Concussed athletes walk slower than non-concussed athletes during cognitive-motor dual-task assessments but not during single-task assessments 2 months after sports concussion: a systematic review and meta-analysis using individual participant data

ObjectivesTo determine whether individuals who sustained a sports concussion would exhibit persistent impairments in gait and quiet standing compared to non-injured controls during a dual-task assessment .DesignSystematic review and meta-analysis using individual participant data (IPD).Data sourcesThe search strategy was applied across seven electronic bibliographic and grey literature databases: MEDLINE, EMBASE, CINAHL, SportDISCUS, PsycINFO, PsycARTICLES and Web of Science, from database inception until June 2017.Eligibility criteria for study selectionStudies were included if; individuals with a sports concussion and non-injured controls were included as participants; a steady-state walking or static postural balance task was used as the primary motor task; dual-task performance was assessed with the addition of a secondary cognitive task; spatiotemporal, kinematic or kinetic outcome variables were reported, and; included studies comprised an observational study design with case–control matching.Data extraction and synthesisOur review is reported in line with the Preferred Reporting Items for Systematic review and Meta-Analyses-IPD Statement. We implemented the Risk of Bias Assessment tool for Non-randomised Studies to undertake an outcome-level risk of bias assessment using a domain-based tool. Study-level data were synthesised in one of three tiers depending on the availability and quality of data: (1) homogeneous IPD; (2) heterogeneous IPD and (3) aggregate data for inclusion in a descriptive synthesis. IPD were aggregated using a ‘one-stage’, random-effects model.Results26 studies were included. IPD were available for 20 included studies. Consistently high and unclear risk of bias was identified for selection, detection, attrition, and reporting biases across studies. Individuals with a recent sports concussion walked with slower average walking speed (χ2=51.7; df=4; p<0.001; mean difference=0.06 m/s; 95% CI: 0.004 to 0.11) and greater frontal plane centre of mass displacement (χ2=10.3; df=4; p=0.036; mean difference −0.0039 m; 95% CI: −0.0075 to −0.0004) than controls when evaluated using a dual-task assessment up to 2 months following concussion.Summary/conclusionsOur IPD evidence synthesis identifies that, when evaluated using a dual-task assessment, individuals who had incurred a sports concussion exhibited impairments in gait that persisted beyond reported standard clinical recovery timelines of 7–10 days. Dual-task assessment (with motion capture) may be a useful clinical assessment to evaluate recovery after sports concussion.Protocol pre-registrationThis systematic review was prospectively registered in PROSPERO CRD42017064861.