Epidemiology of osteoarthritis in Sweden. Register and cohort studies on prevalence and mortality

Abstract: Osteoarthritis (OA) is the most common form of arthritis and often causes pain and functional impairment. Despite its high burden on society, knowledge about the occurrence of OA and its associated mortality is sparse. The epidemiology of OA is difficult to determine due to the patients’ fluctuating symptoms and challenges in detecting and defining the radiographic disease. The specific aims of this thesis were to estimate the current and future prevalence of OA in peripheral joints in Sweden and the mortality associated with OA using the population-based register data and data from clinical examinations.Among the residents of the Skåne region aged ≥45 years in the year 2012, persons who have consulted a physician for OA at least once were identified using the Skåne Healthcare Register (SHR) – a mandatory register containing information on all healthcare visits made in the region. Prevalence of doctor-diagnosed OA (current and expected in the next 20 years) and all-cause mortality in persons with knee or hip OA were estimated. A subset of Skåne residents aged 56 to 84 years who participated in the Malmö Osteoarthritis study (MOA) (n=1527) underwent clinical examination including knee x-ray and knee pain questionnaire in order to assess the presence of knee OA.Of adults aged 45 years or older and living in Skåne (n=531 254), 27% consulted a physician for OA in a peripheral joint at least once between 1999 and 2012 and this proportion is expected to increase to 30% by the year 2032. The 2012 prevalence of the doctor-diagnosed knee OA was 13.8%, of the hip OA – 5.8%, of the hand OA – 3.1% and of the OA in other joints – 12.4%. One in four of persons with the doctor-diagnosed OA consulted for the disease in more than one joint site. In Skåne, the hazard ratios of death in persons with doctor-diagnosed knee or hip OA compared to the general population were 0.87 (95% confidence interval 0.85, 0.89) and 0.90 (0.87, 0.92), respectively. In the MOA study, the prevalence of knee OA giving symptoms (with or without radiographic manifestations) was 15.4%. Of these, 53% consulted healthcare and received a knee OA diagnosis during an 8 year period. A number of sensitivity analyses, including the validation of knee OA diagnoses registered in SHR, quantification of bias and a simulation study assessing the impact of missing diagnostic codes on the derived prevalence estimates, confirmed robustness of the results.This thesis provides a comprehensive view of clinically relevant OA as well as radiographic findings (knee) with or without symptoms in Sweden. Results will aid the healthcare planning and evaluations of the disease burden. The findings also indicate that knee or hip OA are not associated with excess mortality on the population level; possible mechanisms include relatively easy access to care and treatments or maintaining physical activity.