression (NIE of sleep disturbance OR 1.31, 95%CI = 1.02–1.60; NIE of fatigue OR 2.11, 95%CI = 1.50–2.68; NIE of depression OR 1.34, 95%CI = 1.02–1.72). The NDE of sleep disturbance was also significant (OR 2.66, 95%CI,2.07–3.21, Table 4).

Table 4

The mediating effect of turnover intention in the relationship between health problems and actual turnover.

Mediation of turnover intention	Association of sleep disturbance and actual turnover		Association of fatigue and actual turnover		Association of depression and actual turnover
	OR	95% CI	OR	95% CI	OR	95% CI
Total effect	3.50**	2.15–4.80	3.23*	1.02–5.01	1.35	0.33–2.37
Natural direct effect	2.66**	2.07–3.21	1.52	0.59–2.46	1.01	0.28–1.74
Natural indirect effect	1.31*	1.02–1.60	2.11**	1.50–2.68	1.34*	1.02–1.72
Percentage mediated, %	33.5		50.2		98.0

Adjusted for education, marital status, children (yes/no), work unit, and total shift work experience (causal mediation analysis)

*p<0.05

**P<0.01

Discussion

This study investigated the longitudinal relationships between health problems, turnover intention, and actual turnover among shift work nurses. First, we found that turnover intention significantly influenced actual turnover. Second, the participants frequently experienced a variety of health problems that affected their work. Here, the analyses showed that sleep disturbance and fatigue were each associated with turnover intention and turnover, while depression was only associated with turnover intention. Surprisingly, we also found that sleep disturbance had a direct effect on actual turnover, which indicates that participants were prone to abruptly leaving their nursing jobs upon suffering from sleep disturbance, even in cases without existing turnover intention.

Shift work can cause a variety of physical and mental health problems and could become more serious as the period increases [27]. About 75% of Korean nurses work in 8‐hr rotations including day, evening, and night shifts [28], which means that Korean nurses’ health may be vulnerable. In previous studies, it has been reported that nurses’ health problem not only lower the quality of life but also affect the quality of nursing, which in turn can lead to turnover intention [29, 30]. In this study, 22.8% of participants had turnover intention; of this subset, 19.6% ultimately left their jobs within a 12-month period. Meanwhile, the literature shows a variety of different turnover intention rates (4~64%) in various nursing samples [31-35]. Such a wide range may partially be due to the different measurement tools used between studies. While this study considered turnover intention based on responses to items such as “I plan on staying for the next year”, other studies used different reference points, including how often participants thought about turnover in the past [35] and whether they were looking for other jobs [34]. The actual reported turnover rates (2~63%) also differ between studies, which may be due to the various definitions used for turnover [35-39]. While this study defined turnover as an event in which a given participant terminated their employment by leaving the hospital, other studies defined turnover as an event in which the participant transferred to a different work unit or completely left the nursing sector [35, 36]. Above all, such differences may be due to the wide diversity of hospital environments. We recruited participants who worked in tertiary hospitals with a relatively high nurse-to-patient ratio. Most of the participants worked in general wards or intensive care units, taking care of an average of 12 patients in general wards and 2–3 patients in intensive care units. While the severity of patients was high, the salary and working conditions were better than it is in other hospitals in Korea. However, other studies have reported turnover rates in different hospital environments in various countries, regions, and medical institutions [39].

We found a high causality between turnover intention and turnover (OR, 5.54. 95% CI 2.79–10.99). Our causal mediation analysis further revealed that the associations between health problems and actual turnover were partially mediated by turnover intention, meaning that sleep disturbance and/or fatigue may lead to actual turnover by increasing turnover intention. These results support previous studies showing that turnover intention plays mediating roles in the relationships between job satisfaction and actual turnover, quality of work life and actual turnover [36-38].

