Some studies of low back pain have not shown benefits from exercise, whereas others have demonstrated that exercise is associated with a better clinical outcome. Validating the effects of exercise in patients with low back pain poses technical problems, including patient bias, the inability to conduct a blind study of exercise, the various types of exercise programs and the differences in the intensity and duration of exercise for low back pain. Moffett and colleagues performed a randomized trial of exercise for low back pain that took into account whether patients' treatment preferences influenced the outcome.

The 187 patients enrolled in the study were referred by family physicians. Inclusion criteria included an age of 18 to 60 years, back pain lasting at least four weeks but less than six months and no history of physical therapy for the back pain. Patients completed several questionnaires to assess the severity of the back pain and its impact on the patient's ability to function. Patients also completed questionnaires after six and 12 months for assessment of outcome. Most of the patients (63 percent, or 118 of 187 patients) reported a preference for exercise over routine treatment.

Patients were randomly assigned to continue the usual treatment with their family physician or to attend eight one-hour exercise classes during a four-week period. The exercise regimen incorporated stretching, strengthening and low-impact aerobic exercises to encourage normal movement of the spine. Patients were discouraged from following the precept of “let pain be your guide.” They were encouraged to improve and were rewarded with attention and praise. At each session patients were given an educational message that encouraged self-reliance. They also were told that the exercise classes should be viewed as a step toward increasing their own levels of exercise. Over 70 percent of the exercise group attended six to eight classes.

Eighty-nine patients were assigned to the exercise program and 98 were assigned to routine management. Outcomes were evaluated at six weeks in 85 patients in the exercise group and in 94 patients in the usual-treatment group. Follow-up at six months occurred in 77 patients in the exercise group and 86 patients in the control group. At one year, 83 and 88 patients in the exercise group and the usual-treatment group, respectively, were evaluated.

After six weeks, patients in the exercise group reported less distressing pain than the control group and a marginally significant difference in functional disability when compared with the control group. At six months, the difference in disability scores between the two groups was significant. At one-year follow-up, 64 percent of the exercise group and only 35 percent of the control group had improved by at least three points on the disability scale.

Exercise was calculated to save over $250 per patient in direct medical costs and to reduce the number of days lost from work by one half. Patients in the control group took 607 days off work during the 12-month period, compared with 378 days taken off by the intervention group.

The authors conclude that a simple exercise program improves the long-term outcome in patients with low back pain. The benefits observed in this study were independent of patient preference and were sustained over at least one year. The exercise program was highly acceptable to patients and cost-effective in terms of direct costs and in reducing disability.