Does how you do depend on how you think you'll do? A systematic review of the evidence for a relation between patients' recovery expectations and health outcomes

Background: Most clinicians would probably agree that what patients think w ill happen can influence what does happen over the clinical course. Yet des pite useful narrative reviews on expectancy of therapeutic gain and the mec hanisms by which expectancy can affect health outcomes, we were unable to l ocate a systematic review of the predictive relation between patients' reco very expectations and their health outcomes. Methods: We searched MEDLINE for English-language articles published from. 1966 to June 1998 with a title or abstract containing at least 1 of the med ical subject headings (MeSH) "self-assessment," "self-concept" or "attitude to health," or the MeSH subheading "psychology," and at least I word from each of 3 sets: 'patient" and similar words; a form of "expectation," 'beli ef" or 'prediction",- and a form of "recover," 'outcome,' "survival" or "im prove." Relevant articles contained original research data, measured patien ts' recovery expectations, independently measured a subsequent health outco me and analyzed the relation between expectations and outcomes. We assessed internal validity using quality criteria for prognostic studies based on 6 categories (case definition; patient selection; extent of follow-up; objec tive outcome criteria; measurement and reporting of recovery expectations; and analysis). Results: A total of 1243 titles or abstracts were identified through the co mputer search, and 93 full-text articles were retrieved. Forty-one of these articles met the relevance criteria, along with 4 additional articles iden tified through other means. Agreement beyond chance on quality assessments of 18 randomly selected articles was high (kappa = 0.87, p = 0.001). Sixtee n of the 45 articles provided moderate-quality evidence and included a rang e of clinical conditions and study designs; 15 of the 16 showed that positi ve expectations were associated with better health outcomes. The strength o f the relation depended on the clinical conditions and the measures used. Interpretation: Consistency across the studies reviewed and the evidence th ey provided support the need for clinicians to clarify patients' expectatio ns and to assist them in having appropriate expectations of recovery. The u nderstanding of the nature, extent and clinical implications of the relatio n between expectations and outcomes could be enhanced by more conceptually driven and, methodologically sound research, including evaluations of inter vention effectiveness.