In 2007, an estimated 1,444,920 Americans will be diagnosed with cancer and about 559,650 will die from the disease, making cancer the second leading cause of death after heart disease.[1] These estimates do not include noninvasive cancers of any site or most skin cancers, which by themselves amount to almost 1 million cases.[1]

The cancer burden is unevenly distributed in the population, and cancer is considered a disease of the elderly. Patients aged 65 years and older represent only 12% of the population, but they account for 61% of new cancer cases. About 77% of all cancer cases are diagnosed in people over age 55.[1]

Costs of Cancer

Apart from the human toll, there are substantial financial costs associated with cancer. The overall cost of cancer care in 2006 was US$206.3 billion, of which $78.2 billion was the total health expenditure, $17.9 billion was the indirect morbidity cost, and $110.2 billion was attributed to indirect mortality.[1] The costs associated with the management of cancer in the older patient have also been explained in detail in the paper by McKoy and colleagues published in this issue of the journal.[2] These cost considerations are based on the foundation of current medical oncology practice. The critical question is: How much of this practice, and consequently cost associated with it, is informed by reliable evidence?

Research conducted during the past decade convincingly demonstrated a tremendous variation in the practice of medicine—so much so that it is not uncommon for some doctors to undertake a "wait-and-see" policy in a given clinical situation, while in the same setting, others might order a diagnostic test, and still others initiate treatment without testing.[3-7] Many factors—from the structure of local care to the availability of diagnostic technologies to financial incentives—have been invoked in explaining these practice variations in medicine.[3-7] Seen in both the diagnostic and treatment settings, such variations typically result in too little care (underuse, failure to provide an effective service when it would have produced a favorable outcome); too much care (overuse, provision of care when its risks of harm exceed the potential benefits); and the wrong care (misuse, avoidable complications of appropriate care).[8]

Underrepresentation in Trials

Failure to generate and then act on reliable evidence is considered one of the major reasons for suboptimal delivery of modern health-care treatments and other services. Unfortunately, evidence to support the practice of oncology in elderly patients is extremely limited. This is not surprising when only 3% to 5% of all eligible adult cancer patients participate in clinical trials. How can we develop effective interventions for the prevention and treatment of cancer in the elderly if they do not participate in clinical trials, particularly in randomized clinical trials (RCTs), widely considered the most reliable form of medical evidence?[9,10] Practice guidelines, which are ideally based on a systematic review of all RCTs, cannot be developed if the evidence is nonexistent or lacking.[9]

Given the fact that elderly patients are underrepresented in cancer RCTs, the evidence base to guide both current and future prevention and treatment strategies for the elderly is clearly limited. Consequently, we cannot know how much overuse, underuse, or misuse is a part of contemporary medical practice. This also implies that the existing economic analyses really do not measure costs of the "right" medical care but, to a large extent, represent a mixture of overuse, underuse, or misuse of health interventions employed in oncology practice.

Guidelines published by the US Food and Drug Administration for the study of drugs likely to be used in elderly patients state that "drugs should be studied in all age groups, including the geriatric, for which they will have significant utility."[11] Despite these guidelines and the high prevalence of cancer in the elderly population, however, data from several studies show that elderly patients are not adequately represented in research efforts to find innovative treatments for the prevention and treatment of cancer. Again, this makes the applicability of such research findings in the elderly doubtful.[12-15]

Generalizability of Clinical Trials