Background/Purpose: Osteoarthritis (OA) and diabetes commonly co-occur. Potential explanations include common risk factors (aging, obesity) and the effects of OA-related functional limitations on diabetes risk factors (e.g., sedentary behavior exacerbates metabolic syndrome). However, whether or not there is a causal relationship between OA and diabetes is unclear. In a large population-based cohort free of diabetes at baseline, we examined the relationship between self-reported hip and knee OA and incident diabetes.

Methods: A population cohort aged ≥55 years was recruited from 1996-98 and followed through provincial health administrative data to 2014. Subjects with baseline diabetes, rheumatic diseases, and medical conditions associated with functional disability were excluded. Age, sex, height, weight, joint complaints and functional limitations were collected. Hip and knee OA were defined as swelling, pain, or stiffness in any joint lasting 6 weeks in the past 3 months and indication on a joint homunculus that a hip or knee was “troublesome”. Comorbidities were defined using validated algorithms for health administrative data. Using Cox-regressions, we examined the relationship of baseline hip/knee OA (0-2 hips; 0-2 knees) with subsequent incident diabetes as defined from health administrative data (sensitivity – 86%, specificity – 97%), incrementally controlling for age, sex, BMI, preexisting hypertension and cardiovascular disease (CVD), income and prior primary care exposure, and finally walking limitation.

Results:  16,362 participants without baseline diabetes were included: median age 68 years, 61% female and median BMI 25.3 kg/m². 1,637 (10%) individuals met criteria for hip OA, 2,431 (15%) for knee OA, and 3,908 (24%) for walking limitation. Over a median follow-up of 13 years, 3,539 individuals (22%) developed diabetes. Controlling for baseline age, sex, income, BMI, preexisting hypertension and CVD, and prior primary care exposure, a significant dose-response relationship was observed between number of hip/knee joints with OA and incident diabetes: HR for two vs. no OA hips 1.25, 95% CI: 1.08-1.44 (p=0.003); HR for two vs. no OA knees 1.16, 95%CI: 1.04 -1.29 (p=0.008). Further adjustment for walking limitation resulted in attenuation of these relationships, which became non-significant.

Conclusion: In a large population cohort aged ≥55 years free of diabetes at baseline and after controlling for multiple confounders, the presence and burden of hip and knee OA was a significant independent predictor of incident diabetes. This association was explained largely by OA-related walking limitation. Increased attention to management of hip and knee OA with a view to improving mobility has potential to reduce risk of incident diabetes.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/the-impact-of-hip-and-knee-osteoarthritis-on-the-subsequent-risk-of-incident-diabetes-a-population-based-cohort-study/