Predictors of successful cancer prevention programs

Finding the optimal use of health-care resources requires the reliable estimation of costs and consequences. Acquiring these estimates may not be difficult for some common treatments. More difficult is the optimization of resources in the area of diagnostics. Only a few attempts have been made to optimize the use of resources in the area of prevention. Several aspects have to be considered when optimizing the resources for prevention: (1) participation rates in structured prevention programs are low, (2), acquiring data on follow-up and outcomes is difficult, (3) there are concerns about the quality of information available to public, and (4), the public is often unaware of scientific assessments of prevention programs. As prevention programs are costly long-term projects, a strategy to select these programs according to possible predictors of success might be useful. The few analyses of cancer prevention in the literature have been directed towards the most common malignant diseases (as assessed by incidence) such as cancer of the breast, colon, lung and prostate. We argue that incidence is a poor marker for selecting secondary prevention programs. Incidence may be a misleading indicator for two reasons: incidence of disease does not predict efficiency of management or good health outcomes, and incidence does not separate clinically significant from non-significant disease. The traditional strategy is based on the assumption that more screening increases the chance of cure. We propose an alternative outcomes model that suggests better disease management justifies new prevention programs. Indicators for better disease management are effective and efficient treatments as well as high-quality screening (sensitivity and specificity) techniques and possibly "side-effects of prevention programs," which provide early signs of success to motivate the patient's participation, to keep up with the program and finally to succeed.