Citation

Spearing NM, Connelly LB. Spine 2011; 36(25 Suppl): S303-8.

Affiliation

Australian Centre for Economic Research on Health (ACERH UQ), The University of Queensland, Edith Cavell Building, RBWH, Herston Queensland Australia 4029 Professor of Health Economicsk, Australian Centre for Economic Research on Health (ACERH UQ), Centre of National Research on Disability and Rehabilitation Medicine, The University of Queensland, Edith Cavell Building, RBWH, Herston Queensland Australia 4029.

Copyright

(Copyright © 2011, Lippincott Williams and Wilkins)

DOI

10.1097/BRS.0b013e3182388411

PMID

22020600

Abstract

Study Design. Review article.Objective. To explain why the evidence that compensation-related factors lead to worse health outcomes is not compelling, either in general, or in the specific case of whiplash.Summary of Background Data. There is a common view that compensation-related factors lead to worse health outcomes ("the compensation hypothesis"), despite the presence of important, and unresolved, sources of bias. The empirical evidence on this question has ramifications for the design of compensation schemes.Methods. Using studies on whiplash, this paper outlines the methodological problems that impede attempts to confirm or refute the compensation hypothesis.Results. Compensation studies are prone to measurement bias, reverse causation bias, and selection bias. Errors in measurement are largely due to the latent nature of whiplash injuries and health itself, a lack of clarity over the unit of measurement (specific factors, or "compensation"), and a lack of appreciation for the heterogeneous qualities of compensation-related factors and schemes. There has been a failure to acknowledge and empirically address reverse causation bias, or the likelihood that poor health influences the decision to pursue compensation: it is unclear if compensation is a cause or a consequence of poor health, or both. Finally, unresolved selection bias (and hence, confounding) is evident in longitudinal studies and natural experiments. In both cases, between-group differences have not been addressed convincingly.Conclusion. The nature of the relationship between compensation-related factors and health is unclear. Current approaches to testing the compensation hypothesis are prone to several important sources of bias which compromise the validity of their results. Methods that explicitly test the hypothesis and establish whether or not a causal relationship exists between compensation factors and prolonged whiplash symptoms are needed in future studies.

Language: en