Among the health problems that were related to actual turnover, sleep disturbance had the greatest total effect. Surprisingly, the natural direct effect on actual turnover was also statistically significant, which indicates that sleep disturbance may affect actual turnover without increasing turnover intention. Of the 16 nurses who did not have previous turnover intention but ultimately left their jobs, 14 experienced sleep disturbance. This shows that poor sleep quality may induce other problems that influence voluntary job termination. Indeed, previous studies have shown that sleep disturbance affects nursing performance by impairing cognitive function, empathy, and judgment, and increased the potential for errors that threaten patient safety [40-43]. Further, these conditions pose risks that may impact physical and mental health for nurses, including gastrointestinal disorders, cardiovascular disease, obesity, diabetes, and depression [27, 41, 44].

Previous studies have reported a prevalence of sleep disturbance ranging from 57–83% in shift work nurses, which is higher than that found for the general adult population [44, 45]. In this study, a significant number of participants (62%) had sleep disturbance, with an increased prevalence among those with turnover intention (77.7%) and those who left their jobs (84.2%). A number of previous studies have similarly reported associations between sleep disturbance and turnover [14, 15]. As a whole, these findings show that sleep disturbance is a serious health problem for shift work nurses, thus emphasizing the need for multi-faceted organizational support.

In addition, we found that fatigue may cause actual turnover by increasing turnover intention. However, fatigue has received relatively less attention than sleep disturbance due to the known associations between turnover intention and this latter variable [46]. Still, fatigue is a common symptom for shift workers [47, 48]. In this study, 65.4% of all participants and 84.2% of those who left their jobs reported fatigue, which has previously been associated with an increased potential for chronic disease and mental stress [46, 49]. In nurses, fatigue disrupts optimal performance and negatively affects skills while also increasing the risks for medical error, exposure to blood and other body fluids, and musculoskeletal disorder [50]. It is noteworthy that the current results suggest that fatigue may affect both turnover intention and actual turnover. In this regard, managers should work to identify fatigue-related factors and possibly adopt flexible strategies to reduce fatigue among shift work nurses, as this may promote health while reducing actual turnover.

In a previous cross-sectional study, depression was positively associated with turnover intention among nurses [13, 51]; however, a longitudinal cohort study found no such association with actual turnover [52]. Likewise, this study found that depression was associated with turnover intention but not with actual turnover. However, the prevalence of depression is lower in our study than as reported in previous research [53]. We also did not investigate whether participants were taking antidepressants or had been diagnosed with depression. These should be considered when interpreting the results.

Although this study produced new evidence concerning the relationships between health problems, turnover intention, and actual turnover, there were also some limitations. First, we only considered the most commonly reported health problems in shift workers. These health problems could have many other causes besides shift work and may have preceded nurses’ exposure to shift work. In addition, other untested health issues may have had affected turnover intention/turnover. Second, because there is a high turnover rate among novice nurses in Korea, the SWNHT included a relatively high proportion of such participants. In this regard, results may differ when including a higher proportion of experienced nurses. Third, the follow-up period was only 12 months. Additional research is needed to investigate nurse turnover and any related factors over longer periods of time.

Conclusions

In this study, turnover intention was a strong predictor of actual turnover. This highlights the potential connections between turnover intention and actual turnover. Further, turnover intentions led to turnover after some period of time, meaning that turnover was not a sudden event, but a multi-stage process. Finally, there is s strong potential that sleep disturbance may cause turnover, even for nurses without existing turnover intention. This urgently emphasizes the need for early interventions aimed at preventing and alleviating sleep disturbances among shift work nurses.

(SAS7BDAT)

Click here for additional data file.

7 Mar 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Mohammad Hossein Ebrahimi

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified (1) whether consent was informed and (2) what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If the need for consent was waived by the ethics committee, please include this information.

3. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors explored the association between health problems and turnover intention/actual turnover, approaching sleep disturbance, depression and fatigue as health problems, using validated instruments. Sample consists of young women, which may limit the extrapolation of results. However, some comments are needed:

1- What was considered shift work? Was this work organization uniform among the participants?

2- Participants were classified in relation to the total shift work experience as less than one year/more than one year, which can be considered a short time in shift work, especially in relation to the chronic effects on the health of health care workers (HCWs) . What is the reason for this classification? What is the impact of this difference on the analysis of results?

3- Was the variable considered in the analysis for another job concurrently with the current one? Is this information available?

