Impact of nursing shift patterns on work-related musculoskeletal disorders: a systematic review and meta-analysis

Abstract

Musculoskeletal disorders (MSDs) are among the leading causes of disability worldwide and substantially affect the nursing workforce due to high physical demands, biomechanical strain, and irregular work schedules. Growing evidence suggests that shift work, particularly rotating and night schedules, may contribute to work-related musculoskeletal disorders (WRMSDs); however, findings remain inconsistent across studies. This study aimed to examine, through a systematic review and meta-analysis of observational studies, the association between distinct nursing shift patterns (day, rotating, and fixed night shifts) and generalized or region-specific musculoskeletal outcomes among nurses. A systematic search was conducted in PubMed, Scopus, Web of Science, and CINAHL between December 2024 and August 2025. The review was prospectively registered in PROSPERO (CRD420250656652). Studies were eligible if they included samples comprising at least 50% practicing nurses, compared at least two shift types (day, rotating, or fixed night shifts), and reported musculoskeletal outcomes. Observational study designs were included without publication date restrictions. Methodological quality was assessed using the Newcastle–Ottawa Scale for cohort studies and the Joanna Briggs Institute checklist for cross-sectional studies. Risk of bias was evaluated using the ROBINS-E tool. Although the systematic review included a broad range of musculoskeletal outcomes, only musculoskeletal pain outcomes were sufficiently homogeneous to be included in the quantitative synthesis. Random-effects meta-analyses were conducted comparing day versus rotating shifts and day versus fixed night shifts. Twenty-four studies including 102,159 nurses were included in the systematic review, and thirteen studies were incorporated into the meta-analyses. No statistically significant associations were identified between rotating shifts and musculoskeletal pain compared with day shifts (pooled estimate = 1.15; 95% CI: 0.98–1.35; I² = 7.2%). Likewise, fixed night shifts were not significantly associated with musculoskeletal pain compared with day shifts (pooled estimate = 1.09; 95% CI: 0.94–1.27; I² = 81.1%). Stratified analyses across anatomical regions, including neck, back/lumbar, shoulder, and lower limb pain, also showed no statistically significant associations, although several pooled estimates tended to be higher among rotating shift workers. The available evidence suggests that the relationship between shift work and musculoskeletal pain among nurses is complex and likely influenced by multiple occupational, ergonomic, psychosocial, and organizational factors. Although no statistically significant associations were identified in the meta-analysis, rotating and fixed night shifts showed trends toward higher musculoskeletal pain estimates in several anatomical regions. Further longitudinal studies using more standardized definitions of shift work exposure and musculoskeletal outcomes are needed.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12912-026-04857-9.

Background

Musculoskeletal disorders (MSDs) include a wide range of conditions that compromise the structural integrity and functional capacity of muscles, bones, joints, tendons, ligaments and associated connective tissue [54]. According to the World Health Organization (WHO, 2021), these conditions are characterized by persistent pain and functional limitations in mobility and manual dexterity, reducing individuals’ ability to work and engage in social activities. Among non-malignant pain conditions, symptoms arising from damage to the musculoskeletal system are among the most prevalent worldwide.

Musculoskeletal pain constitutes one of the most common and disabling public health challenges worldwide. Estimates from the Global Burden of Disease Study indicate that MSDs, particularly low back and neck pain, rank among the leading contributors to years lived with disability (YLDs) in all world regions [3, 25, 55]. Current data suggest that more than 1.7 billion people are affected globally, with low back pain alone impacting approximately 7.5% of the world’s population over 600 million individuals in 2020 (WHO, 2021). The increasing burden of MSDs is driven by population ageing, sedentary lifestyles and evolving occupational demands [7]. Beyond their clinical implications, MSDs are a considerable socioeconomic burden and one of the primary causes of work absenteeism, functional impairment and increased health service utilization [24].

In occupational settings, musculoskeletal pain is a major public health concern and is a leading cause of sick leave and disability [51]. Work-related musculoskeletal disorders (WRMSDs) due to occupational physical exposure include discomfort, functional limitation, disability or persistent pain affecting the musculoskeletal system [1, 15]. Numerous occupational groups are at an elevated risk of WRMSDs [20]. Nurses are most affected due to the physical, postural and biomechanical demands inherent in clinical practice [14, 49]. Evidence suggests that musculoskeletal disorders are highly prevalent among nurses, affecting multiple body regions. Previous literature has identified multiple occupational risk factors associated with WRMSDs in nurses, including patient handling, repetitive movements, prolonged standing, awkward postures, high physical workload, and psychosocial stressors [6, 14, 48]. Shift work may also contribute to musculoskeletal symptoms through mechanisms related to fatigue, circadian disruption, sleep impairment, and reduced recovery time.

