Objective: To characterize and investigate the relationship between stress and personality in people with rheumatoid arthritis (RA) compared to those with osteoarthritis (OA).
Design: Survey of personality characteristics determined by using Cattell's 16 Personality Factor (16PF) questionnaire and stressful life events at disease onset determined by using the Social Readjustment Rating Scale of Holmes and Rahe.
Setting: Inpatients of an arthritis hospital and outpatients of a clinic, both of which specialize in rheumatic diseases and musculoskeletal problems.
Participants: Diagnosed as "definite" or "classical" RA (N = 128) according to the ARA Diagnostic Criteria for Rheumatoid Arthritis (1958 Revision) or as OA (N = 79) according to radiological and clinical evidence; randomly selected, resulting in a close match for gender, age at disease onset, duration of disease, functional classification, and pattern of disease progress. All individuals approached participated.
Main outcome measures: Twenty personality characteristics, as well as amount of stressful life events at disease onset (Life Change Units).
Results: The RA subjects had significantly (p < .001) more stress at disease onset compared to the OA subjects. A subgroup of high-stress-at-onset RA subjects experienced a higher degree of disease severity (p < .01) than did the RA subjects with no stress at onset. Although the mean personality scores for those with RA were not significantly different from those with OA, the RA personality frequency distributions were different. Some of these differences could be explained by the stress-at-onset subgroup personality characteristics.
Conclusions: As a group, subjects with RA exhibited more stress at disease onset than those with OA. Although mean personality scores were not different between RA and OA subjects, they did exhibit different frequency distributions. There appeared to be a high-stress-at-onset subgroup of RA patients who had a worse disease prognosis and who corresponded to a personality frequency subgroup. The interaction between these variables is more complex than implied by the "RA personality" concept.