4- Regarding the instruments used, what is the previous period considered in the sleep disturbance and depression analyzes (previous week, previous month)?

5- Shift work can impact the health of HCWs, including aggravated by the time of exposure to risk. In this sense, other health problems that may also impact the turnover intention/actual turnover were evaluated, for example: cardiovascular, metabolic, musculoskeletal diseases?

Reviewer #2: This is quite interesting work about the relationship between shift work and some health problems among nurses. It is well written and quite well structured but, in my opinion, it lacks of numerous basic data. There are many points that need to be better clarified.

Which is the the reason why, for novice nurses, data were collected 3 times instead of 2.

It is known that the health problems explored (fatigue, depression, sleep disturbance) may have multiple causes, why they haven't taken into consideration? Moreover, there is no information about the previous presence of those problems among the population studied.

The definition applied for "turnover intention" should be indicated first in the text.

There is no information about the hospital ward where the participants work. I think that this information should be very useful to better understand the reasons for turnover intention.

It is not discussed the relationship between shift work experience and turnover intention, even if the study sample was mainly made of shift workers.

The inclusion criteria of the respondents have not been indicated, even if referred to a previous study, it should be useful to report them in the methods section.

Reviewer #3: Dear authors thank you for this research:

1- I suggest adding female nurses in the title as you only include them and as you mention the turnover intention differ between both genders.

2- could you give us some few details about the number of hospitals from which the nurses were included

3- write the significance level under the tables or the statistical analysis part of the methods.

4- for the questionnaire used please add the total score range for each domain you examine( according to the number of questions and the score points).

5- was a pilot study conducted to assess the clarity of the questionnaires

Good luck

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Giuseppe Buomprisco

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

20 May 2022

Reviewer #1: The authors explored the association between health problems and turnover intention/actual turnover, approaching sleep disturbance, depression and fatigue as health problems, using validated instruments. Sample consists of young women, which may limit the extrapolation of results. However, some comments are needed:

Thank you for your valuable feedback. Our responses to your remarks are presented in bold.

1. What was considered shift work? Was this work organization uniform among the participants?

Thank you for pointing this out. We defined shift work as a combination of 8-hour day, evening, and night shifts with varying numbers of shifts in a row of 1–5 days. All participants in our study were exposed to the same pattern of shift work. We have now added our definition of shift work and inclusion criteria of participants in the Materials and Methods section. (page 6, lines 109-111).

“We defined shift work as a combination of day, evening, and night shifts with varying numbers of shifts in a row of 1–5 days and included shift work female nurses with at least 1 month of experience and no omission in major variables.”

2. Participants were classified in relation to the total shift work experience as less than one year/more than one year, which can be considered a short time in shift work, especially in relation to the chronic effects on the health of health care workers (HCWs) . What is the reason for this classification? What is the impact of this difference on the analysis of results?

Thank you for pointing this out. There are three reasons for classifying the total shift work experience as less than one year/more than one year in our study. First, a previous study reported that the competency of novice nurses improved around 1 year after they started working as nurses, and suggested 1 year as a period to distinguish between novice nurses and experienced nurses . Second, the variables (fatigue, depression, and sleep) we considered in this study are known as chronic health problems in shift workers, but they are also symptoms that appear highest around 6 months after starting shift work. After the period, these health problems either slightly improve or remain over time . Lastly, the turnover rate of nurses in the first year was the highest in Korea. Since the turnover rate was one of the important variables in our study, we thought it was reasonable to divide the subjects into two groups (less than one year/more than one year) considering the aforementioned results.

3. Was the variable considered in the analysis for another job concurrently with the current one? Is this information available?

Thank you for pointing out a very important issue. Although we did not study another job, a previous study reported that shift work was associated with long sleep and increased fatigue in hospital workers with various job titles including nurses . In addition, there was a study that reported that sleep problems and mental disorders are caused by shift work in various occupations, using national data . In a future study, if we consider the same variables in another job, participants’ characteristics in different occupations could be well represented.

4. Regarding the instruments used, what is the previous period considered in the sleep disturbance and depression analyzes (previous week, previous month)?

