Treatment and exercise strategies and their associations with pain and disability: a prospective cohort study of patients with long-lasting low back pain

Objectives To describe self-reported treatment and exercise strategies among patients with long-lasting low back pain (LBP) 1 month after consultation at a specialised hospital-based Medical Spine Clinic and evaluate their associations with changes in pain and disability 1 and 3 months after consultation. Design Prospective cohort study using questionnaire data before consultation (baseline) and 1 and 3 months after consultation. Setting Specialised hospital-based Medical Spine Clinic, Denmark. Participants 1686 patients with long-lasting LBP completed the baseline questionnaire; 908 patients responded at 1 month, of them 623 responded at 3 month. Interventions Patients were categorised by treatment (physiotherapy, chiropractic treatment, physiotherapy+chiropractic treatment and no recommended treatment) and exercise strategy (exercise continued, exercise ceased, exercise initiated and not exercising). Primary and secondary outcome measures Pain was assessed by the numeric rating scale (NRS: 0-10), and disability was assessed by the Oswestry disability index (ODI: 0-100). Results 1-month postconsultation, half of the patients received no recommended treatment; most others received physiotherapy (42%). Nearly half of the patients continued exercise, 28% continued to be inactive, and 22% initiated exercise. For the population as a whole, pain changed by -0.74 (95% CI -0.90; -0.58) and -1.02 (95% CI -1.22; -0.83) points on the NRS at 1- and 3-month follow-up, respectively, and disability by -2.65 (95% CI -3.51; -1.78) and -4.48 (95% CI -5.59; -3.38) points on the ODI. Differences between treatment strategies were small. However, the two groups not exercising improved less compared with those who continued exercise when adjusted for age, sex and baseline level (order of magnitude from 0.07 to 1.18 points on the NRS and from 4.01 to 9.08 points on the ODI). For pain, these group differences were statistically significant at 1 month (p<0.001 for the exercise ceased group and p=0.01 for those not exercising) but not at 3 months (p=0.87, respectively, p=0.21). For disability, the differences were statistically significant both at 1 month (p<0.001 for both groups) and 3 months (p=0.03 and <0.01). Conclusions Mean improvement was negligible, with no differences between treatment strategies. However, patients not exercising showed no or less improvement, highlighting the importance of exercise in managing long-lasting LBP.