The Evidence for Screening

Clinicians and scientists have difficulty proving the intuitive notion that early detection results in longer life, because the observation of selected cases is deceiving and ascertainment of all relevant data for an entire population is difficult. Collections of cancer cases from a practice or screening center do not represent the full spectrum of the disease, so conclusions are not valid for the condition as a whole. Furthermore, watching cases from the day of diagnosis forward almost always makes survival look better for those at early cancer stages because there is more observation time, not necessarily because the process adds years to patients' lives.

People evaluating screening tests hope to minimize bias by designing and conducting studies carefully. Randomized trials reduce concerns about the inference drawn because it is more likely that the intervention and control populations are comparable, but how the study is implemented also influences its quality. Randomization, cause-of-death assessment, selection bias, and population characteristics all can affect inferences drawn from a study, and these factors can also affect whether the conclusions apply to the general population. The risk of these biases differs with study designs and has led the medical community to rank the design types based on the ability to draw valid conclusions. The highest quality results are likely to come from well-conducted randomized trials because they create similar groups and allow the assign ment of an intervention to one of the groups so that selection bias is avoided.14 One persistent concern, however, is that people willing to be in a randomized trial differ from people in the general population. Data from case-control studies are rated below randomized trials but allow for assessment of factors associated with rare conditions. The limitation of this design is that conclusions are affected by how well the cases and controls reflect similar populations and whether all potential confounders and effect modifiers have been identified.15 Despite being considered of lower quality, observational studies can provide powerful insights when conducted where comprehensive data are collected for entire popula-tions.16-19 Such studies are especially important for evaluating whether the impact of screening in practice is consistent with expectations from randomized trials. In fact, some recent analyses show that well-designed and carefully conducted observational studies of interventions may obtain results that are indistinguishable from randomized trials.19,20 Given the rapid progression of technology, observational studies are a necessary part of evaluating new technologies that may be implemented in practice before randomized trials can be conducted.

The following section summarizes the evidence on screening for breast, cervical, colon, and prostate cancers, covering the following topics: (1) epidemiology and natural history; (2) screening test characteristics; (3) review of the literature on benefit; and (4) summary recommendations from the American Cancer Society (ACS) and the U.S. Preventive Services Task Force (USPSTF), the organizations to which U.S. primary care practitioners look most often for recommendations.6,10 Specialty societies such as the American Academy of Family Physicians, American Academy of Obstetrics and Gynecology, American Academy of Physicians, and American College of Gastroenterology have also reviewed the evidence. We have included the recommendations that most strongly influence primary care, recognizing that providers are also strongly influenced by their patients and colleagues.21

The test characteristics discussed in this chapter include sensitivity, specificity, and positive predictive value. "Sensitivity" is the ability of a test to find the screened-for condition when it is present. "Specificity" is the ability of the test to determine that the screened-for condition is absent when the screened person is disease free. The "positive predictive value" (PPV) of a test is the proportion of positive tests due to the condition of interest.

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook


Post a comment