Running and knee joint health: self-reported symptoms, function and osteoarthritis in recreational athletes

Background The prevalence of symptomatic knee osteoarthritis (OA) is growing - largely due to ageing populations and increasing prevalence of obesity. Evidence suggests that other lifestyle factors, including participation in different sports, could be associated with either an increased risk of - or conversely, protection against - knee OA. Running is one of the most popular pastimes in the world. For many years, an association between running and risk of knee OA has been postulated, although currently there is insufficient evidence to conclude whether running has a beneficial or deleterious effect on joint health. Objectives This thesis seeks to examine the prevalence of self-reported knee joint symptoms, function and OA in recreational runners and non-runners. The level of agreement between self-reported running (measured using a novel questionnaire) and contemporaneous running behaviour recorded from a variety of sources (e.g. electronic devices, training diaries) will be identified. Methods This longitudinal study was conducted in the UK and Australia. Baseline questionnaire data was collected between September 2015 and December 2016, with follow-up data collected in Autumn 2017. Self-reported outcomes (e.g. knee symptoms and clinician diagnosed OA) were identified in two baseline questionnaires and one follow-up questionnaire. A sub-cohort of runners submitted running diaries to identify the level of agreement between self-reported running and contemporaneous running activity. A sample of athletes also completed physical testing to objectively assess lower-limb function. Results The prevalence of knee OA did not differ between runners and non-runners (p 0.207). There was no statistically significant difference in Knee Injury and Osteoarthritis Outcome Score (KOOS) for any of its five sub-domains between runners and non-runners. Previous knee injury was associated with prevalence of knee pain in recreational runners (RR 3.55, 3.24-3.90). The knee was the most common site of running-attributed injury, accounting for almost a quarter of all running injuries. There was a trend for self-reported running behaviour to be over-estimated compared to running measured by contemporaneous sources. With increasing age, there was a general decline in performance of functional tests in recreational runners, but this trend was not consistent. Conclusions Further longitudinal data is required to assess the cause and effect relationship between running and onset and/or progression of knee OA. Additional data is also needed to explore the impact of injury in recreational runners, with the aim of developing interventions to reduce their incidence. The trend for over-estimating self-reported running may have implications for the assessment of current and future research that explores running and health outcomes, where accurately identifying the ‘dose’ of running is crucial to the integrity of results