Objective To compare the effects of 4 months of isolated lumbar resistance exercise and total body resistance exercise on walking performance in obese, older adults with chronic low back pain. A secondary analysis examined whether responsiveness to training modulated walking improvement. Design Randomized, controlled trial. Setting Research laboratory affiliated with tertiary care facility. Methods and Intervention Participants (N = 49; 60–85 years) were randomized into a 4-month resistance exercise intervention (TOTRX), lumbar extensor exercise intervention (LEXT), or a control group (CON). Main Outcome Measurements Walking performance, maximal low back strength and leg strength, and average resting and low back pain severity score (from an 11-point numerical pain rating scale; NRSpain) were collected at baseline and month 4. Results The TOTRX and LEXT improved lumbar extensor strength relative to CON, and the TOTRX (P < .05). NRSpain scores at month 4 were lowest in the TOTRX group compared with the LEXT and CON groups, respectively (2.0 ± 1.7 points vs 3.7 ± 2.6 points and 4.6 ± 2.4 points; P < .006). A total of 53% and 67% of participants in the TOTRX and LEXT groups were responders who made lumbar extensor strength gains that achieved ≥20% greater than baseline values. Although the TOTRX demonstrated the greatest improvement in walking endurance among the intervention groups, this did not reach significance (10.1 ± 12.2% improvement in TOTRX vs 7.4 ± 30.0% LEXT and −1.7 ± 17.4% CON; P = .11). Gait speed increased most in the TOTRX (9.0 ± 13.5%) compared with the LEXT and CON groups (P < .05). The change in lumbar extensor strength explained 10.6% of the variance of the regression model for the change in walking endurance (P = .024). Conclusions The use of LEXT and TOTRX produced similar modest improvements in patients’ walking endurance. Lumbar extensor strength gain compared with leg strength gain is a moderate but important contributor to walking endurance in obese older adults with chronic low back pain. Responders to resistance exercise programs (event those with only lumbar extension exercise) who make at least a 20% improvement in strength can expect better improvement in walking endurance than those who do not achieve this strength improvement.