Comparison of 3 manual therapy techniques in individuals with low back pain who meet a clinical prediction rule


Davenport: 0000-0001-5772-7727

Document Type


Publication Title










First Page


Last Page


Publication Date



Study Design. Randomized clinical trial.

Objective. The purpose of this randomized clinical trial was to examine the generalizability of 3 different manual therapy techniques in a patient population with low back pain that satisfy a clinical prediction rule (CPR).

Summary of Background Data. Recently a CPR that identifies patients with LBP who are likely to respond rapidly and dramatically to thrust manipulation has been developed and validated. The generalizability of the CPR requires further investigation.

Methods. A total of 112 patients were enrolled in the trial and provided demographic information and completed a number of self-report questionnaires including the Oswestry Disability Questionnaire (ODQ) and the Numerical Pain Rating Scale (NPRS) at baseline, 1-week, 4-weeks, and 6-months. Patients were randomly assigned to receive 1 of the 3 manual therapy techniques for 2 consecutive treatment sessions followed by exercise regimen for an additional 3 sessions. We examined the primary aim using a linear mixed model for repeated measures, using the ODQ and NPRS as dependent variables. The hypothesis of interest was the group by time interaction, which was further explored with pair-wise comparisons of the estimated marginal means.

Results. There was a significant group x time interaction for the ODQ ( P < 0.001) and NPRS scores ( P = 0.001). Pair-wise comparisons revealed no differences between the supine thrust manipulation and side-lying thrust manipulation at any follow-up period. Significant differences in the ODQ and NPRS existed at each follow-up between the thrust manipulation and the nonthrust manipulation groups at 1-week and 4-weeks. There was also a significant difference in ODQ scores at 6-months in favor of the thrust groups.

Conclusion. The results of the study support the generalizability of the CPR to another thrust manipulation technique, but not to the nonthrust manipulation technique that was used in this study. In general, our results also provided support that the CPR can be generalized to different settings from which it was derived and validated. However, additional research is needed to examine this issue.