RESUMEN
OBJECTIVE: This study aimed to examine the presence of distinct trajectories of adherence to home-based exercise recommendations among people with low back pain (LBP). This study also aimed to identify differences in baseline characteristics among groups. METHODS: This study was a secondary analysis of a prospective, multicenter cluster randomized controlled trial investigating the cost-effectiveness of a stratified blended physical therapist intervention compared to usual care physical therapy in patients with LBP. The intervention group received usual care with integrated support via a smartphone app. A total of 208 patients were recruited from 58 primary care physical therapist practices. Baseline data included patient characteristics, physical functioning, pain intensity, physical activity, fear avoidance, pain catastrophizing, self-efficacy, self-management ability, and health-related quality of life. The Exercise Adherence Scale (score range = 0-100) was used to measure adherence during each treatment session. Latent class growth analysis was used to estimate trajectories of adherence. RESULTS: Adherence data were available from 173 out of 208 patients (83%). Data were collected during an average of 5.1 (standard deviation [SD] = 2.5) treatment sessions, with total treatment duration of 51 (SD = 41.7) days. Three trajectory classes were identified: "declining adherence" (12%), "stable adherence" (45%), and "increasing adherence" (43%). No differences in baseline characteristic were found between groups. CONCLUSION: Three adherence trajectories to exercise recommendations were identified in patients with LBP. However, baseline characteristics cannot identify a patient's trajectory group. IMPACT: Despite the presence of distinct trajectories of adherence in patients with LBP, physical therapists should not attempt to place a patient in a trajectory group at the start of treatment. Instead, adherence should be closely monitored as treatment progresses and supported when required as part of an ongoing process.
