How big is the effect of spinal manipulation on the pressure pain threshold and for how long does it last? - secondary analysis of data from a systematic review.

Background: Spinal manipulation (SM) has been shown in a systematic review to have a statistically significant effect on the pressure pain threshold (PPT) in asymptomatic subjects, when SM is compared to a sham intervention. The magnitude and duration of this effect is unclear. Objectives: To determine the effect-size of SM in asymptomatic subjects and its duration. Method: This is a secondary analysis of data from a previous review. We sought to compare the effect-sizes in the various articles but had to calculate them ourselves, at different follow-up time measurements. Effect-sizes (Cohen's d or Hedge's g coefficient) were considered low, medium, and large, at the cut points of 0.2, 0.5, and 0.8, respectively. Results: Effect-sizes were reported in 6/8 studies, but all had calculated 'within-group' changes, not 'between-group' differences. Immediately after SM, only one study of four (with four measurements) had a statistically significant 'medium' effect size (d = 0.56; 95% CI: 00.4-1.08 to d = 0.70; 95% CI:0.18-1.22). Five minutes after SM, 4/5 studies found a statistically significant 'medium to large' effect-size (d = 0.51; 95% CI: 0.04-0.98 to d = 1.24; 95% CI: 0.28-2.20). Ten minutes after SM, two studies reported a 'medium' effect-size with statistical significance (d = 0.58; 95% CI: 0.11-1.05 to d = 0.80; 95% CI: 0.12-1.48). We drew no conclusions for the effect-sizes at one minute and thirty minutes after SM, as no between-group statistical difference was found. Conclusion: Authors need to revise their approach to 'effect size'. Our calculations showed that the effect-size of SM on PPT may go from 'medium' to 'large' within the first five minutes but appears to diminish again within ten minutes. Research of this type should collect information for longer periods and compare results to other interventions to put results into perspective.