Previous systematic reviews have reported that between 60% and 80% of nurses experience low back pain, while pain in the knees, shoulders, hips, or neck affects approximately 30–50% of this workforce [15, 16, 45]. More recent evidence confirms that the burden of WRMSDs remains substantial and widely distributed across anatomical regions in nursing populations [21, 56]. Recent meta-analytic evidence also suggests substantial geographic variability in WRMSD prevalence among nurses across continents and healthcare systems, although consistently high prevalence rates have been reported worldwide [21].

Given that WRMSDs contribute to nurse absenteeism and may exacerbate the projected global shortage of nursing professionals, which is expected to reach 4.1 million by 2030 (WHO, 2020), it is essential to identify aggravating occupational factors and develop targeted preventive strategies. Shift work has been proposed as a potential contributing factor to musculoskeletal symptoms in nurses. Several studies have suggested a possible association between rotating or irregular shifts and musculoskeletal pain affecting multiple body regions [33, 56]. A previous meta-analysis by Chang and Peng [11] reported an increased prevalence of work-related musculoskeletal disorders among nurses working rotating and irregular shifts compared with fixed day shifts, with a pooled odds ratio (OR) of 1.40 for low back pain. However, important methodological and conceptual aspects merit further consideration. Previous reviews frequently combined heterogeneous healthcare populations, limiting the specificity of findings for nursing professionals. In addition, rotating schedules and fixed night shifts were commonly grouped into a single exposure category despite representing potentially distinct occupational and circadian conditions. Furthermore, prior evidence focused predominantly on isolated anatomical regions, particularly low back pain, with limited consideration of generalized or multi-site musculoskeletal burden. Consequently, the extent to which different nursing shift schedules are associated with broader patterns of musculoskeletal symptoms remains unclear.

Therefore, an updated synthesis specifically focused on nurses and based on a more differentiated characterization of shift schedules and musculoskeletal outcomes is needed. The present study aims to: (1) restrict the analysis to predominantly nursing populations, (2) distinguish between day, rotating, and fixed night shifts as separate occupational exposures, and (3) examine both generalized and region-specific musculoskeletal outcomes. By addressing these methodological limitations, this study seeks to provide a more clinically and occupationally relevant understanding of the association between shift work and musculoskeletal health in nurses.

The aim of this study is to examine, through a systematic review and meta-analysis of observational studies, the association between different nursing shift patterns and generalized or region-specific musculoskeletal disorders among nurses.

Methods

Design

This study is a systematic review and meta-analysis examining the association between shift work and musculoskeletal disorders among nurses. The review was registered prospectively in PROSPERO (CRD420250656652). Methodology followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [41] and adhered to the Meta-analysis and Systematic Reviews of Observational Studies (MOOSE) recommendations [50].

Deviations from the registered protocol were documented and are reported transparently in this manuscript. Although the protocol initially specified a predefined set of musculoskeletal outcomes, during the study selection and data extraction phases a broader range of outcomes was identified across the included studies. To avoid selective reporting and to ensure a comprehensive synthesis, all relevant musculoskeletal outcomes consistently reported in the literature were included in the analysis.

Selection criteria

Selection criteria were based on predefined criteria using PECOS (Population, Exposure, Comparison, Outcome, Study Design) [37] (supplementary appendix 1). Based on the population, main inclusion criteria were studies with samples of more than 50% practicing nurses. All studies with samples that included other healthcare professionals at a higher percentage than nurses or those focusing on nursing students were excluded. No restrictions regarding age or sex were applied. Studies that analyzed nursing shifts and classified them into day or night shifts were included, for example: fixed day or night shifts, morning and afternoon shifts, or rotating shifts that included night shifts. All studies that analyzed a single nursing shift showed data exclusively on hours worked, or that focused on the COVID-19 period were excluded. Outcome studies that analyzed musculoskeletal problems in nurses, such as chronic pain, lower back pain, pain in the upper or lower limbs, muscle strains, or multisite musculoskeletal symptoms, were included. Studies that focused on mental fatigue or did not distinguish between mental and physical fatigue were excluded.

