Both genetic and environmental factors probably influence the severity of Spondyloarthropathies. Hip involvement, which may be used as a marker of disease severity, is more frequent in spondyloarthropathies developed in North Africa. The objective of this study was determine the predisposing factors of hip involvement in spondyloarthropathy in North Africa. Patients fulfilled the Amor or European Spondyloarthropathy Study Group (ESSG) criteria of spondyloarthropathy. The study was retrospective, cross-sectional, multicentre and carried out in North Africa. The data collected were demographic data, socio-cultural factors and clinical presentation at onset. The risk of hip involvement with regard to disease duration was estimated using Kaplan-Meier's method. The predictive value of each variable with regard to time to hip involvement was evaluated using a uni- and then a multivariate Cox proportional hazard model. Five hundred and eighteen patients were included. The risk of hip involvement was estimated at 39+/-3% after 10 yr disease duration. The factors picked up by the multivariate analysis were: diagnostic delay less than 7 yr, age at onset below 24 yr and a combination of 'lower social class' and 'no refrigerator at home'. This study confirms the high prevalence of hip involvement during the course of spondyloarthropathy in North Africa and suggests a role of environmental factors in its appearance.