Continental Umbrella Review and Meta-Analysis of Work-Related Musculoskeletal Disorders Prevalence Among Healthcare Professionals

Abstract

Work-related musculoskeletal disorders (WMSDs) have a significant impact on healthcare professionals. The aim of this study was to conduct an umbrella review and meta-analysis to examine the overall body area prevalence of WMSDs by continents, according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Mendeley, PubMed/Medline, Science.gov, ScienceDirect, and Google Scholar were screened without date limitation to identify relevant meta-analyses. The selection, quality appraisal, and data extraction process were performed independently by two reviewers. Ten meta-analyses were included from the 3853 unique records, for a total of 100,211 participants, including dentists, nurses, surgeons, and mixed healthcare professionals. High heterogeneity (Cochran’s Q test and I² statistic) was observed. The largest number of meta-analyses was performed among nurses. Subgroup analysis by continent revealed an imbalance in the number of works, with Asia being the most documented. The analysis of prevalence rates was complete in Asia (overall and nine body areas), and partial in Europe (neck, shoulder, wrist) and Africa (lower back only). A ranking of the most exposed areas by continent was proposed. The lower back was the most exposed area (Africa: 54.5%; Asia: 56.6%). It would be relevant in future work to consider the numerous cross-sectional studies in order to improve subgroup analyses by continent and, thus, complete and strengthen the initial results presented in this first umbrella review.

1. Introduction

Professional activities affect the health and quality of life of workers, the origin of which may be physical, psychological, socio-economic, etc. [1,2]. Among the physical risk factors reported, awkward postures, repeated movements, carrying heavy loads, and inadequate workplace ergonomics are the most frequently mentioned. Repeated exposure to these risk factors increases the chances of developing musculoskeletal disorders [3,4]. Stiffness, pain, or changes in body structures like muscles, tendons, ligaments, joints, peripheral nerves, and supporting blood vessels are the main musculoskeletal symptoms [5]. They lead to various pathologies such as osteoarthrosis, degenerative spine disease, carpal tunnel syndrome, tendinitis, and other chronic pain, such as cervical or low back pain [6,7]. The Global Burden of Disease 2019 counted nearly 1.71 billion people affected by musculoskeletal disorders and diseases [8]. When musculoskeletal disorders originate or are aggravated in the context of occupational activities, they are called work-related musculoskeletal disorders (WMSDs). They lead to absenteeism at work, which has significant direct and indirect financial consequences for different countries [9,10].

Healthcare professionals are highly impacted by WMSD [11]. This issue was addressed in cross-sectional studies to assess the overall prevalence and by body area of WMSDs. Numerous works have reported an average overall prevalence between 60 and 90% for all professionals: nurses [12], physiotherapists [13], midwives [14], dentists [15], surgeons [16], and osteopaths [17]. The use of questionnaires allowed for a deeperevaluation by quantifying the WMSD prevalence for each region of the body, classically divided into nine areas: neck, upper back, lower back, shoulder, elbow, wrist, hip, knee, and ankle. The two areas most exposed to WMSD among all health professionals are the lower back, neck, and shoulder [18,19,20,21], with a prevalence above 40%.

In order to synthesize the numerous published surveys, authors have proposed systematic reviews to provide an overview for healthcare professionals [22,23,24,25] or a mixed sample [11]. To complement the qualitative analysis of systematic reviews, authors have carried out meta-analyses to provide a quantitative synthesis of the overall prevalence and by body area. The data have been pooled for a country [26,27], a continent [28,29,30,31,32], or worldwide [33,34,35,36,37].

Several factors, such as working conditions or the socio-economic environment, often linked to a country or continent, can influence the prevalence of WMSDs. As reported, worldwide data remain limited to one healthcare profession or one geographical area. To our knowledge, there is no synthesis that investigates the effect of the geographical area on the WMSD prevalence among all healthcare professionals. Such a study could contribute to a better understanding of the general impact of WMSDs to implement ergonomic program policies for the prevention of WMSDs.

