Charles Krauthammer today delivers a useful dose of cold water on the fantasy that preventive care saves money. There are a few measures that probably do save money; mammography is one. Mammography does not, obviously, prevent breast cancer but it does detect the disease at an early stage and probably saves money by avoiding the treatment of late incurable breast cancer in more women. Kaiser, an HMO with a loyal membership, was a pioneer in routine mammography and this is the most likely incentive. More on the issue of preventive care and cost here.
This brings up the difference between prevention and screening for early diagnosis. It also raises the issue of cost-benefit analysis in medicine. I have spent some time studying this subject. The analysis of such issues is called decision theory. It is very commonly used in business to decide on which investments are more likely to make money. The decision tree typically uses software that calculates the costs associated with all possible decisions that could be made. An example is here. The decision tree begins with an initial set of facts, the simpler the better. In medicine, it might be a symptom like rectal bleeding.
A patient comes in with rectal bleeding. How do we proceed ? I used to teach this sort of thing to medical students. Cost-benefit analysis can be part of this process. For example, it sometimes seems to me that junior doctors choose a spiral CT scan as the initial diagnostic event far too often. I know that one third or more of cardiology fellows, graduate physicians in training, cannot hear a systolic heart murmur. This is the most basic examination in cardiology. The stethoscope was the first instrument that allowed doctors to examine the interior of the body. Yet one-third of budding cardiology specialist are incompetent with it. Why ? They have spent years ordering echocardiograms on every patient with the least suspicion of heart disease. Here is a useful role for cost-benefit analysis.
I don’t want to conduct a seminar on this subject. If you want one, there are journal articles going back 30 years on the topic. This is where we get into Bayesian Algebra. Anyone who gets into this subject must understand the basic principles of diagnostic tests, true positive, false positive, true negative and false negative. A true positive test is positive in a patient who has the disease. False positive is, of course, positive in a patient who does not have the disease. Two groups of patients have the disease, true positives and false negatives. From this sort of thing, one calculates the sensitivity and specificity of tests. A test is the most sensitive that has the most true positives. A test that is the most specific, has the most true negatives.
That, of course, has nothing to do with prevention but prevention, aside from immunization, is not medicine anyway. It is public health plus cultural norms. A century after Columbus’ first voyage brought syphilis back to Europe, one third of the residents of Paris had the disease. The importance of virginity in a bride may have arisen from such data.
I have serious doubts that 98% of the advocates of health reform would have any idea what I’m talking about here.
Tags: health care, medicine, science
That is so good to know! I am glad people care enough to do something helpful in this situation.
[…] post: A dose of cold water « A Brief History… Tagged as: 2009-at-859-pm, brief-history, disease, Display, europe, health, health-reform, […]