Abstract
Purpose
The importance of interpreting clinical trial results in terms of the benefits a treatment may offer to individuals with chronic pain is becoming more widely recognized. The clinical meaningfulness of group differences can better be described by looking at the percentages of responders in each treatment group, rather than between group mean differences. We have reassessed the outcomes of a clinical trial for chronic low back pain (LBP) from this new perspective.
Methods
The randomized clinical trial examined short- (12 weeks) and long-term (52 weeks) efficacy of high-dose, supervised trunk exercise (SET), spinal manipulative therapy (SMT), and a course of home exercise and self-care advice (HEA) for the treatment of LBP ≥ 6 weeks duration. We calculated response to care at 2 levels for 2 variables: numeric back pain scores (NRS) and Roland-Morris Disability (RMD), and at 3 timepoints (Weeks 12, 26 and 52). The 2 levels were calculated as percent improvement from baseline ≥ 30% and ≥75%. Finally, we calculated the relative proportion (± 95% confidence intervals) of the sample (∼100 per treatment group) that achieved each level of improvement.
Conclusion
In general, 60%-80% of patients had ≥ 30% improvements in pain @ week 12, while 20%-30% showed ≥75% improvement. At week 52, the percentage of patients with ≥ 30% improvement dropped slightly, while ≥75% improvement increased slightly. For RMD scores, 30%-50% of patients showed ≥75% improvement at week 52. We will present these results and compare them to the previous analysis of between group mean differences.
Conflicts of interest: None
Sources of funding: United States Health Resources and Services Administration
