Injury Profile in Youth Female Athletes: A Systematic Review and Meta-Analysis

Abstract

Background: An increasing number of epidemiological studies assessing the incidence, prevalence and severity of injury in youth female sport are available. However, no study has sought to synthesise the current evidence base across all youth female sport. As such, a systematic review and meta-analysis of injury in this cohort is necessary to understand the diversity of injury and its associated burden between sports in addition to identifying the density of research available.

Objective: To conduct a systematic review and meta-analysis of epidemiological data of injuries in youth female athletes with particular attention to injury incidence, mean days lost and injury burden.

Methods: Searches were performed in PubMed, EBSCO (SPORTDiscus with Full Text MEDLINE, APA PsycINFO, CINAHL, Academic Search Complete) and Cochrane databases. Studies were considered if they reported time-loss injury incidence or prevalence in youth female (≤ 19 years old) athletes. Study quality and risk of bias were assessed using STROBE-SIIS extension, Newcastle-Ottawa Scale, and funnel plots, respectively. Injury incidence and burden rate data were modelled using a mixed-effect Poisson regression model. Days lost data were modelled using a generalised linear mixed model.

Results: Thirty-two studies were included. The overall incidence rate, mean days lost per injury, and burden rate were 4.4 injuries per 1000 h (95% CI 3.3-5.9), 10 days (95% CI 6-15), and 46 days per 1000 h (95% CI 23-92), respectively. Forty percent of athletes sustained at least one time-loss injury. Competitive level was a significant moderator of match and training injury incidence, with elite youth athletes presenting greater pooled injury incidence estimates than non-elite athletes (p = 0.0315 and p = 0.0047, respectively). The influence of moderators on days lost and injury burden could not be determined due to an insufficient number of studies for analysis.

Conclusion: Despite a broad inclusion criterion, there is limited injury surveillance research available across youth female sport. Outside of soccer, little research density is evidenced with single studies available in popular team sports such as Australian football and rugby union. Insufficient study numbers reporting mean days lost and injury burden data were available for analysis, and pooled days lost data could only be estimated for soccer. This highlights a need for future research to report days lost data alongside injury number and exposure so burden can be calculated and the full risk of injury to youth female athletes can be identified.

Fig. 1

PRISMA flow diagram for study selection process

Fig. 2

Incidence of total injuries (with 95% confidence intervals) by sport. Study reference, study setting and total number of injuries are provided for each study. The location of the diamond represents the estimated incidence rate and the width reflects the precision of the estimate. The dashed line represents the prediction interval and shows the range of the true effect in 95% of study settings. RE random effects

Fig. 3

Incidence of match injuries (with 95% confidence intervals) by sport. Study reference, study setting and total number of injuries are provided for each study. The location of the diamond represents the estimated incidence rate and the width reflects the precision of the estimate. The dashed line represents the prediction interval and shows the range of the true effect in 95% of study settings. RE random effects

Fig. 4

Incidence of training injuries (with 95% confidence intervals) by sport. Study reference, study setting and total number of injuries are provided for each study. The location of the diamond represents the estimated incidence rate and the width reflects the precision of the estimate. The dashed line represents the prediction interval and shows the range of the true effect in 95% of study settings. RE random effects

Fig. 5

Meta-analysed proportions for injury mechanism (A), injury onset (B), injury severity time-bins (C), injury region (D), injury location (E) and injury type (F). Proportions of each study were combined in the meta-analysis. Error bars represent 95% confidence intervals of estimate proportions. Number: total number of injuries, (): number of studies and []: percentage of total estimated proportion