Whether preparing turkey, Chanukah candles, or comet-viewing telescopes this long weekend, let us all be thankful for not being burned in the flames of the recent cholesterol guideline controversy. The problems surrounding the new ACC/AHA guidelines for reducing heart disease risk through cholesterol-lowering pharmacologic treatment and the accompanying risk calculator for heart disease are complex and not yet completely explicated. One contributing problem, however, appears to be the decision by the guideline developers to rely primarily on evidence from randomized controlled trials while discounting powerful evidence from observational studies. The controversy provides an important reminder about the limited external validity that trials often bring and the corresponding expansive external validity that well-designed observational and quasi-experimental studies can offer – particularly when crafting guidelines and risk prediction models designed for population-level health impact.
Perhaps this guideline kerfuffle provides an opportunity to focus more scientific and professional attention on the growing array of non-pharmacologic and non-clinical strategies for heart disease prevention. After all, credible evidence indicates that clinical therapeutics and public health interventions each contributed roughly equal shares to the 50% reduction in U.S. heart attack mortality between 1980 and 2000. Empirical evidence continues to mount about the effectiveness of public health programs and policies in improving physical activity and healthy eating, two powerful behavioral mechanisms for reducing heart disease risks (which are recognized in other ACC/AHA guidelines). And we have a promising national campaign underway now, the Million Hearts Initiative, that promotes a balanced prevention strategy using both clinical and community-level interventions.
To be sure, we need more and stronger evidence about the population-level health and economic effects of these types of public health approaches, and about the organizational and financing strategies that work best to scale and spread these strategies to reach the population groups at greatest risk. And we need more head-to-head comparisons of how clinical prevention strategies perform alone and in combination with community-based strategies (CER studies in public health). This is where our work in public health services and systems research (PHSSR) comes into play, including empirical studies in public health economics. This type of research, perhaps, can help to reduce the errors in models and in guidelines that crop up when taking a too-narrow view of the types of evidence that are relevant for shaping heart disease prevention strategies.