Abstract
CrossFit® is a fitness program characterised by founder Greg Glassman as “constantly varied, high-intensity, functional movement” and is typically coupled with other lifestyle recommendations, most prominently those related to diet and nutrition. The overarching aim of this thesis was to advance the scientific understanding of injury epidemiology associated with CrossFit participation. To address this aim, a thesis consisting of ten chapters is presented. Following the introductory material of Chapter 1, the thesis commences with a scoping review of studies investigating injury and CrossFit participation. The scoping review (Chapter 2) provides a broad perspective of the research ‘landscape’ and locates the intersection points between rank in an evidence-hierarchy and step/s of the Translating Research into Injury Prevention Practice framework (TRIPP). A key finding of the scoping review was identification of a discrete subgroup of studies reporting the epidemiology of injury associated with CrossFit participation. In order to systematically review these studies, an appropriate tool to assess risk-of-bias in cross-sectional surveys of self-reported injury was required. Chapter 3 reports the development of an appraisal tool consisting of previously published items modified for the intended context of CrossFit. The new appraisal tool, labelled the ‘Risk-of-bias Appraisal Tool for Cross-sectional studies (‘RIBAT-C’), was subsequently deployed within a systematic review of studies reporting injury epidemiology in CrossFit participants (Chapter 4). A best-evidence synthesis of six studies identified an injury incidence within the range 0.27 to 3.3 injuries per 1000 h of exposure but this apparently low value must be interpreted cautiously and in light of the high risk-of-bias evident in these studies. Taken together, the conclusions of the scoping review (Chapter 2) and systematic review (Chapter 4) indicate a need to better understand injury epidemiology in CrossFit participants (TRIPP step 1) including consideration of aetiological factors (TRIPP step 2). In preparation for a prospective observational cohort study, Chapter 5 presents a narrative review of selected methodological issues involved in conducting an observational study including the role of workload, movement quality, and previous injury on subsequent injury risk. Movement quality is a putative risk factor for musculoskeletal injury that has been evaluated by clinicians using various assessment tools, including, most prominently, the Functional Movement Screen (FMS). Two systematic reviews, the first of inter- and intra-rater reliability of FMS scores (Chapter 6), and the second investigating the strength of association between FMS scores and subsequent injury (Chapter 7) were undertaken to investigate the FMS. The findings of all previous chapters then converge on a prospective observational cohort study designed to address the aim of investigating injury epidemiology (particularly injury incidence), and identifying potential risk factors that may be associated with injury aetiology in CrossFit participants. Chapter 8 reports the findings of pilot work undertaken to consider the feasibility of community-based data collection before Chapter 9 reports a prospective observational cohort study of 30 adult recreational CrossFit participants monitored over a 26-week period. The key findings of Chapter 9 were: a) within the CrossFit population, measures of injury severity incorporating reduction in participation, rather than time-loss, are likely to be important, b) the average weekly prevalence for both gradual and acute onset injuries of substantial severity was ~ 3.6% (95%CI 0 to 10.2), c) the incidence of gradual onset injuries of substantial severity was 3.1 per 1000 h exposure (95%CI 1.1 to 5.1); and the incidence of acute onset injuries of substantial severity was 2.7 per 1000 h exposure (95%CI 1.2 to 4.2), d) participants in this study demonstrated low variance in acute:chronic workload ratio (ACWR) with the mean for all participants approximating 1.1 (95% CI 1.03 to 1.14), e) the stability of ACWR observed in this sample is attributable to three determinants, firstly, the uniformity of session intensity which was, on average, 6.4 out of 10 (95%CI 6.1 to 6.7), aligning with a Borg descriptor between “hard” (6 out of 10) and “very hard” (7 out of 10); secondly, the near uniform duration of 60 min for each workout session (consistent with typical CrossFit affiliate practices for group classes); and finally, in this sample, the low variation in number of weekly sessions (95%CI 3.7 to 4 sessions per week). If the stability of these parameters is typical of recreational CrossFit participants generally, it suggests that ACWR may not be a viable injury risk factor in this population. The magnitude of effect for association between the strongest of the movement quality indices and subsequent injury risk was ‘small to moderate’, with the lower and upper limit of the confidence interval ranging between ‘trivial’ and ‘large’. Notwithstanding the limited precision associated with the small sample, the magnitude of effect observed here is arguably insufficient to justify use of the movement quality indices investigated here as isolated indicators of injury risk. Finally, Chapter 10 provides an overview of findings arising from the thesis including recommendations for advancing scholarship in the area of injury prevention research specific to CrossFit and analogous forms of exercise activity.