Thus, we aim to propose an umbrella review and meta-analysis to quantify the worldwide WMSD prevalence among healthcare professionals. The associated research question was as follows: “Is there an effect of continent on the prevalence of WMSDs overall and by body area among healthcare professionals?”. The analysis focused on the nine body areas and the overall prevalence.

2. Materials and Methods

2.1. Search Strategy

This umbrella review protocol was registered on PROSPERO with the number CRD420251052696, and no ethical approval was required. This search was conducted between 5 and 11April 2025in five free databases, i.e., Mendeley, PubMed/Medline, Science.gov, ScienceDirect, and Google Scholar, looking for meta-analyses that investigated the prevalence of WMSD among healthcare professionals. The search strategy for each database, i.e.,the keyword combination and filters applied, is presented in

Table 1. Search strategy for each database.

Database	Search Keyword Combinations	Applied Filters
Mendeley	musculoskeletal disorder AND (healthcare professional OR surgeon OR nurse OR physiotherapist OR dentist OR midwife OR physical therapist) AND meta-analysis	None
PubMed/MedLine	musculoskeletal disorder AND (healthcare professional OR surgeon OR nurse OR physiotherapist OR dentist OR midwife OR physical therapist) AND meta-analysis	Article type: Meta-Analysis Article language: English
Science.gov	musculoskeletal disorder AND (healthcare professional OR surgeon OR nurse OR physiotherapist OR dentist OR midwife OR physical therapist) AND meta-analysis	Top results
ScienceDirect	musculoskeletal disorder AND (healthcare professional OR surgeon OR nurse OR physiotherapist OR dentist OR midwife OR physical therapist) AND meta-analysis	Article type: Review articles Languages: English
Google Scholar	“meta-analysis” prevalence “systematic review”“body area” healthcare OR professional OR surgeon OR nurse OR physiotherapist OR dentist OR midwife OR physical therapist “work-related musculoskeletal disorders”	None

. The articles corresponding to the search in each database were compiled in a synthesis table in Microsoft Excel 2020, and duplicates were identified and removed using the software function. Then, each single entry was evaluated on the basis of its title and abstract separately by two reviewers (P.G. and J.J.-B.). Articles that did not meet the inclusion/exclusion criteria were excluded. The list of remaining articles from each reviewer was compared to establish the list of articles to be sought for retrieval and full evaluation. Both reviewers evaluated separately the full-text to build the final list of meta-analyses to be included. Any discrepancies at each stage of the selection process were resolved by consensus and a further review of the articles.

2.2. Inclusion and Exclusion Criteria

The Population, Exposure, Comparator, and Outcomes (PECO [38]) guidelines were used to define the inclusion criteria. For the present meta–meta-analysis, the population was all healthcare professionals; the exposure was 12-month WMSD prevalence; the data were compared between the five continents, and the outcomes were overall WMSD prevalence and by body area as a percentage. Nine body areas were studied in the assessment of WMSDs: neck, upper and lower back, shoulder, elbow, wrist, hip, knee, and ankle. WMSDs were considered to be present when symptoms such as pain and discomfort lasting at least one week or occurring at least once a month over the last 12 months were present [39]. As the aim was to conduct an umbrella review, only meta-analyses were included.

The following criteria led to the exclusion of potential meta-analyses: (1) the studies were not peer-reviewed meta-analyses, (2) the studies were not written in English, (3) the study population was not composed of healthcare professionals, (4) the sample was a mixed population with no possibility of distinguishing healthcare professionals from other professions, (5) the study results did not provide the overall and body-area prevalence of WMSDs as a percentage.