To be included, studies had to compare or analyze musculoskeletal problems between different shifts. Observational and descriptive studies, cross-sectional studies, case-control studies, or cohort studies both in Spanish and English were included. Review, intervention and protocol studies were excluded with no restrictions applied regarding the country of origin or publication date. No publication date limit was applied.

Search strategy

The literature review was conducted between December 2024 and August 2025 by consulting the databases PubMed, Web of Science, Scopus and CINAHL. The key words used for developing the research strategy were related to three areas: nursing, shift work schedule and musculoskeletal disorders. MeSH terms and their combination with Boolean operators AND and OR are shown in Appendix 2. We used the Rayyan.ai application for data management. Rayyan.ai enabled the researchers to import articles from the databases, identify and discard duplicates, assign reviewers for independent and blinded screening, select articles, monitor the percentage of discrepancies, and access and review full-text documents [44].

Data extraction

After discarding duplicate studies, two independent researchers (CRZ and NMG) read the title and abstract, later applying the inclusion and exclusion criteria. A third researcher then solved discrepancies to select studies for full-text reading. The first two researchers read the full texts, and further discrepancies were solved by a third author (JBC). Final studies were included by consensus.

Three authors (CRZ, NMG and JBC) then analyzed the general characteristics, results and methodological quality of these studies and extracted the relevant information. Year of publication, lead author, sample size, participant characteristics, study design, musculoskeletal disorder, affected body region, work shift type, assessment tools, follow-up duration were extracted (see Appendix 3). This data, previously agreed on by the three reviewers, was registered on a specially designed Microsoft Excel spreadsheet using a predefined template.

The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analyses was applied for cohort studies [57] and The Joanna Briggs Institute (JBI, 2017) critical appraisal checklist for analytical and cross-sectional studies was used. (see Appendix 4 for data transfer by authors NMG and CRZ).

Statistical analysis

Shifts examined included day shifts (e.g., 7:00 AM–3:00 PM and extended daytime schedules such as 7:00 AM–7:00 PM), rotating shifts involving morning–afternoon–night rotations (7:00 AM–3:00 PM, 3:00 PM–11:00 PM, 11:00 PM–7:00 AM, and 7:00 PM–7:00 AM), and fixed night shifts (7:00 PM–7:00 AM or 11:00 PM–7:00 AM).

Two independent meta-analyses were conducted to compare day versus rotating shifts and day versus fixed night shifts using random-effects models. Only outcomes with sufficient data were included in the analyses.

Although the systematic review included a broader range of musculoskeletal outcomes, only musculoskeletal pain outcomes were eligible for quantitative synthesis, as these were the only outcomes reported with sufficient consistency across studies.

Musculoskeletal pain outcomes were categorized according to anatomical region (neck, back/lumbar, shoulder, and lower limbs). Studies reporting pain affecting more than one anatomical region, aggregated musculoskeletal pain outcomes, or the number of painful body sites were classified as generalized or multi-site musculoskeletal pain for analytical purposes.

Included studies reported different outcome assessment periods, including point prevalence, 3-month prevalence, and 12-month prevalence, as well as different effect measures, including odds ratios (ORs), risk ratios (RRs), and hazard ratios (HRs). To enable quantitative synthesis, reported estimates were transformed to a common logarithmic scale and pooled as relative measures of association whenever sufficient information was available. Given the variability in temporal definitions, study designs, and outcome assessment periods across studies, pooled estimates were interpreted as general measures of association rather than as specific epidemiological parameters. Forest plots were generated to illustrate the effect estimates were reported with 95% confidence intervals (95% CIs).

Random-effects models were applied to account for between-study variability. Heterogeneity was assessed using the I² statistic. Due to methodological heterogeneity across observational studies, pooled estimates should be interpreted cautiously.

Values greater than 1 indicated a higher likelihood of musculoskeletal pain in the exposed shift group, whereas values below 1 indicated a lower likelihood. All analyses were performed using Meta Analysis Online Software (Fekete & Győrffy, 2025).

Risk of bias assessment

The risk of bias in each study was independently assessed by two authors using the ROBINS-E tool for non-randomized studies of exposure [23]. The ROBINS-E assessment was used exclusively to evaluate methodological quality and risk of bias and was not applied as a criterion for study exclusion. This tool evaluates risk of bias across seven domains relevant to observational exposure studies, including bias due to confounding, exposure measurement, participant selection, post-exposure interventions, missing data, outcome measurement, and selective reporting. Studies were classified as low, moderate, or high risk of bias according to the criteria established by the tool. For this review, studies deemed to have low or moderate risk of bias were considered to provide more reliable evidence, whereas studies classified as high risk of bias were interpreted with caution due to their greater susceptibility to confounding and methodological limitations.

