PRINCIPLES OF CANCER SCREENING

PREVENTION AND SCREENING

The goal of any screening program is to reduce the burden of suffering from disease either in an individual or in a population. Understanding the concept of prevention is central to understanding the different ways in which the natural history of disease may be altered. Prevention may be classified as primary, secondary, and tertiary(Table 1).

When most hear the term prevention, particularly when applied to cancer, they think of finding cancer at an early stage and instituting treatment when it

WHAT SHOULD BE SCREENED FOR

Thus far, this article has dealt primarily with the benefits, risks, and costs of diagnostic and treatment strategies, rather than with specific disease characteristics that allow for successful early intervention. The prevalence of a disease is important not only for its influence on the operating characteristics of a test, but also for how it affects potential outcomes of treatment for a given population. Clinicians have the greatest opportunity to affect the health of a population if they

ABSOLUTE VERSUS RELATIVE RISK REDUCTION

In reviewing results of clinical trials, it is important to make a distinction between the relative difference between the comparison groups and the absolute difference in the event rates. A treatment that reduces mortality by only 1% in a disease state that causes 100,000 deaths saves 1000 lives, whereas a treatment that may reduce mortality by 50% in a disease causing 100 deaths results in only 50 lives saved. The relative risk reduction is the percentage reduction in the measured end point

GRADING OF EVIDENCE

Evidence from the medical literature used to support an argument for or against a screening strategy must be assessed for its validity and weight. The best experimental evidence is from well-designed, randomized, controlled trials. Random allocation before the clinical intervention eliminates bias and increases the likelihood that differences between the groups are, in fact, a result of the intervention. A cohort study is usually an observational study in which allocation of subjects to the

SYSTEMS

Although there is emerging consensus in preventive medicine regarding the priorities for prevention, there is less consensus on how best to implement preventive measures and achieve measurable results at the level of the individual practitioner or practice group. Managed-care delivery systems exert an influence on planning cost-effective approaches to both preventive care and treatment of illness. Despite physicians' awareness of and agreement with guidelines for preventive care, there is a gap

RESOURCES

The accumulation of knowledge in preventive medicine continues at a pace that makes it difficult to be familiar with all of the latest scientific developments. What resources are available to the busy practitioner to help incorporate the latest developments into daily practice? Ideally a practitioner can use the approach of the Evidence-Based-Medicine Group¹¹ and retrieve relevant, well-designed clinical trials and review the primary evidence. Although this may be time-consuming, it provides

ACKNOWLEDGMENTS

The author would like to thank Mark Levine, Dennis Plante, and Anne-Lise Jacobsen for review of the manuscript and helpful comments.