2.3. Quality Assessment

The PRISMA 2020 checklist (27 items) was applied to perform the quality appraisal of each meta-analysis [40]. A 3-point scale was used to fill each item: 0, 1, or 2 for an item not, partially, or completely fulfilled, respectively. The sum of all points gives a score out of 84 points (due to all the sub-items) converted into a percentage. The quality was defined in accordance with Hermanson and Choi’s classification [41]: a score of less than 50% (<42 points) means a high quality; a score of 80% or more (≥68 points) means a low quality. Otherwise (score between 42 and 67 points, 50–80%), the quality is medium. Two reviewers (P.G. and J.J.-B.) performed the quality assessment separately and compared their results. In the event of any discrepancies, they discussed (eventually with a third reviewer) and agreed on the final score.

2.4. Data Extraction

The extracted data were compiled into 3 parts. The first covered general information on each meta-analysis: name of first author, country, continent, year of publication, and number of studies included in the meta-analysis with their distribution by country and continent. The second part contains demographic data on healthcare professionals: sample size, healthcare profession, age, experience, and number of hours worked per week. The final section covers WMSD prevalence rates by body area: neck, upper back, lower back, shoulder, elbow, wrist, hip, knee, and ankle, as well as overall WMSD prevalence. For each prevalence reported as a percentage, the number of studies and the number of participants (if available) were indicated. Data were compiled in two summary tables. Empty boxes indicated missing information. All data in this umbrella review were reported following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines.

2.5. Overlap Analysis

Particular attention was paid to the overlap of primary studies between meta-analyses. The presence of a primary study several times in the same analysis overstates its sample size and falsely leads to greater precision in the analysis [42]. The corrected coverage area (CCA) is used to measure the degree of overlap between meta-analyses:

$$CCA= {\displaystyle \frac{N-r}{r\times c-r}}$$

where N corresponds to the total number of times primary publications appeared in reviews (inclusive of double-counting); r is the number of unique primary publications, and c is the number of meta-analyses compared.

Hennessy et al. [43] proposed a 5-step procedure for dealing with overlap: (1) creation of the 2-to-2 meta-analysis comparison matrix; (2) calculation of the N, c, and r parameters and then the CCA; (3) individual pairing review; (4) triangulation of data sources; (5) transparent reporting of the process and results. Pieper et al. [44] have proposed a classification to qualify the degree of overlap based on the CCA: 0–5% corresponds to a low overlap;6–10% corresponds to a moderate overlap; 11–15% corresponds to a high overlap, and >15% corresponds to a very high overlap. The Graphical Representation of Overlap for Overviews (GROOVE) [45] was used to display the results of overlap assessment.

If overlap is defined as very high, the approach used in the literature is to retain only one of the meta-analyses [42,43,46]. Selection is performedbasedon (1) quality (the meta-analysis with the highest quality is retained), (2) on the year of publication (the most recent is retained), and finally, on the number of primary studies (the one with the most is selected).

2.6. Data and Statistical Analysis

For each body area and overall prevalence, a meta-analysis with a forest plot was constructed to obtain pooled prevalence rates and 95% CIs among healthcare professionals for each continent using the method of Neyeloff et al. [47]. Cochran’s Q test and the I² statistic were used to quantify heterogeneity [48]. The significance level was p(Q) ≤ 0.1. The I² statistic was interpreted on a 4-level scale: low (0–40%), moderate (30–60%), substantial (50–90%), and high heterogeneity (75–100%) [48]. The model used to pool prevalence was selected from p(Q) and I². If p(Q) ≥ 0.1 and I² ≤ 50%, a fixed-effects model was applied (i.e., low heterogeneity). Otherwise, if p(Q) < 0.1 or I²> 50%, a random-effects model was selected (i.e., significant heterogeneity).

To answer the question, “Is there an effect of continent on the prevalence of WMSDs overall and by body area among healthcare professionals?”, subgroup analyses were performed. Healthcare professionals were grouped by continent according to their country of practice. Subgroup analysis was performed only if the number of meta-analyses was greater than or equal to 2.

