Despite strong evidence that arthritis symptoms and disability can be reduced through physical activity, less than one quarter of U.S. adults with arthritis meet recommended physical activity guidelines. Research supports the use of self-efficacy enhancing strategies in interventions to improve symptoms and increase physical activity in older adults with arthritis. The aim of this study was to re-evaluate an intervention program called Choosing Arthritis Appropriate Physical Activity (CASAPA) from post-intervention to 1-month follow-up, to determine if improvements seen at post-intervention were maintained, and to examine the predictive ability of physical activity self-efficacy subtypes and intention for predicting physical activity. CASAPA was 4-week, psycho-behavioral, educational intervention designed to increase knowledge, skills, and confidence in selecting arthritis appropriate physical activity, delivered to an ethnically diverse, low-SES sample of older adults (n = 143, M age = 66.93), recruited from non-residential community senior centers in San Diego County. Participants completed written questionnaires at baseline, postintervention, and 1-month follow-up to measure self-efficacy for arthritis related exercise, for overcoming barriers to exercise, and for choosing appropriate physical activity. McNemar's test was used to determine if improvements in self-efficacy measures seen at postintervention in the intervention group were maintained at 1-month follow-up. Linear regression was used to test the predictive ability of the self-efficacy subtypes and intention for physical activity. Results show no significant change in self-efficacy levels for the intervention or control group, suggesting that the intervention group maintained their improved self-efficacy, and remained higher than controls at 1-month follow-up. Results also show that neither the self-efficacy subtypes nor intention at post-intervention were significant predictors of physical activity at follow-up. Findings support evidence that arthritis related self-efficacy subtypes can be increased and maintained for some time following an intervention, but that further research is needed to understand the roles of these self-efficacy subtypes and physical activity intention on physical activity in older adults with arthritis.