Formal sensitivity analyses based on risk-of-bias categories were not performed because the number of studies included within each anatomical subgroup was limited and the variability in ROBINS-E classifications across studies was low, which could have resulted in unstable and difficult-to-interpret estimates.

Results

Figure 1 shows the flow chart corresponding to the selection of studies included in the systematic review. From the search of databases and registers, 1.685 references were identified (PubMed = 224, Scopus = 617, Web of Science = 690 and CINAHL = 154) and 1 additional register. After removing 760 duplicates, two researchers (CRZ and NMG) examined 925 studies by reading their titles and abstracts, of which 845 were excluded for not meeting the inclusion criteria. Consensus on articles to be included was reached based on eligibility criteria.

Fig. 1.

Flow chart diagram

Eighty-one full reports were selected for retrieval, although one could not be obtained. Finally, 80 full reports were evaluated. 56 were excluded for the following reasons: less than 50% of the nursing population in the study sample (n = 5), studies in languages other than English or Spanish (n = 4), studies where work shifts were not considered (n = 12), where musculoskeletal disorders were not addressed (n = 18), those where associations between shifts and musculoskeletal disorders were not analyzed (n = 15) and review articles or RCT (n = 2). See Appendix 5 for reasons why the 56 studies were excluded.

Finally, 24 studies met inclusion criteria and were incorporated into the systematic review. However, only 13 studies were eligible for inclusion in the two meta-analyses because essential quantitative data required to standardize the reported measures of association were insufficiently reported in the original publications, and additional information could not be obtained.

Study characteristics

The twenty-four studies analyzed included 102.159 participants (see Table 1). The sample size ranged from 55 participants [39] to 51.406 participants [63]. Data collection spanned over three decades from the late 1980s to 2020. The oldest field data study was that of Feskanich et al. [19], with records from 1988 up until 2000. The most recent were those of Schultz et al. [46], with data collected between 2019 and 2020, and Zhang et al. [63], covering the period from 2016 to 2017. The highest concentration of research was between 2000 and 2020, a period that accounts for more than 80% of the studies included.

Table 1.

Characteristics of included studies

First author (year)	Country	Study design	Sample size (n)	Title
Guan et al. [22]	China	Cross-sectional	2.458	Occupational factors causing pain among nurses in mainland China
Kędzierska et al. [31]	Poland	Cross-sectional	200	The influence of shift work on health behaviors and self-reported health of nurses
Ali et al. [4]	Pakistan	Cross-sectional	154	Impact of morning and rotational duties on physical health of nurses working in tertiary care hospitals of Karachi
Schultz et al. [46]	Brazil	Cross-sectional	83	Factors related to musculoskeletal pain of nurses in hospital settings
Unruh and Asi, [53]	USA	Cross-sectional	414	Determinants of workplace injuries and violence in new RNs
Nicoletti et al. [39]	Switzerland	Cohort	55	Physical workload, trapezius activity, and neck pain in nurses’ day and night shifts
Zhao et al. [65]	Australia, New Zealand, UK	Cohort	928	The Effects of Shift Work and Interaction Between Shift Work and Overweight/Obesity on Low Back Pain in Nurses Results from a Longitudinal Study
Quevedo and Zúñiga, [43]	Chile	Cross-sectional	70	Low bone mineral density in rotating-shift workers
Feskanich et al. [19]	USA	Cohort	38.062	Nightshift work and fracture risk: The Nurses’ Health Study
Lipscomb et al. [35]	USA	Cross-sectional	1.163	Work-schedule characteristics and musculoskeletal disorders of registered nurses
Zhang et al. [63]	China	Cross-sectional	51.406	Career development and occupational disease in nurses: A cross-sectional study
Katsifaraki et al. [30]	Norway	Cross-sectional	1.829	Pain complaints after consecutive nights and quick returns in Norwegian nurses
Ovayolu et al. [40]	Turkey	Cross-sectional	114	Frequency and severity of low back pain in intensive care nurses
Farzaneh et al. [17]	Iran	Cross-sectional	80	The survey of job injuries and mental health disorders among clinical nurses from ergonomics aspect
Moscato et al. [38]	Italy	Cross-sectional	185	Musculoskeletal injuries among operating room nurses: A multicenter survey
Sveinsdóttir, [52]	Iceland	Cross-sectional	348	Self-assessed quality of sleep, occupational health, working environment, illness experience and job satisfaction of female nurses working different combination of shifts
Matre et al. [36]	Norway	Cross-sectional	1.558	Pain complaints associated with quick returns and insomnia among Norwegian nurses
Katsifaraki et al. [29]	Norway	Cohort	679	Sleep duration mediates abdominal and lower-extremity pain after night work in nurses
Yongcheng et al. [60]	China	Cross-sectional	692	The associations of work style and physical exercise with risk of MSDs in nurses
Abedini et al. [2]	Iran	Cross-sectional	400	Patient manual handling risk assessment among hospital nurses
Cameron et al. [10]	Canada	Cross-sectional	303	Musculoskeletal problems experienced by older nurses in hospital settings
Passali et al. [42]	Greece	Cross-sectional	394	Work-related musculoskeletal disorders in Greek nursing professionals
Asghari et al. [5]	Iran	Cross-sectional	144	Musculoskeletal pain in operating room nurses: Associations with quality of work life, working posture, socio-demographic and job characteristics
Buja et al. [9]	Italy	Cross-sectional	440	Strain and health implications of nurses’ shift work

