Characteristics of subjects self-reporting arthritis in a population health survey: distinguishing between types of arthritis.

OBJECTIVES Arthritis is a broad term covering disparate diseases with varying prognoses. Epidemiological surveys are important tools for arthritis research, but they either fail to specify arthritis subtypes or they provide self-reported arthritis data that are potentially misclassified. This limits their use for research about arthritis subgroups. This study describes and compares characteristics of subjects self-reporting subtypes of arthritis in a Canadian epidemiological survey. We also consider the feasibility of developing methods for distinguishing subtypes of arthritis in such population surveys. METHODS Using data from 119,904 adult participants in the Canadian Community Health Survey (CCHS) cycle 3.1, we identified those self-reporting one of four subtypes of arthritis and compared the four groups with regard to socio-demographic status, lifestyle and health characteristics, medication use, health care utilization and functional outcomes. Cross-tabulations of weighted prevalence were estimated and tested for statistical significance using the chi-square test. RESULTS Descriptive results showed very few distinguishing characteristics across self-reported arthritis subtypes on 34 investigated variables. Participants with osteoarthritis were more likely to be older and female than other groups. Statistical testing showed no difference between rheumatoid arthritis, osteoarthritis and "other" type of arthritis for physical activity level, health conditions, medication use, health care utilization and functional limitations. DISCUSSION Characteristics of subjects who self-report different types of arthritis in a typical population health survey (CCHS) are not sufficiently dissimilar to justify valid data analyses and interpretation by arthritis subgroup. Future studies might focus on identifying and implementing supplemental questionnaire items in epidemiological population surveys.