3. Results

3.1. Database Search Results

A total of 3853 articles were identified by scanning the five databases, and the Excel function identified and removed 72 duplicates. Of the 3781 unique articles, only 32 were retained for full-text review. The remaining 3781 articles were excluded for the following reasons: lack of meta-analysis, non-healthcare population, and no reported prevalence. Twenty-five were removed after full review because prevalence was not presented by continent or not expressed as a percentage, the sample included multiple non-healthcare professions, or sample data for meta-analysis were not reported. To the remaining seven articles, three were added by citation searching and snowballing. Thus, the present umbrella review included 10 meta-analyses for a total of 100,211 healthcare professionals. The complete search process is illustrated by the flowchart in Figure 1.

3.2. Quality Appraisal

The quality of each meta-analysis was assessed using the PRISMA 2020 checklist. The mean score was 69.0 ± 6.3 points (min: 51 points; max: 80 points), or 82.1 ± 7.5% of items and sub-items completed. Six meta-analyses presented a high quality (≥68 points or 80%) [27,28,29,32,49,50] and four a medium quality (between 42 and 67 points or 50% to 80%) [26,31,51,52]. No meta-analysis presented a low quality. The detailed assessment is presented in Appendix A.

3.3. Study Characteristics

The meta-analysis by Gorce et al. [32] on surgeons presented the prevalence data of MSDs separately for America, Asia, and Europe. Thus, these data were considered as three different studies and arepresented on three separate lines in

Table 2. Demographic characteristics of the 18 included studies.

Authors	Country	Number of Studies	Sample Size	Healthcare Profession	Age of Participants	Context of this Study
						Continent	Number of Studies	Number of Countries
Africa
Aremu et al., 2023 [31]	Uganda	18	4486	Healthcare professionals	30–40	Africa	18	12
Kasa et al., 2020 [50]	Ethiopia	19	6110	Nurses	-	Africa	19	11
America
Gorce et al., 2023 [32]	France	17	4935	Surgeons	46.5 ± 4.2	America	17	2
Asia
Al Amer 2020 [49]	Saudi Arabia	18	5345	Healthcare professionals	23.7 ± 1.6	Asia	18	1
Gorce et al., 2023 [32]	France	4	628	Surgeons	39.9 ± 4.8	Asia	4	3
Jacquier-Bret et al., 2025 [29]	France	40	19,903	Nurses	31.3 ± 3.4	Asia	40	10
Saberipour et al., 2019 [51]	Iran	33	11,995	Nurses	-	Asia	33	1
Soroush et al., 2018 [26]	Iran	41	16,350	Nurses	33.09	Asia	41	1
Wang et al., 2024 [27]	China	23	21,042	Nurses	22–60	Asia	23	1
Zakerjafari et al., 2018 [52]	Iran	23	2531	Dentists	27.4–42.2	Asia	23	1
Europe
Gorce et al., 2023 [32]	France	12	1733	Surgeons	44.2 ± 5.9	Europe	12	9
Gorce et al., 2025 [28]	France	12	5153	Nurses	27.9–47.6	Europe	12	7

, which presents the descriptive information from each meta-analysis. Therefore, Table 2 contains 12 lines distributed by continent as follows: two studies in Africa [31,50], one study in America [32], seven studies in Asia [26,27,29,32,49,51,52], and two studies in Europe [28,32]. Each continent was represented by healthcare professionals from several countries: 14 countries in Africa, 2 countries in America, 10 countries in Asia, and 7 in Europe. The overall sample of 100,211 health professionals included 2531dentists (1 study), 80,553 nurses (six studies), 7296 surgeons (three studies), and 9831 mixed healthcare professionals (two studies).

A large variability was observed between meta-analyses for the number of cross-sectional studies included (4 to 40) and the number of participants (628 to 21,041). Although anthropometric data were often available in cross-sectional studies, mean values were rarely reported in meta-analyses, which did not allow for a synthesis in the present analysis. Ten meta-analyses reported the age of participants (22 to 60 years), with only five meta-analyses for which the mean value was computed. Height, weight, and body mass data were too rarely reported and are, therefore, not included in Table 2.

Table 3. WMSD prevalence by body area and overall prevalence for each included meta-analysis.