Selected studies were from Poland [31], Switzerland [39], Norway ([29, 30, 36], Italy [9, 38], Greece [42], Iceland [52], and the United Kingdom [65]. Eight Middle Eastern and four Asian studies were carried out, three in China [22, 60, 63], three in Iran [2, 5, 17], one in Pakistan [4], and one in Turkey [40]. Four North American studies were located in the United States [19, 35, 53] and Canada [10] and two in South America, Brazil [46] and Chile [43]. One Oceanic study, conducted in Australia, New Zealand, and the United Kingdom [65] was also selected.

Half of the reviewed studies reported a numerical mean age, with an overall mean age of 41.1 years. The lowest mean age was observed in the study by Abedini et al. [2] (30.8 years) and the highest in Ovayolu et al. [40] (56 years). Most studies were multicenter (n = 21; 87.5%) and cross-sectional in design. The most frequently evaluated variables were related to musculoskeletal pain, particularly neck pain, back pain, and shoulder pain. The Nordic Musculoskeletal Questionnaire (NMQ) was the most widely used measurement tool (six studies), followed by ad hoc questionnaires and other validated scales such as the Visual Analogue Scale (VAS) and the Short Form Health Survey (SF-36). Fewer studies examined musculoskeletal outcomes other than pain, including muscle strains, low bone mineral density, osteoporosis, hip and wrist fractures, and broader work-related musculoskeletal disorders [19, 43, 53, 60].

More than half of the studies (n = 13) found significant differences between different work shifts, generally to the detriment of night or rotating shifts [2, 10, 19, 22, 30, 31, 35, 43, 46, 52, 60, 63, 65]). Findings, however, were not uniform. While half of the studies identified positive associations between shift type and the presence of pain or physical discomfort, a similar number reported null, negative, or mixed results ([4, 5, 9, 17, 36, 38, 39, 40, 53]), reflecting high variability in the effects of shift work on the musculoskeletal health of nursing staff (see Appendix 3).

Quality of studies

The methodological quality of the studies was measured using the NOS scale for cohort studies and the JBI critical appraisal checklist for cross-sectional studies (See Appendix 4). Studies included are of acceptable methodological quality with the longitudinal designs proving more robust than the cross-sectional designs.

The cohort studies are of moderate to high methodological quality, with scores between 6 and 9 out of a total of 9 on the NOS. The study by Nicoletti et al. [39] scored 6 points, presenting limitations in the representativeness of the cohort and the description of the unexposed group. Feskanich et al. [19] scored highest (9/9), standing out for its high population representativeness, exhaustive control of confounding factors, and adequate longitudinal follow-up characteristics to reinforce the validity of findings. The studies by Zhao et al. [65] and Katsifaraki et al. [29, 30]) scored between 8 and 9, demonstrating adequate control of variables such as age, Body Mass Index (BMI), smoking, sleep, and psychosocial factors, along with sufficient follow-up for the observation of outcomes. Some of them relied on self-reports for the evaluation of results however, which may have caused information bias.

