Socio-demographic factors, reproductive history and risk of osteoarthritis in a cohort of 4.6 million Danish women and men

Summary Objective Studies addressing possible socio-demographic and reproductive factors in the aetiology of osteoarthritis (OA) are few. We studied possible influences of educational level, household income, marital status and parenting patterns on OA risk overall and at anatomical sites. Method We linked national register data about socio-demographic variables, reproductive histories and OA hospital contacts to a cohort of 4.6 million Danes. Ratios of first OA hospitalisation rates (RRs) were calculated using Poisson regression. Results Overall, 100,437 women and 92,020 men had a first OA hospital contact during 91.5 million person-years between 1982 and 2008. Short education, low income and married status were significantly associated with increased OA risk, and persons with children were at higher risk of OA overall (RR=1.10 in women; RR=1.22 in men), OA knee (RRs 1.14; 1.28), OA back (RRs 1.18; 1.33), and OA hand (RRs 1.21; 1.43), but not of OA hip (RRs 0.96; 1.00) than persons without children. The RR of OA overall increased by a factor of 1.05 in women and 1.04 in men per additional child, most notably for OA knee in women (1.10 per child). Conclusion Risk of OA hospitalisation was highest among married persons and persons with short education or low income. The similar or even stronger associations with reproductive factors in men than women suggest that unmeasured lifestyle factors rather than biological factors associated with pregnancy might explain the higher OA risk in persons with children. However, the particularly strong association between parity and risk of OA knee in women is compatible with a role of pregnancy-associated factors.