Authors		WMSD Prevalence by Body Area									WMSD Overall
		Neck	Upper Back	Lower Back	Shoulder	Elbow	Wrist	Hip	Knee	Ankle
Africa
Aremu et al., 2023 [31]	Prev. (%)	27.9%	22.7%	44.8%	23.7%	6.5%	20.9%	9.2%	18.1%	14.6%	-
	Nb studies	17—He	16—He	16—He	17—He	16—He	18—He	17—He	16—He	17—He	-
	Nb subjects	4283	3930	3930	4283	3930	4486	3930	3930	3930	-
Kasa et al., 2020 [50]	Prev. (%)	-	-	64.1%	-	-	-	-	-	-	-
	Nb studies	-	-	19—He	-	-	-	-	-	-	-
	Nb subjects	-	-	6110	-	-	-	-	-	-	-
America
Gorce et al., 2023 [32]	Prev. (%)	39.3%	-	-	36.0%	-	27.2%	-	-	-	-
	Nb studies	17—He	-	-	13—He	-	16—He	-	-	-	-
	Nb subjects	4935	-	-	3528	-	4554	-	-	-	-
Asia
Al Amer 2020 [49]	Prev. (%)	-	-	65.0%	-	-	-	-	-	-	-
	Nb studies	-	-	15—He	-	-	-	-	-	-	-
	Nb subjects	-	-	4032	-	-	-	-	-	-	-
Gorce et al., 2023 [32]	Prev. (%)	50.4%	-	-	35.6%	-	25.8%	-	-	-	-
	Nb studies	4—He	-	-	4—He	-	3—He	-	-	-	-
	Nb subjects	628	-	-	628	-	493	-	-	-	-
Jacquier-Bret et al., 2025 [29]	Prev. (%)	45.7%	34.1%	58.4%	43.0%	14.9%	32.2%	22.5%	40.7%	33.4%	84.3%
	Nb studies	39—He	32—He	38—He	37—He	33—He	32—He	30—He	35—He	33—He	29—He
	Nb subjects	19,306	15,331	19,186	18,666	17,160	16,932	16,477	17,800	17,160	16,145
Saberipouret al., 2019 [51]	Prev. (%)	-	-	60.5%	-	-	-	-	-	-	84.2%
	Nb studies	-	-	31—He	-	-	-	-	-	-	14—He
	Nb subjects	-	-	11,743	-	-	-	-	-	-	4490
Soroush et al., 2018 [26]	Prev. (%)	47.8%	44.6%	61.0%	42.8%	17.5%	36.8%	24.6%	46.5%	43.7%	-
	Nb studies	25—He	17—He	38—He	16—He	17—He	16—He	12—He	21—He	13—He	-
	Nb subjects	9137	6079	15,835	5452	6555	5619	4420	7728	4525	-
Wang et al., 2024 [27]	Prev. (%)	58.0%	-	35.0%	49.0%	15.0%	25.0%	21.0%	31.0%	30.0%	79.0%
	Nb studies	16—He	-	17—He	16—He	16—He	16—He	14—He	16—He	15—He	19—He
	Nb subjects	-	-	-	-	-	-	-	-	-	21,042
Zakerjafariet al., 2018 [52]	Prev. (%)	51.9%	-	37.3%	33.2%	12.9%	33.7%	11.9%	17.6%	12.9%	-
	Nb studies	21—He	-	18—He	17—He	11—He	18—He	11—He	11—He	8—He	-
	Nb subjects	2371	-	1500	1954	898	2053	854	852	643	-
Europe
Gorce et al., 2023 [32]	Prev. (%)	54.1%	-	-	51.4%	-	31.8%	-	-	-	-
	Nb studies	10—He	-	-	10—He	-	6—He	-	-	-	-
	Nb subjects	1540	-	-	1634	-	694	-	-	-	-
Gorce et al., 2025 [28]	Prev. (%)	49.9%	46.3%	61.4%	39.3%	13.4%	32.3%	20.8%	36.6%	24.7%	87.8%
	Nb studies	11—He	6—He	12—He	12—He	7—He	10—He	6—He	9—He	9—He	10—He
	Nb subjects	3757	2778	5153	5153	3070	4451	2849	3526	4230	3877