We apologize for not clearly describing the period. The period suggested in the instruments is the last 2 weeks for sleep disturbance and 1 week for fatigue and depression. We added information about the period specified in instruments in the Materials and Methods section. (page 7, lines 138-139, lines 143-144, page 8, lines 151-153).

“Higher scores indicate lower sleep quality and scores above 10 indicate sleep disturbance during the past 2 weeks [21].”

“Fatigue was measured using the Fatigue Severity Scale (FSS), which comprised of nine items concerning the degree of fatigue over the previous week.”

“Depression was assessed using the shortened Center for Epidemiological Studies Depression Scale (CES-D), which comprised of 10 items concerning depressive feelings and thoughts during the previous week.”

5. Shift work can impact the health of HCWs, including aggravated by the time of exposure to risk. In this sense, other health problems that may also impact the turnover intention/actual turnover were evaluated, for example: cardiovascular, metabolic, musculoskeletal diseases?

Unfortunately, we did not consider a cardiovascular or metabolic disease. The participants of our study were young female nurses with an average age of 26 which also corresponds to those of Korean average female nurses. Considering these conditions, it was expected that the prevalence of cardiovascular or metabolic diseases was very low and that the association with turnover could not be explored. We previously reported, however, that there was no association with turnover intention in musculoskeletal diseases .

Reviewer #2: This is quite interesting work about the relationship between shift work and some health problems among nurses. It is well written and quite well structured but, in my opinion, it lacks of numerous basic data. There are many points that need to be better clarified.

Thank you for your valuable feedback. Our responses to your remarks are presented in bold.

1. Which is the reason why, for novice nurses, data were collected 3 times instead of 2.

Thank you for your interest in our study. The reason we conducted the survey three times for novice nurses is that, unlike experienced nurses, one additional survey was conducted to explore the health status before exposure to shift work, which was excluded on the part of the experienced nurses. The other two surveys were conducted after six months with a time lag of 12 months for novice nurses, which is similar to that of experienced nurses (Figure 1).

Figure 1. Schematic overview of the Shift Work Nurses’ Health and Turnover (SWNHT) study. Reprinted from “Association between Health Problems and Turnover Intention in Shift Work Nurses: Health Problem Clustering”, by J-S. Ki et al., 2020, Int. J. Environ. Res. Public Health, 17(12), 4532.

2. It is known that the health problems explored (fatigue, depression, sleep disturbance) may have multiple causes, why they haven't taken into consideration? Moreover, there is no information about the previous presence of those problems among the population studied.

Thank you for pointing out a very important issue. We agree with the reviewer. We assumed that the health problems explored in our study (sleep disturbance, fatigue, and depression) were caused mainly by shift work. However, as you pointed out, these health problems could have many other causes besides shift work and some of these problems could have preceded the shift nurses’ exposure to shift work. We now have added this as a limitation in the Discussion section as the reviewer suggested (page 18, lines 310-314).

“First, we only considered the most commonly reported health problems in shift workers. These health problems may have many other causes besides shift work and may have preceded nurses’ exposure to shift work. In addition, other untested health issues may have had affected turnover intention/turnover.”

3. The definition applied for "turnover intention" should be indicated first in the text.

We apologize for the omission of the definition. We have added the definition of the turnover intention in the Materials and Methods section (page 7, lines 125-126).

“Turnover intention is defined as an employee's voluntary resignation intention or attempt[18, 19].”

4. There is no information about the hospital ward where the participants work. I think that this information should be very useful to better understand the reasons for turnover intention.

Thank you for your suggestion. We added information about the hospitals where the participants work in the Discussion section (page 16, lines 257-263 ).

“We recruited participants who worked in tertiary hospitals with a relatively high nurse-to-patient ratio. Most of the participants worked in general wards or intensive care units, which took care of an average of 12 patients in general wards and 2-3 patients in intensive care units. While the severity of patients was high, the salary and working conditions were superior to other hospitals in Korea. However, other studies have reported turnover rates in different hospital environments in various countries, regions, and medical institutions [39].”

5. It is not discussed the relationship between shift work experience and turnover intention, even if the study sample was mainly made of shift workers.

