To investigate whether there is familial aggregation in the prevalence of osteoarthritis, and to determine the risk factors of osteoarthritis in rural population of Shanxi Province, thereby providing new ideas for early prevention and treatment of osteoarthritis.

The resident households in rural communities of Shanxi Province were included as the subjects in the study by random cluster sampling method. Sasinformat (SAS) 9.2 software was used for univariate analysis by t-test or chi-square test. The generalized linear mixed model was used to determine the familial aggregation of osteoarthritis.

The valid final number of households was 1 664 with a total of 4 891 subjects. Excluding those under 40 years old, and re-including the households with 2 or more family members into the model, there were 642 households and 2 351 subjects in total. According to univariate analysis and generalized linear mixed model analysis, age, body mass index (BMI), education level, as well as the heating and ventilation in residential environment were the influence factors of osteoarthritis. Family as a random effect was shown to be statistically significant in the total population and in the population over 40 years old. A total of 10.97% and 18.42% variation in the disease outcome were caused by families, indicating that there was a familial aggregation in osteoarthritis.

Many factors can affect the prevalence of osteoarthritis in rural population of Shanxi Province, and familial aggregation exists in osteoarthritis. However, familial aggregation may be caused by genetic factors, or living environment and habits, which needs validation in further studies. Nonetheless, the study on familial aggregation of osteoarthritis provides new ideas and perspectives for early prevention and treatment of osteoarthritis.

Osteoarthritis; Risk factors; Random effect; Generalized linear mixed model; Familial aggregation