Prev.: prevalence; He: heterogeneous data distribution, application of a random effects model; Nb: number.

presents the pooled overall WMSD prevalence from each meta-analysis as well as the prevalence of the nine body areas. When data were available, the number of studies, the computation model used, and the corresponding sample size were also reported. A large variability was observed in the number of WMSD prevalence reported in the meta-analyses: two studies reported all 10 prevalence rates [28,29]; two studies reported prevalence for all nine areas but did not report the overall prevalence [26,31] one study was missing an area and the overall prevalence [27]; one study had two missing prevalence rates [53], and six studies reported between 1 and 3 prevalence rates only [28,29,32,49,50,51]. The neck, lower back, and wrist were the most studied areas, with nine values reported.

3.4. Overlap Assessment and Study Selection

The construction of the two-by-two meta-analysis comparison matrix resulted in 242 primary cross-sectional studies. The CCA computation qualified the overall overlap as low, with a mean value of 0.87% (10 meta-analyses, 261 citations across all studies, 242 primary studies) with a range between 0.0% and 9.3%. The distribution of overlap across the 45 comparisons was as follows: 93.33% (n = 42) had a low overlap, and 6.67% (n = 3) had a moderate overlap (Figure 2). The detailed overlap analysis ispresented using the GROOVE tool [45] in Appendix B.

3.5. Umbrella Review Results

No meta-analysis could be conducted for America due to the presence of only one study for this continent. Two studies were included for Africa. The meta-analysis could only be performed for the lower back. Three meta-analyses could be carried out for Europe (neck, shoulder, and wrist), while all areas could be analyzed for Asia. All results were presented in forest plot diagrams by continent. For each study, the authors, their country of origin, the category of healthcare professionals studied, the sample size, the prevalence with its 95% confidence interval, and the weight were indicated.

3.5.1. Overall Prevalence

Only three meta-analyses reported an overall WMSD prevalence for healthcare professionals: two studies in Asia and one in Europe. The prevalence using a fixed-effects model was pooled only for Asia (84.3%; 95% CI: 83.0–85.5%, n = 20,635), as displayed in Figure 3.

3.5.2. Prevalence by Body Area

The studies by Chenna et al. [34] and Wang et al. [27] in Asia did not report sample sizes in the presentation of prevalence results. Therefore, they were removed from the subgroup analysis for Asia. The following sections detail the results for the nine body areas by continent and worldwide. The Q and I² statistics parameters revealed high heterogeneity across body areas and continents. Thus, a random-effects model was applied to pool all prevalence values, except for the wrist in Europe.

Neck

Subgroup analyses reported a prevalence of 48.4% (95% CI: 45.9–51.0%, n = 31,442) in Asia and 51.7% (95% CI: 47.7–55.8%, n = 5297) in Europe (Figure 4).

Upper Back

The upper back WMSD prevalence (Figure 5) was pooled only for Asia (39.3%, 95% CI: 29.0–49.6%, n = 21,410).

Lower Back

The lower back was the most studied area (seven meta-analyses) and the only body area for which a subgroup analysis could be conducted in Africa. Figure 6 depicts the results for Africa and Asia. The pooled prevalence for Africa (54.5%, 95% CI: 35.5–73.4%, n = 10,040) and Asia (56.6%, 95% CI: 51.3–61.9%, n = 52,296) were similar.

Shoulder

The shoulder WMSD prevalence was reported for Asia and Europe (Figure 7). The prevalence pooled in Asia and Europe was 38.9% (95% CI: 34.6–43.3%, n = 26,700) and 45.3% (95% CI: 33.4–57.1%, n = 6787), respectively.

Elbow

The elbow WMSD prevalence was 15.3% (95% CI: 13.1–17.5%, n = 24,613, Figure 8), pooled only for Asia from three meta-analyses.

