ON RECONSIDERING COMMUNITY-BASED HEALTH PROMOTION

Merzel and D’Afflitti are to be commended for their excellent overview of community-based health promotion studies.1 The human immunodeficiency virus (HIV) prevention studies are a model for possible successful community intervention programs. These “successful” community intervention programs to change HIV risk behavior are not unique. They represent the classic “shoe leather” epidemiological prevention approach for control of sexually transmitted diseases by identifying high-risk populations, mode of transmission of the disease, recruitment of local or neighborhood advocates, and a defined intervention strategy.

There have been previous successful community-based public health programs to control chronic diseases that serve as models for successful programs. Perhaps the earliest and most successful was the elimination of pellagra by dietary modifications in high-risk communities by Goldberger et al. in the early 1900s.2 Community-based tuberculosis programs were very successful in delivering isoniazid and other therapies to high-risk communities to reduce the number of new cases.3 Community-based programs were successful in reducing the risk of rheumatic heart disease and recurrent rheumatic fever among children with rheumatic fever even in very high-risk communities.4

The Hypertension and Detection Follow-up Program in the 1980s was a very successful community-based randomized trial to test antihypertensive therapy and reduction of stroke, congestive heart failure, coronary heart disease, and total mortality. The trial used health counselors to work with participants and demonstrated substantial improvement in adherence to antihypertensive therapy, reduction in blood pressure, and decrease in risk of vascular disease even among the lowestincome minority participants.5,6

These successful community studies have several characteristics in common: (1) a strong public health and preventive medicine science base7,8; (2) selection of an appropriate population in which to implement the intervention (i.e., common-source population, total population, or selected higher-risk subsample of population); (3) use of an intervention that has been shown to have an impact (efficacy) on defined measurable outcome and that will have a meaningful impact on the health of at-risk populations; (4) sufficient funding to provide an adequate, well-trained, and dedicated field staff to deliver the effective therapies; and (5) community support. I would strongly suggest a careful review of these successful public health programs of the past before we focus on new behavioral theories for community programs.9

Finally, the statement that these programs do not require evaluations with regard to reduction of incidence or mortality of disease or objective evidence of changes in risk factors is a serious mistake. Such an assumption could be the death knell of community-based public health programs in the future, especially since there is now greater emphasis on evidence-based outcomes as criteria for support.8,10 Unfortunately, many programs have good processes but are too weak to have any meaningful objective outcomes.