Objective: There is a common perception that injury compensation has a negative impact on health status, and systematic reviews supporting this thesis have been used to influence policy and practice decisions. This study evaluates the quality of the empirical evidence of a negative correlation between injury compensation and health outcomes, based on systematic reviews involving both veri. able and non-veri. able injuries. Design: Systematic meta-review (a "review of reviews''). Data sources: PubMED, CINAHL, EMBASE, PEDro, PsycInfo, EconLit, Lexis, ABI/INFORM, The Cochrane Library, and the AHRQ EPC were searched from the date of their inception to August 2008, and hand searches were conducted.
Review methods: Selection criteria were established a priori. Included systematic reviews examined the impact of compensation on health, involved adults, were published in English and used a range of outcome measures. Two investigators independently applied standard instruments to evaluate the methodological quality of the included reviews. Data on compensation scheme design (i.e., the intervention) and outcome measures were also extracted. Results: Eleven systematic reviews involving veri. able and non-veri. able injuries met the inclusion criteria. Nine reviews reported an association between compensation and poor health outcomes. All of them were affected by the generally low quality of the primary (observational) research in this field, the heterogeneous nature of compensation laws (schemes) and legal processes for seeking compensation, and the difficulties in measuring compensation in relation to health. Conclusion: Notwithstanding the limitations of the research in this field, one higher quality review examining a single compensation process and relying on primary studies using health outcome (rather than proxy) measures found strong evidence of no association between litigation and poor health following whiplash, challenging the general belief that legal processes have a negative impact on health status. Moves to alter scheme design and limit access to compensation on the basis that it is "bad for health'' are therefore premature, as evidence of such an association is unclear. (C) 2009 Elsevier Ltd. All rights reserved.