Objective: To evaluate the effectiveness of Stanford's Chronic Disease Self-Management Program (CDSMP) for chronic low back pain (LBP) in older Americans. Design: Randomized controlled trial. Setting: Community-based program offered at 12 locations. Subjects: Community-dwelling seniors (n = 109) aged 60 and older with chronic LBP of mechanical origin. Methods: Patients were randomly allocated to the CDSMP or to a 6-month, wait-list control group. The program included one 2.5-hour session per week for 6 weeks. Outcomes evaluated at 6 months included 100-point modified Von Korff pain and disability scales; days with pain and disability; SF-36 general health, energy-fatigue, and emotional well-being scales; 2 scales from the Arthritis Self-Efficacy Scale, self-care attitudes/behaviors, and health services utilization. Results: For pain at 6 months, the primary outcome, the adjusted mean difference between the program and control, was −1.0 (P = .835). There was a sizable advantage for the program in disability averaged over the course of the entire 6-month study (−9.2, P = .027), but not at the 6-month follow-up (−5.8, P = .278). There was an interaction between intervention and baseline disability days favoring the program for higher baseline values (P = .007). The CDSMP affected emotional well-being (7.6, P = .037) and energy-fatigue (5.1, P = .274). There were no differences for self-efficacy, pain days, and general health. Conclusion: There was no advantage for the CDSMP over a wait-list control for improving pain, general health, self-efficacy, and self-care attitudes in older Americans with chronic LBP. A benefit was suggested for emotional well-being, fatigue, functional disability, and days with disability.