OBJECTIVE: Genetic factors are important in the aetiology of hip osteoarthritis (OA), but studies are limited by cross-sectional design and poor association with clinically important disease. Identifying cohorts with progressive OA will facilitate development of OA biomarkers. Using a middle-aged cohort with genetic predisposition to hip OA and a control group, we compared the prevalence of clinical and radiographic hip OA and incidence of progression over 5 years. DESIGN: 123 individuals (mean age 52 years) with a family history of total hip arthroplasty (THA) ('sibkids') were compared with 80 (mean age 54 years) controls. The prevalence of radiographic OA [scored according to Kellgren & Lawrence (K&L)], clinical features, and incidence of clinical progression over a 5-year period were compared. A multivariate logistic regression model was used to adjust for confounders. RESULTS: Sibkids had odds ratios (ORs) of 2.7 [95% confidence interval (CI) 1.1-6.3, P = 0.02] for hip OA (K&L grade >/=2), 3.4 (1.4-8.4, P = 0.008) for clinical signs, and 2.1 (0.8-5.8, P = 0.14) for signs and symptoms. Over 5 years, sibkids had ORs of 4.7 (1.7-13.2, P = 0.003) for the development of signs, and 3.2 (1.0-10.3, P = 0.047) for the development of signs and symptoms. DISCUSSION: Compared to a control group and after adjustment for confounders, individuals with genetic predisposition to end-stage hip OA have higher prevalence of OA, clinical features, and progression. In addition to structural degeneration, the inherited risk may include predisposition to pain. Genetically-loaded cohorts are useful to develop hip OA biomarkers, as they develop progressive disease at a young age.