Wrist

The wrist WMSD prevalence was investigated in Asia and Europe, as shown in Figure 9. The pooled values were very close for the two continents: 32.6% (95% CI: 29.4–35.9%, n = 25,097) and 32.2% (95% CI: 30.7–33.8%, n = 5145) for Asia and Europe, respectively.

Hip

The hip WMSD prevalence was investigated in three meta-analyses in Asia (Figure 10). The pooled prevalence was 19.8% (95% CI: 14.5–25.1%, n = 21,751).

Knee

Figure 11 shows the results of knee WMSD prevalence for the three meta-analyses. The overall pooled prevalence among the 26,380 Asian healthcare professionals was 35.0% (95% CI: 23.7–46.4%).

Ankle

The ankle WMSD prevalence was presented in Figure 12 only for Asia. The prevalence was 30.1% (95% CI: 17.3–42.8%, n = 22,328) from three meta-analyses.

3.5.3. WMSD Prevalence Synthesis

Only Asia showed prevalence for all nine body areas. Values could only be pooled for one body area in Africa and three in Europe. The prevalence of WMSDs by continent for each body area is presented in Figure 13. The lower back was the most exposed area across all continents, with a prevalence ranging from 54.5% in Africa to 56.6% in Asia, followed by the neck (48.4% and 51.7% for Asia and Europe, respectively). Prevalence was higher in Europe than in Asia for neck and shoulder pain.

Table 4. Classification of most exposed body areas by continent.

Ranking	Asia	Europe
1	Lower back (56.6%)	Neck (51.7%)
2	Neck (48.4%)	Shoulder (43.5%)
3	Upper back (39.3%)	Wrist (32.2%)
4	Shoulder (38.9%)
5	Knee (35.0%)
6	Wrist (32.6%)
7	Ankle (30.1%)
8	Hip (19.8%)
9	Elbow (15.3%)

complements this information by showing the ranking of all body areas in Asia and for the three evaluated body areas in Europe.

4. Discussion

The present study summarizes the overall and by body area WMSD prevalence among healthcare professionals. To our knowledge, this is the first umbrella review and meta-analysis on the issue of WMSDs in healthcare professionals by continent. It is based on an unlimited date search, which identified 10 meta-analyses for a total of 242 unique non-overlapping cross-sectional studies.

4.1. WMSD Prevalence—Overall and by Body Area

The prevalence of WMSD among professionals from different continents, pooled from available meta-analyses, was estimated between 79% and 87.8%, based on more than 45,000 participants. Although the number of professionals surveyed is significant, this value remains little reported in the literature since only 4 of the 10 meta-analyses provided this information. As a result, the analysis by continent could only be conducted for Asia (84.3%). It is noted that the four available meta-analyses (three in Asia and one in Europe) only concerned nurses. These prevalence values appear higher than those reported for other health professions: 78.4% for dentists [34] or 56.4% for emergency medical workers [53]. It is clear that the low number of meta-analyses available tohealthcare professionals, as well as the representativeness of the different professions, hasan impact on the results that an umbrella can provide.

Analysis by body area highlighted that the lower back and neck were the most exposed areas. Regardless of the continent, healthcare professionals were heavily impacted at the cervical and lumbar spine level. Their work with patients, whether involving manipulation, administering care, or performing surgical procedures, induces a heavy strain on this part of the body. Many studies reported a high prevalence of low back pain (>70%) among nurses [54], dentists [55], physiotherapists [56], or surgeons [57]. Similarly, a high prevalence of neck pain has been observed among surgeons [58], dentists [59], nurses [60], or, more generally, healthcare professionals [61]. Hip and elbow appeared as the least impacted areas with a prevalence of approximately 15–20%. The constraints of different health professions are mainly associated with the upper part of the body, which may explain the lower prevalence oflower limb pain. However, some pathologies, such as osteoarthritis or varicose veins, can appear by the repetition of awkward postures such as maintaining static posture for long periods during operations (i.e., in dentists [62] or surgeons [63]) or numerous displacements, for example, in nurses [64].

