Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Health Beliefs, Disease Severity, and Patient Adherence: A Meta-Analysis

DiMatteo, M Robin PhD; Haskard, Kelly B. MA; Williams, Summer L. MA

doi: 10.1097/MLR.0b013e318032937e
Original Article

Background: A large body of empirical data exists on the prediction of patient adherence from subjective and objective assessments of health status and disease severity. This work can be summarized with meta-analysis.

Objectives: Retrieval and summary analysis of r effect sizes and moderators of the relationship between patient adherence and patients’: (1) beliefs in disease threat; (2) rated health status (by physician, self, or parent); and (3) objective disease severity.

Methods: Comprehensive search of published literature (1948–2005) yielding 116 articles, with 143 separate effect sizes. Calculation of robust, generalizable random effects model statistics, and detailed examination of study diversity with moderator analyses.

Results: Adherence is significantly positively correlated with patients’ beliefs in the severity of the disease to be prevented or treated (“disease threat”). Better patient adherence is associated with objectively poorer health only for patients experiencing disease conditions lower in seriousness (according to the Seriousness of Illness Rating Scale). Among conditions higher in seriousness, worse adherence is associated with objectively poorer health. Similar patterns exist when health status is rated by patients themselves, and by parents in pediatric samples.

Conclusions: Results suggest that the objective severity of patients’ disease conditions, and their awareness of this severity, can predict their adherence. Patients who are most severely ill with serious diseases may be at greatest risk for nonadherence to treatment. Findings can contribute to greater provider awareness of the potential for patient nonadherence, and to better targeting of health messages and treatment advice by providers.

Supplemental Digital Content is Available in the Text.

From the Department of Psychology, University of California, Riverside, California.

Supported by a Robert Wood Johnson Foundation Investigator Award in Health Policy Research, by National Institute on Aging grant 5R03AG27552-02, and by the Committee on Research of the U.C. Riverside Academic Senate.

The views expressed in this paper are those of the authors alone and do not imply endorsement by the funding sources.

Reprints: M. Robin DiMatteo, PhD, Department of Psychology, University of California, 900 University Avenue, Riverside, CA 92521. E-mail:

© 2007 Lippincott Williams & Wilkins, Inc.