Inclusion criteria of the studies included in this review were as follows: (1) studies should state a clear and similar definition of overuse injuries (e.g. injuries without a specific identifiable event, or gradual onset, or resulting from repetitive microtrauma); (2) studies should specifically report the prevalence of overuse injuries, and; (3) the population included should be athletes of any sport, regardless of professional or amateur level and age group.

Electronic database searches were performed from the earliest record to February 2021 using MEDLINE and EMBASE (via OvidSP), SPORTDiscus, and CINAHL (via EBSCO), without language or date restriction. Additional hand searching was conducted by reviewing articles published in relevant sports journals and the reference list of all eligible studies. The terms used in the search strategy were related to 'prevalence', 'overuse injury', 'athletes', 'sports'. The complete search strategy is presented in the Supplemental online material.

For studies selection, after removing duplicates, two independent reviewers (MFF and TMN) screened all titles and abstracts identified by the search strategy for eligibility. Then, full texts for all potential eligible studies were retrieved. The same two reviewers independently evaluated all full texts, and studies fulfilling the inclusion criteria were included. Any disagreements were resolved by a third reviewer (FOM).

For data extraction, two reviewers (MFF and TMN) independently extracted the characteristics of all included studies (e.g. study design, setting, study population, overuse injuries definition, how overuse injuries were measured and their point- or period-prevalence), with a third reviewer (FOM) resolving any discrepancies. Point-prevalence is defined as the number of people with a current overuse injuries at a specific point in time and period-prevalence is defined as the number of people with overuse injuries over a specified period of time.20 For prevalence data, we extracted the total proportion of overuse injury and, when available, the proportion of overuse injury for individual and team sports and for each anatomical area. Moreover, when available, we reported prevalence of substantial overuse injuries, which is defined as injuries leading to moderate or severe reduction in training volume or performance, or complete inability to participate in sports.

Two independent reviewers (MFF and TMN) assessed risk of bias for each included study, using The Joanna Briggs Institute Prevalence Critical Appraisal Tool.21 This tool consists of 9 items specifically designed to assess the methodological quality of prevalence studies. Each item is rated as ‘yes’, ‘no’, or ‘unclear’ according to information available in each study, with a maximum score of 9 points, one point for each “yes” answer. The higher the score, the better the methodological quality. A third reviewer (FOM) solved potential disagreements. To assess if sample size of each included study was appropriate (e.g. item 3), we used the equation provided in a previous study,22 where p was the expected prevalence (29.3%), which was defined based on the results of a previous study,23 Z was the confidence level (1.96), and d was the precision (5.0%). Sample size estimation resulted in an appropriate sample size of 318 participants. A fourth reviewer (RAR) solved potential disagreements regarding risk of bias assessment.

The I² statistics was used to assess among-study heterogeneity and pooling estimated prevalence of overuse injuries in athletes of different sports, separated in individual and team sports, following the Cochrane recommendations.18 For pooling, random-effects model was used when I² ≥ 50.0%. Findings from pooling and individual studies were presented in forest plots with 95% confidence interval (CI). We planned to analyze separately the point- and period-prevalence.

Sensitivity analysis was performed to compare whether prevalence of overuse injuries in specific body regions (i.e. shoulder, lower back, knee, and hip/groin/thigh) differ between individual and team sports. All analyses were performed using Comprehensive Meta-Analysis, version 2.2.04 (Biostat, Inc.©, Englewood, New Jersey).

The GRADE (Grading of Recommendations Assessment, Development and Evaluation) system was used to summarize the overall quality of the evidence. This tool provides four levels ranging from high-quality evidence to very low-quality evidence.24 Scoring of evidence started at high-quality evidence, which was downgraded one level for each of the following pre-specified criteria: (1) poor methodological quality (average risk of bias ≤ 5 points out of 9); (2) imprecision (when ≥ 50% of studies did not present appropriate sample size – minimum of 318 participants); (3) indirectness (when more than 25% of studies used non-standard measurements of overuse injuries); and (4) inconsistency (if I² ≥ 50.0%).25,26 Single studies (n < 318 participants) were considered inconsistent and imprecise (that is, sparse data) and provided low quality evidence. This could be further downgraded to very low quality evidence if studies were assessed as having high risk of bias. Due to the small number of studies (n=3) reporting point-prevalence of overuse injuries in individual and team sports, our meta-analysis and sensitivity analysis focused on period-prevalence. Two independent reviewers (MFF and TMN) assessed the quality of evidence using GRADE and between-reviewer discrepancies were resolved by a third reviewer (FOM).

