To assess the effectiveness of low-intensity laser therapy in the treatment of musculoskeletal low back pain. A double-masked, placebo-controlled, randomized clinical trial. A physical medicine and rehabilitation clinic. Sixty-three ambulatory men and women between the ages of 18 and 70yrs with symptomatic nonradiating low back pain of more than 30 days' duration and normal neurologic examination results. Subjects were bloc randomized into two groups with a computer-generated schedule. All underwent irradiation for 90 seconds at eight symmetric points along the lumbosacral spine three times a week for 4 weeks by a masked therapist. The sole difference between the groups was that the probes of a 1.06 microm neodymium:yttrium-aluminum-garnet laser emitted 542mW/cm2 for the treated subjects and were inactive for the control subjects. Subject's perception of benefit, level of function as assessed by the Oswestry Disability Questionnaire, and lumbar mobility. The treated group had a time-dependent improvement in two of the three outcome measures: perception of benefit and level of function. These results were most marked at the midpoint evaluation (p < .005, p < .01) and end of treatment (p < .017, p < .001) but tended to lessen at the 1-month follow-up (p < .10, p < .004). Lumbar mobility did not differ between the groups at any time. All tests were two-sample t tests with unequal variances. Treatment with low-intensity 1.06 microm laser irradiation produced a moderate reduction in pain and improvement in function in patients with musculoskeletal low back pain. Benefits, however, were limited and decreased with time. Further research is warranted.