A prospective, randomized 5-year follow-up study of functional restoration in chronic low back pain patients

A functional restoration (FR) program, dealing with a combination of intensive physical and ergonomic training, psychological pain management, and patient education, was tested in two randomized, parallel group studies. In one of these patients following the FR program were compared with a non-treated control group (project A), and in the other with patients on two less intensive treatment programs (project B). A total of 238 chronic low back pain patients participated in the two studies, 106 entering project A and 132 project B. Patients from the two projects were comparable except that the patients in project A were recruited from all over the country, whereas patients in project B all were living in and around Copenhagen. Thirteen patients never started any treatment, and 20 patients (9%) dropped out during the treatment period. Of the 207 who completed treatment, 89% returned a mailed questionnaire 5 years later. This was the case for 55% of the drop-outs. The questions referred to work situation, pain level, activities of daily living, days of sick leave, contact with health care professionals, physical activity, use of medication, and a subjective overall assessment. The results show that in project A the treated group reported significantly fewer contacts with the health care system and significantly fewer days of sick leave over the 5-year follow-up period compared to the control group. In all other parameters, including work ability, there was no statistically significant difference between the two groups. In project B, patients treated in the FR program did significantly better in most measured parameters, except in leg pain, use of pain medication and sport activity, where no significant differences were found between groups. The overall result shows a positive long-term effect of the FR program, but it also shows the necessity of testing a given treatment in different projects and designs, among other things due to statistical variations.