Association between compensation status and outcomes in spine surgery: a meta-analysis of 31 studies

BACKGROUND CONTEXT: Numerous studies have demonstrated poorer outcomes in patients with Workers' compensation (WC) when compared with those without WC following treatment of various of health conditions, including spine disorders. It is thus important to consider compensation status when assessing treatment outcomes in spine surgery. However, reported strengths of association have varied significantly (1.31-7.22). PURPOSE: The objective of this study was to evaluate the association of unsatisfactory outcomes on compensation status in spine surgery patients. STUDY DESIGN/SETTING: A meta-analysis was performed. PATIENT SAMPLE: Patient sample is not applicable in this study. OUTCOME MEASURE: Demographics, type of surgery, country, follow-up time, patient satisfaction, return to work and non-union events were the outcome measures. METHODS: Both prospective and retrospective studies that compared outcomes between compensated and non-compensated patients in spine surgery were included. Two independent investigators extracted outcome data. The meta-analysis was performed using Revman software. Random effects model was used to calculate risk ratio (RR, 95% confidence interval [CI]) for dichotomous variables. RESULTS: Thirty-one studies (13 prospective; 18 retrospective) with a total of 3,567 patients were included in the analysis. Follow-up time varied from 4 months to 10 years. Twelve studies involved only decompression; the rest were fusion. Overall RR of an unsatisfactory outcome was 2.12 [1.74, 2.58; p<.001] in patients with WC when compared with those without WC after surgery. The RR ofan unsatisfactory outcome in patients with WC, compared with those without, was 2.09 [1.38, 3.17]; p<.01 among studies from Europe and Australia, and 2.14 [1.48, 2.60]; p<.01 among US studies. The RR of decompression-only procedures was 2.53 [1.85, 3.47]; p<.01, and 1.79 [1.45, 2.21]; p<.01 for fusion. Forty-three percent (209 of 491) of patients with WC did not return to work versus 17% (214 of 1250) of those without WC (RR 2.07 [1.43, 2.98]; p<.001). Twenty-five percent (74 of 292) and 13.5% (39 of 287) of patients had non-union in the compensated and non-compensated groups, respectively. This was not statistically significant (RR 1.33 [0.92, 1.91]; p=.07). CONCLUSIONS: Workers' compensation patients have a two-fold increased risk of an unsatisfactory outcome compared with non-compensated patients after surgery. This association was consistent when studies were grouped by country or procedure. Compensation status must be considered in all surgical intervention studies. (C) 2015 Elsevier Inc. All rights reserved.