The revised Oswestry Low Back Pain Questionnaire (ROLBPQ) and Roland-Morris Activity Scale (RMAS) were compared in a randomized controlled trial of chiropractic manipulation, stroking massage, corset and transcutaneous muscular stimulation (TMS). This trial employed specific inclusion and exclusion criteria, including nonspecific low back pain for a duration of 3 wk to 6 months and ages between 18 and 55. We had the opportunity to ask 85 patients to answer the questionnaires. Sixty-three patients, who completed the initial and final evaluations, were used for data analysis. Both ROLBPQ and RMAS showed good internal consistency with alpha coefficients ranging from .77 to .93. Both instruments showed a significant difference between the chiropractic manipulation and massage groups (p less than .05). RMAS was able to further show significant differences between the chiropractic manipulation and TMS groups, and between the corset and massage groups, but the ROLBPQ failed to do so. RMAS also showed that chiropractic manipulation had a better but nonsignificant result than corset, possibly due to insufficient sample size and/or duration of treatment. We conclude that both instruments are reliable for measuring low back pain disability, and chiropractic manipulation has a superior short-term benefit when compared to stroking massage and TMS in subacute low back pain patients. In addition, it appears that RMAS is preferable in a clinical trial situation for subacute low back pain because it is more sensitive than ROLBPQ to detect changes.