Searches retrieved 4807 titles and, after excluding duplicates, we identified 2069 references that had the title analyzed. Following the title, abstracts, and full-texts screening, 17 studies were included in this review. Included studies were published between the years of 2007 and 2021. The main reasons for excluding potential full texts (n = 63) were non-athlete participants (n = 4), no reporting of prevalence estimates (n = 53), and no adequate definition of overuse injuries (n= 6) (Fig. 1). Table 1

Fig. 1.

PRISMA flow-chart of studies through the review.

(0.45MB).

Most of the included studies (n = 10) included data from both sex,10,27–35 four included only male athletes,12,36–38 three included only females participants.39–41 Regarding the study design, thirteen were prospective cohort studies,12,27–30,32,35–41 two studies were descriptive epidemiological studies,31,33 and two were cross-sectional studies.10,34 For overuse injuries measurement, nine studies used the Oslo Sports Trauma Research Center questionnaire (OSTRC),12,27,28,30,32,33,36,39,40 four used the Oslo Sports Trauma Research Center questionnaire on health problems (OSTRC-H),29,35,37,41 one used an injury record form,31 two used a questionnaire on injuries originally developed by the authors,10,34 one used medical record data.38 Thirteen studies reported period prevalence,10,12,27,28,30,32,33,35–39,41 among them, three studies reported pre-competition prevalence,27,40,41 two studies reported pre- and during the season prevalence,35,36 four studies reported prevalence during the season,12,32,33,37 and five studies did not specify when the prevalence measure was taken.10,28,30,38,39 In addition, four studies reported point-prevalence.29,31,34,40 Twelve studies reported prevalence of overuse injuries separated by anatomical areas,12,27,28,30,32,33,35–37,39–41 one study reported separated team and individual sports,10 and twelve studies reported the prevalence of substantial overuse injuries.12,27–30,32,34–36,39–41

Mean ± standard deviation methodological quality of the included studies was 5.8 ± 1.3 of 9 (ranging from 0 to 9). Seven studies scored ≤5 of 9.10,31,32,34,36–38 Methodological quality items for each included study are reported in Table 2. Of seventeen studies, nine did not use sample frame appropriate to address the target population,12,31,32,34–36,38,40,41 seventeen did not sample study participants in an appropriate way,10,12,27–41 fourteen did not have appropriate sample size,10,12,27–29,31–33,35,37–41 five did not describe subjects and the setting in detail, 32,34-37 one did not analyze data with sufficient coverage of the identified sample,31 four did not use valid methods for the identification of overuse injuries,10,31,34,38 no study present measures in a standard and reliable way for all participants, three did not have appropriate statistical analysis,34,36,37 and three did not reach appropriate response rate.10,29,31

Pooling of 24 704 participants, 22 748 from individual sports and 1956 from team sports, from the 17 included studies was done to estimate the prevalence of overuse injury in individual and team sports separately (Fig. 2). According to the GRADE system, pooling of thirteen period-prevalence studies provided low-quality evidence that the prevalence of overuse injuries in individual sports was 42.0% (95% CI: 30.0, 55.0) and team sports was 33.0% (95% CI: 21.0, 49.0). The evidence was downgrade from high- to low-quality because of imprecision and inconsistency. Due to the small number of studies (n=3) reporting point-prevalence of overuse injuries in individual and team sports, we were not able to conduct a meta-analysis to estimate the point-prevalence.

Fig.2.

Meta-analysis for the 13 studies reporting period-prevalence of overuse injuries in team and individual sports.

(0.35MB).

Sensitivity analysis showed that the period-prevalence estimates for individual and team sports did not significantly differ across body regions. The lack of statistically significant difference is evidenced by the overlap of 95% CIs between the period-prevalence estimates between individual and team sports. Further details can be seen in Fig. 3.

Fig. 3.

Sensitivity analysis of period-prevalence for body regions in individual and team sports. Adjusted I² was presented when analysis using random-effects model.

(0.39MB).