Thank you for your suggestion. We have now added the relationship between shift work and turnover intention in the Discussion section (page 15, lines 240-244 ).

“Shift work can cause a variety of physical and mental health problems and could become more serious as the period increases [27]. About 75% of Korean nurses work in 8‐hr rotations including day, evening, and night shifts[28], which means that Korean nurses' health may be vulnerable. In previous studies, it has been reported that nurses’ health problems not only lower the quality of life but also affect the quality of nursing, which in turn can lead to turnover intention [29, 30].”

6. The inclusion criteria of the respondents have not been indicated, even if referred to a previous study, it should be useful to report them in the methods section.

Thank you for pointing this out. we have now added our definition of shift work and inclusion criteria of participants in the Materials and Methods section. (page 6, lines 109-114).

“We defined shift work as a combination of day, evening, and night shifts with varying numbers of shifts in a row of 1–5 days and included shift work female nurses with at least 1 month of experience and no omission in major variables. According to these criteria, this subset included a final sample size of 491 participants, excluding 12 with no engagement in eight-hour rotational shift work and one who did not answer items related to the major variables.”

Reviewer #3: Dear authors thank you for this research:

Thank you for your valuable feedback. Our responses to your remarks are presented in bold.

1. I suggest adding female nurses in the title as you only include them and as you mention the turnover intention differ between both genders.

Thank you for pointing this out. We agree with the reviewer’s opinion. We have changed the title to “Health problems, turnover intention, and actual turnover among shift work female nurses: Analyzing data from a prospective longitudinal study”.

2. could you give us some few details about the number of hospitals from which the nurses were included

All data were collected at two tertiary hospitals in Seoul, South Korea. The number of shift work nurses in the two study hospitals averaged 1,700 with an average of 1,800 beds. For this study, we analyzed data from 244 and 247 nurses, respectively. We have added these details in the Materials and Methods section (page 6, lines 101-103)

“The number of shift work nurses in the two study hospitals was approximately 1,400 and 1,900, and data from 244 and 247 nurses were analyzed for this study, respectively.”

3. write the significance level under the tables or the statistical analysis part of the methods.

Thank you for your suggestion. We have now added the significance level under the tables (Table 1, Table 2, Table 3, Table 4).

4. for the questionnaire used please add the total score range for each domain you examine (according to the number of questions and the score points).

Thank you. We have added the total score range for the key variables.

Sleep Disturbance (page 7, lines 137-139)

“The Korean ISI comprised of seven items that are rated in a 5-point scale (0–4 points), and the score ranges from 0 to 28. Higher scores indicate lower sleep quality and scores above 10 indicate sleep disturbance during the past 2 weeks [21].”

Fatigue (page 7, lines 144-147)

“Each item is rated in a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree), and average scores obtained by dividing the total score(range 9-63) by the number of items indicates the degree of fatigue; the cutoff point for fatigue is more than 4 points on average [23].”

Depression (page 8, lines 153-155)

“Each item is rated in a 4-point scale (0 = less than one day, 3 = about five to seven days), with higher total scores (range 0-30) indicating more depressive symptoms; total scores of 10 or above indicate depression [24].”

5. was a pilot study conducted to assess the clarity of the questionnaires

Yes, you are right. We previously conducted a pilot study and published it.

(Baek J, Choi-Kwon S. Sleep Patterns, Alertness and Fatigue of Shift Nurses according to Circadian Types. Journal of Korean Biological Nursing Science. 2017;19(3):198-205. https://doi.org/10.7586/jkbns.2017.19.3.198)

Submitted filename: Response_to_Reviewers_Ki.docx

Click here for additional data file.

22 Jun 2022

Health problems, turnover intention, and actual turnover among shift work female nurses: Analyzing data from a prospective longitudinal study

PONE-D-21-32791R1

Dear Dr. Choi-Kwon,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Mohammad Hossein Ebrahimi

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

28 Jun 2022

PONE-D-21-32791R1

Health problems, turnover intention, and actual turnover among shift work female nurses: Analyzing data from a prospective longitudinal study

Dear Dr. Choi-Kwon:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Mohammad Hossein Ebrahimi

Academic Editor

PLOS ONE