JOINT ANALYSIS OF 3 US COMMUNITY INTERVENTION TRIALS FOR REDUCTION OFCARDIOVASCULAR-DISEASE RISK

During the 1980s three comprehensive community-based heart disease pre vention trials were conducted in the United States. The Stanford Five- City Project, Minnesota Heart Health Program, and Pawtucket Heart Heal th Program involved 12 cities; six received a 5-8 year multifactorial risk reduction program. This analysis pools data from the three studie s to delineate the common intervention effects with greater sample siz e and power than could be attained by the single studies. Time trends were estimated for cigarette smoking, blood pressure, total cholestero l, body mass index, and coronary heart disease mortality risk in women and men aged 25-64 years. The joint estimates of intervention effect were in the expected direction in nine of 12 gender-specific compariso ns; however, these were not statistically significant. The results ill ustrate the analytic challenges of evaluating community-based preventi on trials and point to the smaller than expected net differences, rath er than small sample size, as the reason for few statistically signifi cant effects in the three U.S. prevention trials. (C) 1997 Elsevier Sc ience Inc.