In the cross-sectional studies, evaluated using the JBI checklist, methodological quality was more heterogeneous. Most, such as Guan et al.[22], Kędzierska et al. [31], Schultz et al. [46], and Lipscomb et al. [35], clearly describe the inclusion criteria, study setting and applied valid and reliable measurement methods. Several studies used multivariate statistical models. Some studies had shortcomings in the identification and management of confounding factors, such as Passali et al. [42]. This could affect the causal interpretation of the observed associations. Other studies, such as Ali et al. [4], Moscato et al. [38], and Asghari et al. [5], stood out for their adequate validation of measurement instruments, strengthening the reliability of their results.

Risk of bias assessment of included studies

ROBINS-E tool for non-randomized studies of exposure was used to assess the risk of bias in all 24 studies included in the systematic review and meta-analysis (see Fig. 2). Thirteen (54.2%) of the twenty-four studies had a low risk of bias. Three studies had the lowest level of risk of bias because all their assessments were positive in all domains [30, 35, 63]. One study (Unruh et al. 2018) had a high risk of bias due to problems in the study design. The main risk of bias in the studies is related to bias in the selection of reported outcomes, which is a problem in 66.7% of the studies. This is followed by bias arising from outcome measurement, which occurs in 58.3% of studies, and bias due to confounding factors, occurring in 50% of studies.

Fig. 2.

Risk of bias

Effects of nursing shifts on musculoskeletal pain

Two independent meta-analyses were carried out using a random-effects model, comparing day shifts to rotating shifts, day shifts to fixed night shifts. These analyses included only those body regions for which sufficient data were available to determine the shift associated with higher likelihood of musculoskeletal pain. Outcomes included neck pain, back/lumbar pain, shoulder pain, lower limb pain, and generalized or multi-site musculoskeletal pain.

Results of musculoskeletal symptoms observed in daytime vs. rotating shifts

The first meta-analysis comparing day and rotating shifts included seven studies (Fig. 3). The overall pooled estimate showed no statistically significant association between rotating shifts and musculoskeletal pain outcomes compared with day shifts (effect size = 1.15; 95% CI: 0.98–1.35), with low between-study heterogeneity (I² = 7.2%).

Fig. 3.

Analysis of pain observed in daytime vs. rotating shifts. Note. Values greater than 1 indicate a higher likelihood of musculoskeletal pain in rotating shifts compared with day shifts

For musculoskeletal pain in multiple locations, this was the area with the largest effect size (1.41; 95% CI: 0.76–2.60), although the wide confidence intervals indicated considerable uncertainty surrounding the estimated effect.

Stratified analyses by anatomical region likewise showed no statistically significant associations. Back/lumbar pain showed a pooled effect size of 1.07 (95% CI: 0.71–1.61), whereas shoulder pain yielded the highest regional estimate (effect size = 1.33; 95% CI: 0.98–1.80). Neck pain showed a pooled effect size of 1.19 (95% CI: 0.85–1.66), while lower limb pain was associated with a pooled effect size of 0.88 (95% CI: 0.62–1.26).

Results of musculoskeletal symptoms observed in daytime vs. fixed night shifts

The second meta-analysis comparing day and fixed night shifts included six studies (Fig. 4). The overall pooled estimate comparing fixed night shifts with day shifts showed no statistically significant association with musculoskeletal pain outcomes (effect size = 1.09; 95% CI: 0.94–1.27), although substantial between-study heterogeneity was observed (I² = 81.1%).

Fig. 4.

Analysis of pain observed in Daytime vs. Fixed Night Shifts. Note. Values greater than 1 indicate a higher likelihood of musculoskeletal pain in rotating shifts compared with day shifts

Although no statistical significance was found, musculoskeletal pain in multiple locations again had the largest combined effect size (1.14; 95% CI: 0.83–1.55).

Stratified analyses by anatomical region likewise revealed no statistically significant associations. Back/lumbar pain showed a pooled effect size of 1.26 (95% CI: 0.65–2.43), while shoulder pain yielded a pooled effect size of 1.20 (95% CI: 0.58–2.45). Lower limb pain was associated with a pooled effect size of 1.06 (95% CI: 0.50–2.23), whereas neck pain showed an estimate below unity (effect size = 0.94; 95% CI: 0.83–1.07).

Discussion

This systematic review suggests possible shift work may be associated with different musculoskeletal outcomes among nurses; however, the quantitative meta-analysis, restricted to musculoskeletal pain outcomes, did not demonstrate a statistically significant association between rotating or fixed night shifts and musculoskeletal pain. Nevertheless, several pooled estimates tended to favor higher musculoskeletal pain prevalence in rotating or fixed night shifts, particularly for shoulder, neck, and generalized or multi-site musculoskeletal pain, although confidence intervals were wide and crossed the null value.

