OBJECTIVE

To study the relationship between the methodological quality and other characteristics of reviews of spinal manipulation for low back pain on the one hand and the reviewers' conclusions on the effectiveness of manipulation on the other hand.

DATA SOURCES

Reviews identified by MEDLINE search, citation tracking, library search, and correspondence with experts.

STUDY SELECTION

English- or Dutch-language reviews published up to 1993 dealing with spinal manipulation for low back pain that include at least two randomized clinical trials (RCTs).

DATA EXTRACTION

Methodological quality was assessed using a standardized criteria list applied independently by two assessors (range, 0% to 100%). Other extracted characteristics were the comprehensiveness of the search, selective citation of studies, language, inclusion of non-RCTs, type of publication, reviewers' professional backgrounds, and publication in a spinal manipulation journal or book. The reviewers' conclusions were classified as negative, neutral, or positive.

DATA SYNTHESIS

A total of 51 reviews were assessed, 17 of which were neutral and 34 positive. The methodological quality was low, with a median score of 23%. Nine of the 10 methodologically best reviews were positive. Other factors associated with a positive reviewers' conclusion were review of spinal manipulation only, inclusion of a spinal manipulator in the review team, and a comprehensive literature search.

CONCLUSIONS

The majority of the reviews concluded that spinal manipulation is an effective treatment for low back pain. Although, in particular, the reviews with a relatively high methodological quality had a positive conclusion, strong conclusions were precluded by the overall low quality of the reviews. More empirical research on the review methods applied to other therapies in other professional fields is needed to further explore our findings about the factors related to a positive reviewers' conclusion.