Thesis (Ph.D.)--Boston University / PLEASE NOTE: Boston University Libraries did not receive an Authorization To Manage form for this thesis or dissertation. It is therefore not openly accessible, though it may be available by request. If you are the author or principal advisor of this work and would like to request open access for it, please contact us at open-help@bu.edu. Thank you. / The purposes of this study were to describe the relationship of sociodemographic and biomedical characteristics to health perceptions among individuals with rheumatoid arthritis (RA) and to identify which functional measures are associated with one's health perceptions. Data were extracted on 395 individuals with classical or definite RA for a secondary data analysis. These data had been collected between 1978 and 1982 using the Arthritis Impact Measurement Scales (AIMS), a reliable and valid self-administered questionnaire on health status. Sociodemographic information included age, sex, current marital status, occupation, income, and engagement in a productive role. Biomedical data included a respondent's disease duration, disease severity, and comorbid conditions. A dependent variable which classified health perceptions (HP) was constructed from two AIMS questions concerning self-assessed current health and belief in the ability to resist illness. Using subjects' sociodemographic characteristics and biomedical indicators as determinants of health perception classifications (HPC), repondents were grouped into nodes through a recursive partitioning technique (CART). Income, disease activity, current marital status, age, and comorbidity were identified by CART as important and interactive determinants of HPC. In the second phase of analysis, differences in function between HP groups on the nine AIMS scales were explored using subjects' self-assigned HP groups, while controlling for sociodemographic and biomedical characteristics and their potential interactive effects as they had been identified by the CART algorithm. Analysis of these differences demonstrated that the explanatory sociodemographic and biomedical variables used in this study both explained and masked significant differences between HP groups on certain functional measures. Finally, the analysis determined which functional measures were most associated with self-assigned HP using a stepwise logistic regression model. Anxiety, physical activity and household activity formed an overall explanatory model of HP. Physical activity and depression were associated with current health perceptions. Perceived resistance to illness was related to subjects' levels of anxiety and household activity. / 2031-01-01