Despite the absence of statistically significant associations, several methodological and clinical aspects warrant further consideration. The present study specifically examined the relationship between nursing shift patterns and WRMSDs; however, this relationship is likely multifactorial and may interact with ergonomic, psychosocial, organizational, and sleep-related factors [6, 34, 61]. In addition to physical workload and repetitive biomechanical exposure, shift work may also contribute indirectly to musculoskeletal symptoms through mechanisms related to fatigue, circadian disruption, impaired sleep quality, and reduced recovery time between shifts [26]. Additionally, several occupational hazards, including awkward postures, repetitive movements, patient handling, and shift work itself may contribute simultaneously to musculoskeletal symptoms among nurses [32].

Previous evidence suggests that the effects of shift work in nurses may extend beyond musculoskeletal outcomes. Systematic reviews and meta-analyses have consistently reported poorer sleep quality, sleep disturbances, and circadian disruption among nurses working rotating or night schedules [12, 13, 62]. In parallel, impaired sleep quality and circadian misalignment have been associated with musculoskeletal pain, fatigue, and reduced recovery capacity, which may partially contribute to the development or persistence of WRMSDs in nursing populations ([26, 30, 64]). Several studies have suggested that biomechanical mechanisms underlying WRMSDs may be more strongly influenced by ergonomic workplace factors than by shift organization alone [14, 28, 47].

Earlier cross-sectional studies had already suggested that rotating and irregular schedules may be associated with a higher frequency of musculoskeletal complaints among nurses [22, 31, 46, 60], although both the magnitude and anatomical distribution of symptoms varied considerably across studies [11, 15].

A previous meta-analysis by Chang and Peng [11] reported a significant association between rotating or irregular shifts and work-related musculoskeletal disorders, particularly for low back pain. However, rotating and irregular schedules were analyzed as a single combined exposure, limiting the ability to distinguish between potentially different occupational and circadian characteristics associated with specific shift patterns. In contrast, the present study differentiated between day, rotating, and fixed night shifts, allowing a more detailed assessment of how distinct work schedules may relate to generalized and region-specific musculoskeletal outcomes among nurses.

Interestingly, the more differentiated approach adopted in the present review may also help explain why the observed associations were less consistent than those reported previously. When rotating and fixed night schedules were analysed separately, no statistically significant associations were identified for most anatomical regions or generalized musculoskeletal outcomes. This finding suggests that the relationship between shift work and WRMSDs may be more complex and less uniform than suggested by analyses grouping heterogeneous shift patterns into a single exposure category, particularly when considering the multifactorial interaction between physical workload, psychosocial stressors, and organizational factors previously described in nursing populations [48].

In addition, substantial variability was identified in the classification and organization of shift schedules across studies. Differences in shift duration, frequency of night work, quick returns, overtime exposure, and rotation systems may contribute to the heterogeneity observed between studies and complicate direct comparisons across occupational settings [8, 56]. Furthermore, previous international meta-analyses have also highlighted considerable heterogeneity in WRMSD prevalence according to geographical, organizational, and ergonomic contexts among nursing populations [21]. Such variability highlights the need for more standardized definitions of shift work exposure in future observational research.

Several limitations should be considered when interpreting the findings of this review. First, the included nursing populations encompassed a wide range of professional roles, clinical settings, and work demands, potentially contributing to variability in biomechanical exposure and musculoskeletal outcomes. Second, substantial methodological heterogeneity existed across studies, including differences in musculoskeletal outcome definitions, assessment instruments, and recall periods, which may limit comparability between studies and influence the pooled estimates.

Third, most included studies were cross-sectional and relied on self-reported measures, increasing the possibility of recall and information bias and limiting causal inference. In addition, not all studies adequately controlled for relevant confounding factors such as age, workload, or ergonomic exposure.

Fourth, several subgroup meta-analyses included a limited number of studies and showed moderate to high heterogeneity, which may have reduced the stability of pooled estimates and contributed to the wide confidence intervals observed in some analyses. In addition, 11 of the 24 studies included in the systematic review could not be incorporated into the meta-analyses because the original publications did not report sufficient quantitative information to allow harmonization of the reported measures of association. This reduced the comprehensiveness of the quantitative synthesis and may have limited the precision and representativeness of the pooled estimates. Although risk of bias was assessed using the ROBINS-E tool, studies were not excluded based on risk-of-bias ratings, and formal sensitivity analyses stratified according to risk-of-bias categories were not performed due to the limited number of studies within anatomical subgroups and the low variability in risk-of-bias ratings across studies. Consequently, the robustness of pooled estimates according to study quality could not be formally explored through sensitivity analyses.

