Mammography and Early Breast Cancer Detection by Alan B. Hollingsworth M.D

Author:Alan B. Hollingsworth, M.D.
Language: eng
Format: epub
Publisher: McFarland & Company, Inc., Publishers
Published: 2016-08-04T16:00:00+00:00

In the great hierarchy of evidence, none of these approaches will equal the level of evidence achieved through a prospective, randomized trial. Yet are they to be dismissed entirely? If so, why did investigators bother doing the research in the first place? And, just as meta-analyses of prospective RCTs can claim many notches for its belt, there are notches for the observational studies as well. For instance, it was a case-control study that first demonstrated the link between tobacco smoking and lung cancer.

In a case-control study, one group with a “condition or treatment” is compared to a “hand-picked” group that is intended as comparable in all respects, but without the “condition or treatment.” When the measured outcome in a case-control study is death rate (as opposed to survival rate) one actually circumvents three of the Four Horsemen: Lead Time Bias, Length Bias and Overdiagnosis. The remaining weakness is selection bias. That is, how was the control group selected and how do the participants match up to those in the treatment group? Some case-control studies will use larger numbers in their control group than their treatment group, or even use two or more control groups, to bolster claims of validity. In truth, however, only the randomization process throttles selection bias.

There is often a temporal aspect to observational studies as well, in that they are sometimes performed as a prelude to prospective RCTs, serving as the theoretical basis upon which a prospective randomized trial is justified, organized, and funded. It is impossible, however, to perform prospective RCTs for everything we do in medicine, so observational studies, or circumstantial evidence, is all we’ll ever have for most of what we do.

There are hundreds and hundreds of these observational studies related to mammographic screening and the early detection of breast cancer. It requires much of a lifelong career to assimilate the information. Is it any wonder that the Task Force relegates them to a cursory afterthought, given that this same Task Force is trying to keep up with nearly 100 other preventive health measures?

In this battle over screening mammography, the weaponry is based in numbers. These numbers may appear to have been forged into the tempered steel of sophisticated statistical analysis, but they are more accurately described as plastic. Recall how MD Anderson’s emeritus professor of radiology, the late Dr. Gerald Dodd, Jr., warned us about the encroachment of non-clinical forces into the practice of medicine? Today, we are at the mercy of our non-clinical colleagues who toy with the numbers on their fingertips. Task Force members, by the way, are mostly clinicians who have a special interest and training in epidemiology and public health. But even the Task Force will outsource much of the number games to biostatisticians. So an important question to ask is “Who pays the biostatisticians, those who torture the data until it talks?”

Of the biases other than the Four Horsemen, one that gets the spotlight today is financial bias. Clinicians who present their papers at scientific meetings are now

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.