Matched comparison of health care use and costs 3-10 years after a youth sport-related knee injury

Abstract

Purpose: Joint injury is a known risk factor for development of osteoarthritis (OA). Post-traumatic osteoarthritis (PTOA) is common in the knee joint and has been shown to develop in young to middle-aged adults 12-20 years post joint injury. Currently, there is little known about health care utilization early in the period between joint injury and PTOA onset. Health care resource use associated with OA may start prior to OA diagnosis, and understanding this pattern could be useful both planning secondary prevention interventions as well as health services planning. The purpose of this study was to determine if adolescents and young adults with a 3-10 year history of knee injury differ from age, sex and sport matched controls with respect to healthcare utilization and associated costs in the first 3-10 years post injury. Methods: Participants included the first 100 participants (50 matched-pairs) in the Alberta Youth Prevention of Early Osteoarthritis (PrE-OA Study) cohort study. This included 50 youth who suffered a sport-related intra-articular knee injury in the past 3-10 years recruited from participants of previous studies conducted at the Sport Injury Prevention Research Centre and Sports Medicine Clinic at the University of Calgary, and 50 age, sex and sport-matched uninjured controls. Intra-articular injury was defined as a clinical diagnosis of knee ligament, meniscal or other intra-articular tibio-femoral or patella-femoral injury (i.e., osteochondral, fractures etc.) that required both medical consultation with a physician or physiotherapist and resulted in a disruption of regular sport participation. Uninjured participants must not have experienced a knee injury requiring medical consultation in the past 3-10 years. All participants completed a study questionnaire that provided demographic information, and information about health-care utilization in the previous twelve months. Specifically, participants were asked to report all hospitalizations and surgeries, as well as all visits to the emergency department, family physician or sports medicine physician, physiotherapist, chiropractor, massage therapists and/or athletic therapist. Participants also reported any diagnostic imaging (e.g., MRI, CT, or x-ray) studies over the twelve months. Utilization of health services was costed in 2013 dollars using Alberta health system data sources and allied health services pricing from a large multidisciplinary rehab clinic. Descriptive statistics (means or proportions (95% CI) or median (range)) were calculated for all demographic and healthcare outcomes by study group as applicable. A Wilcoxon signed-rank test for matched groups (Z) was used to compare total number of health care visits and total healthcare costs over the study period (alpha=0.05). Two sensitivity analyses were conducted. The first re-examines the health care visit and cost comparisons excluding hospitalization and surgery. The second repeats these comparisons excluding one cost outlier from the analysis. Results: The median participant age was 22 years (range 15-26) and 42% were female. Previously injured participants had a greater median number of visits per year (3.5, range 0-31) than uninjured participants (2.5, range 0-22). However, these differences in median visits were not statistically significant (Z = -0.694; p-value = 0.492). Similarly, previously injury participants incurred greater median total health care costs ($335, range $0 - $62,233) over the twelve month study period compared to uninjured participants ($264, range $0 - $14,697). These differences were in median health care costs were not statistically significant (Z= -0.626; p-value = 0.534). Hospitalizations or surgery accounted for more than 50% of the total health care costs for both study groups, with the portion of costs associated with hospitalization or surgery 15% greater for previously injured (67%) compared to uninjured (52%) participants. Sensitivity analyses looking at first excluding hospitalization and surgery costs, and second removing a cost outlier did not change the overall conclusion that there is no statistically significant difference in utilization and total costs between uninjured and previously injured participants. Conclusions: To our knowledge, these are the first reported findings regarding health care utilization and costs in the first ten years following knee injury in those at risk of developing of PTOA. Based on point estimates in this sample, the health care and allied health care visits and their associated costs were higher in those who had sustained an earlier knee injury than those who had not, however these differences were not statistically significant. Next steps will be to conduct this analysis including an additional 50 cases and controls to increase the precision of these estimates. We speculate that there will be increased health care utilization prior to diagnosis of OA in those with a previous injury, but perhaps not as early as the 3-10 year post injury period which our study captured.