Retention of blinding at follow-up in a randomized clinical study using a sham-control cervical manipulation...

Participants in clinical trials of spinal manipulation have not been rigorously blinded to group assignment. This study reports on secondary analyses of the retention of participant blinding beyond the immediate posttreatment time frame following a single-session, randomized clinical study. A novel control cervical manipulation procedure that has previously been shown to be therapeutically inert was contrasted with a typical manipulation procedure.

A randomized clinical study of a single session of typical vs sham-control manipulation in patients with chronic neck pain was conducted. Findings of self-reported group registration at 24 to 48 hours posttreatment were computed. The Blinding Index (BI) of Bang et al was then applied to both the immediate and post-24- to 48-hour results.

Twenty-four to 48 hours after treatment, 94% and 22% of participants in the typical and control groups, respectively, correctly identified their group assignment. When analyzed with the BI of Bang et al, the immediate posttreatment BI for the group receiving a typical manipulation was 0.22 (95% confidence interval [CI], -0.03 to 0.47); for the group receiving a control manipulation, it was 0.19 (95% CI, -0.06 to 0.43). The BI at post-24 hours was as follows: typical = 0.75 (95% CI, 0.59-0.91) and control = -0.34 (95% CI, -0.58 to -0.11).

This study found that the novel sham-control cervical manipulation procedure may be effective in blinding sham group allocation up to 48 hours posttreatment. It appears that, at 48 hours posttreatment, the modified form of the typical cervical manipulation was not. The sham-control procedure appears to be a promising procedure for future clinical trials.