The analysis by continent revealed that the prevalence ofneck and shoulder conditions was higher in Europe than in Asia by 3% and 6%, respectively. This could be partly explained by the choice of inclusion criteria in the meta-analysis of Gorce et al. [32] carried out among surgeons in Asia and Europe. Indeed, this study considered only cross-sectional studies, in which surgeons used video or robotic assistance. However, it has been shown that the use of assistance promotes the appearance of WMSDs, particularly in the neck and shoulder [35]. On the other hand, the weight of this meta-analysis in Europe was much higher than its weight in Asia for the neck (43.8% vs. 13.1%) and shoulder (49.2% vs. 20.9%). This difference comes from a significant gap in sample sizes for each continent: participants in Europe represented 1540/5297 (29.6%) and 1634/6787 (24.1%), respectively, for the neck and shoulder, while in Asia, the ratios were 628/31442 (2.0%) and 628/26700 (2.4%). These promising results should be confirmed by additional data from a larger number of countries and covering all health professions. These investigations should be continued on worldwide cross-sectional studies to subsequently conduct meta-analyses that will contribute to umbrella reviews and thus knowledge on WMSD among health professionals.

4.2. Limitations and Recommendations for Future Work

The first limitation was the number of available meta-analyses that address the issue of WMSDs by continent among healthcare professionals. However, there are several meta-analyses in Asia concerning nurses, dentists, surgeons, and healthcare professionals. In Africa and Europe, the number of meta-analyses is very limited (two per continent), covering nurses and healthcare professionals or nurses and surgeons, respectively. In addition, in the meta-analyses, not all body areas were always effectively included. Therefore, it would be interesting in future work to rigorously consider the numerous cross-sectional studies to propose meta-analyses by continent and, thus, complete and strengthen the initial results presented in this first umbrella review.

The second limitation concerns the representativeness of different health professions in meta-analyses. Currently, there are far more studies on nurses than on other professions (half of the meta-analyses concern nurses). Therefore, the prevalence data better reflect the specificities of this profession. It would be advisable to continue the work on dentists, physiotherapists, surgeons, midwives, etc., to have a more balanced distribution of health professions.

The third limitation was heterogeneity. A large disparity was observed in the demographic and anthropometric characteristics of the participants (age, experience, gender, height, weight, etc.), sample sizes, and working or environmental conditions (cultural, economic, political). Because all of these parameters could affect prevalence measures, interpretation should be made with caution, even if the random-effects model is robust to heterogeneity. Future studies could be conducted using subgroup analyses both in cross-sectional studies and in meta-analyses to modulate this heterogeneity.

Other limitations are methodological. The PRISMA method was used to select the meta-analyses for the umbrella review. The choice of keywords and inclusion/exclusion criteria (including language and databases) may have limited the search and, thus, omitted some studies. Another methodological limitation concerns the method of collecting prevalence data. There may be disparities in the types of questionnaires used or in the definition of prevalence itself (assessment duration over 7 days, 12 months, career, etc.), which are not always mentioned in the studies. This parameter may be one cause of variability in the assessment of overall prevalence and by body area. In the future, it would be relevant to precisely define prevalence and the tool used to facilitate the exploitation of data in subgroup analyses of meta-analyses and umbrella reviews.

5. Conclusions

This umbrella review and meta-analysis revealed an imbalance in the number of works carried out on different continents. Asia is the most documented and the only one for which prevalence analysis could be carried out on all body areas and overall. In Europe, only the neck, shoulder, and wrist could be assessed. In Africa, analysis could only be performed on the lower back. The results showed that meta-analyses on nurses were predominant and that the lower back was the most exposed area among healthcare professionals. It would be relevant in future work to consider the numerous cross-sectional studies to conduct subgroup analyses by continent and, thus, complete and strengthen the initial results presented in this first umbrella review.