A meta-analysis of studies of the Health Belief Model in Taiwan

Abstract

Health Belief Model (HBM) has been used for near twenty years in Taiwan. Although more then 100 studies have been published, the effect of HBM in health behavior is still unknown. The purpose of this study is to apply meta-analysis to determine the effect sizes of HBM and the relationship between four main variables and health behaviors in studying health behaviors in Taiwan. 　　The researcher will use computerized bibliographic databases and the ancestry approaches to search for studies that were designed based on HBM, conducted in Taiwan and published from 1974 to 2003. For data analysis, the coding sheet and codebook will be developed by the research to collect required information from each recruited study. Comprehensive Meta- Analysis software designed by Borenstein & Rothstein in 1999 and Rosenthal’s method will be used to analyze the data. The results show: (1) the total explanatory of four main variables in HBM is 22% which achieves significant level; (2) the effect of HBM in health behavior has significant differences while different sources, research designs, types of behavior, ways of performance, and types of independent variables; (3) sources, research designs, types of behavior, types of dependent variables, and types of independent variables need to rely on the interactions with other variables to moderate effect sizes; (4) no matter what sources, research designs, types of behavior, ways of performance, and types of independent variables are, perceived barriers of taking action are the strongest variable. 　　Conclusion: HBM can apply to explain or predict health behaviors in Taiwan. This study suggests to adding cues and self-efficacy in future studies to increase whole explanatory or prediction. Moreover, perceived barriers of taking action are the strongest variable. Therefore, professional should remove perceived barriers of taking action to increase health behaviors.

Date of Award	2004
Original language	English
Supervisor	Wei-Ming Luh (Supervisor)