Finally, the generalizability of the findings may be limited by differences in healthcare systems, staffing conditions, and organizational practices across countries and clinical contexts. Furthermore, restricting the review to nursing populations improved occupational specificity but may have reduced the number of eligible studies available for quantitative synthesis.

Although no statistically significant associations were identified, the high prevalence and multifactorial nature of WRMSDs among nurses support the importance of occupational health strategies aimed at improving ergonomic conditions, recovery opportunities, sleep hygiene, and workload management across different shift schedules. Preventive approaches incorporating physical activity promotion, ergonomic interventions, and sleep-health education may help reduce fatigue accumulation and improve recovery between shifts [64]. Future research should prioritize prospective cohort designs, standardized definitions of shift work exposure, and more homogeneous musculoskeletal outcome measures. Further investigation into the physiological, psychosocial, ergonomic, and organizational mechanisms potentially linking shift work and WRMSDs would contribute to a more comprehensive understanding of musculoskeletal health among nurses.

Conclusion

This systematic review and meta-analysis examined the association between shift work and musculoskeletal pain across different anatomical regions among nurses. Although several pooled estimates tended to be higher among rotating shift workers, no statistically significant associations were identified.

Differences in exposure definitions, musculoskeletal outcome assessment methods, and study designs limited comparability between studies and require cautious interpretation of the pooled estimates. In addition, persistent or chronic pain outcomes and generalized musculoskeletal pain were not consistently defined across studies, limiting the interpretation of these specific outcomes.

Furthermore, potential mechanisms linking shift work and musculoskeletal pain, including circadian disruption, sleep disturbances, fatigue, and biomechanical load, were not directly assessed in most included studies.

Overall, the available evidence suggests that the relationship between shift work and musculoskeletal pain among nurses is complex and potentially influenced by multiple occupational and organizational factors. Future research should prioritize larger longitudinal studies using more standardized definitions of shift work exposure and musculoskeletal outcomes.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

MSDs

Musculoskeletal disorders

WRMSDs

Work-related musculoskeletal disorders

JBI

Joanna Briggs Institute

OR

Odds ratio

HRs

Hazard Ratios

WHO

World Health Organization

YLDs:

Years lived with disability

BMI

Body Mass Index

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

MOOSE

Meta-analysis and Systematic Reviews of Observational Studies

PECOS

Population, Exposure, Comparison, Outcome, Study Design

NOS

Newcastle-Ottawa Scale

RR

Relative Risk

Author contributions

Cayetana Ruiz-Zaldibar: Writing – review C editing, Writing – original draft, Methodology, Formal analysis, Data curation, Conceptualization. Noemí Mayoral- Gonzalo: Review C editing, Supervision, Methodology, Data curation, Conceptualization. David Pérez-Manchón, Gema Mata-González and Raquel Moreno-Sánchez: Writing – review C editing, Supervision, Conceptualization. Jaime Barrio-Cortés: Writing – review C editing, Methodology, Formal analysis, Data curation. Víctor Jiménez Díaz-Benito: Writing – review C editing, Supervision, Methodology, Formal analysis, Data curation, Conceptualization.

Funding

This study has received no financial support.

Data availability

The datasets generated and/or analysed during the current study are available from the corresponding author.

Declarations

Declaration of Generative AI and AI-assisted technologies in the writing process

The authors declare the use of the Rayyan.ai platform for managing the systematic review, as well as the use of ChatGPT to refine and clarify concepts related to the language in which the article is written

Ethics approval

Not applicable.

Consent for publication

Not applicable.

CRediT authorship contribution statement

Cayetana Ruiz-Zaldibar: Writing – review & editing, Writing – original draft, Methodology, Formal analysis, Data curation, Conceptualization. Noemí Mayoral- Gonzalo: Review & editing, Supervision, Methodology, Data curation, Conceptualization. David Pérez-Manchón, Gema Mata-González and Raquel Moreno-Sánchez: Writing – review & editing, Supervision, Conceptualization. Jaime Barrio-Cortés: Writing – review & editing, Methodology, Formal analysis, Data curation. Víctor Jiménez Díaz-Benito: Writing – review & editing, Supervision, Methodology, Formal analysis, Data curation, Conceptualization.

Competing interests

The authors declare no competing interests.