The STarT Back Screening Tool (SBST) was developed to screen and to classify patients with low back pain (LBP) into subgroups for the risk of having a poor prognosis. However, this classification at baseline does not take into account variables that can influence the prognosis during treatment or over time. The objectives:
- To investigate the changes in the risk subgroup measured by the SBST over a period of 6 months
- To assess the long-term predictive ability of the SBST when administered at different time points.
Patients with chronic nonspecific LBP (n=148) receiving physical therapy care as part of a randomized trial were analyzed. Pain intensity, disability, global perceived effect and the SBST were collected at baseline, 5 weeks, 3 and 6 months. Changes in the SBST risk classification were calculated. Hierarchical linear regression models adjusted for potential confounders were built to analyze the predictive capabilities of the SBST when administered at different time points. A large proportion of patients (60.8%) changed their risk subgroup after receiving physical therapy care. The SBST improved the prediction for all 6-month outcomes when using the 5 weeks risk subgroup and the difference between baseline and 5 week subgroup, after controlling for potential confounders. The SBST at baseline did not improve the predictive ability of the models after adjusting for confounders.
This study shows that many patients change SBST risk subgroup after receiving physical therapy care and the predictive ability of the SBST in patients with chronic LBP increases when administered at different time points.