Context: Compensation, whether through workers' compensation or through litigation, has been associated with poor outcome after surgery; however, this association has not been examined by meta-analysis.
Objective: To investigate the association between compensation status and outcome after surgery.
Data sources: We searched MEDLINE (1966-2003), EMBASE (1980-2003), CINAHL, the Cochrane Controlled Trials Register, and reference lists of retrieved articles and textbooks, and we contacted experts in the field.
Study selection: The review included any trial of surgical intervention in which compensation status was reported and results were compared according to that status. No restrictions were placed on study design, language, or publication date. Studies were selected by 2 unblinded independent reviewers.
Data extraction: Two reviewers independently extracted data on study type, study quality, surgical procedure, outcome, country of origin, length and completeness of follow-up, and compensation type.
Data synthesis: Two hundred eleven studies satisfied the inclusion criteria. Of these, 175 stated that the presence of compensation (workers' compensation with or without litigation) was associated with a worse outcome, 35 found no difference or did not describe a difference, and 1 described a benefit associated with compensation. A meta-analysis of 129 studies with available data (n = 20,498 patients) revealed the summary odds ratio for an unsatisfactory outcome in compensated patients to be 3.79 (95% confidence interval, 3.28-4.37 by random-effects model). Grouping studies by country, procedure, length of follow-up, completeness of follow-up, study type, and type of compensation showed the association to be consistent for all subgroups.
Conclusions: Compensation status is associated with poor outcome after surgery. This effect is significant, clinically important, and consistent. Because data were obtained from observational studies and were not homogeneous, the summary effect should be interpreted with caution. Compensation status should be considered a potential confounder in all studies of surgical intervention. Determination of the mechanism for this association requires further study.