| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Point |
1 2010 Shady Ln., Green Bay, WI 54313.
Received February 18, 2003; accepted after revision May 15, 2003.
Address correspondence to R. Levatter
(dximgr{at}aol.com).
Methodologic Concern
Baker assumes that abdominal CT screening will be done without IV contrast material. It is no doubt true, for reasons of efficiency, safety, and cost, that unenhanced CT is currently the predominant technique. Similarly, at one time single-contrast barium enema was the predominant radiology technique for imaging the colon. Pointing out that single-contrast barium enemas are not sensitive examinations is not, in itself, an argument against radiologic screening of the colon. Similarly, arguing, correctly, that many things can be missed on an unenhanced screening CT of the abdomen is not an argument against abdominal screening CT in general. It is at most an argument for improving the screening protocol used in many imaging centers.
A Basic Economic Principle
Baker, along with many other physicians who write on this topic, pays little notice to a widespread foundational premise of modern economics, the subjectivity of value [6, 7]. It is widely understood by the economics profession that different people value the same commodity or service differently, which explains why one person is willing to pay more than another for the same good. What follows from this economic truism? Baker attempts to show that the chance of finding life-threatening disease on screening CT of the abdomen in sufficient time that the natural course of the disease can be modified is remote. Even if that is true, the subjectivity of economic value implies that nothing certain can be deduced from that medical fact in determining the willingness of people (who fully understand Dr. Baker's point) to pay for screening CT.
Some examples of the subjective nature of economic value will clarify my point:
One: Antibiotics are life saving, but their value is practically nil to Christian Scientists. Explaining the objective likelihood of penicillin curing his pneumonia will not induce the Christian Scientist to value it more highly.
Two: Chemically, Kosher wine is identical to non-Kosher wine. Pointing out this objective, scientific fact to a devout Jew will not cause him to value non-Kosher wine the same as Kosher wine.
Three: Imagine two men with identical headache symptoms. One is an air traffic controller and the other, a professional boxer. We would all understand if the air traffic controller is more concerned about his headache and therefore places more value on a head CT, than the professional boxer, despite the fact the professional boxer has a greater pretest probability of abnormality [8].
Four: Cochlear implant surgery is now available, allowing some deaf people to hear. Most deaf people find this to be a godsend, but some object, arguing that deafness is not a disability but a culture and that such surgery amounts to attempted cultural genocide. The objective facts of this surgery are the same for two deaf people, who may nonetheless value it differently because only one of them wishes to correct his deafness.
I trust these examples make clear that economists, discussing the subjectivity of value, do not use the term "subjective" as synonymous with "arbitrary" or "capricious." "Subjective" refers instead to the economic actor, the individual subject from whom value originates. This insight—independently developed in the l870s by Karl Menger, Leon Walras, and Stanley Jevons— initiated the paradigm of marginal utility into economics and sounded the intellectual death knell for objective theories of value, including the most famous, Marx's labor theory. Subjective data, representing personal and ever-changing preferences of millions of people in the marketplace, are hidden to physicians but cannot be ignored.
So just as the decision to get a leather interior in your new car hinges in part on objectively measurable characteristics of leather but in the final analysis is a function of how valuable you find the look, feel, and smell of leather, so too the decision as to whether to get a screening CT hinges in part on the statistical likelihood of finding disease early but in the final analysis is a function of how valuable you find the minimization of risk (or the substitution of one type of risk for another) offered by screening. As such, the decision is more an economic one than a medical one, and therefore specialists in medicine have no unique insight into the question or its answer.
Further Economic Interlude
But one aspect of the economics of screening CT is wholly unlike the economics of standard medicine and deserves focused comment. In standard medicine as practiced in the United States today for most patients, third-party payments divorce those who receive the benefit from those who pay for the benefit, creating easily anticipated economic distortions. Those paying for but not receiving benefits wish to limit costs; those receiving but not paying for the benefits wish to maximize service. With screening CT, there is no disconnect between the recipient of the service and the purchaser of the service; thus, only those who believe the cost is worth the benefit will buy the service. There have been no studies indicating any significant fraction of these willing customers believe, ex post, that their ex ante decision to purchase was unwise. This is simply another way of saying that screening CT is providing value to those who voluntarily choose to undergo and pay for the procedure.
The same, unfortunately, cannot be said for routine insurance-covered procedures we perform daily, for which many patients— should they have to pay for the examinations themselves—would not find the cost worth the benefit. Some may think, here, that the patient, in his role as employee, really does pay the full cost, because the insurance benefit is obtained at the expense of forgoing a higher salary. But this assumption does not take into account the perverse and economically distortive consequence of paying for health care as an employee benefit rather than purchasing health insurance on one's own.
It is not the power of group purchasing that pushes people to seek health insurance as an employee benefit; one could form any number of voluntary associations to achieve that purpose, buying health insurance through one's church, synagogue, PTA, professional organization, and so forth. In fact, a century ago it was commonplace to purchase group insurance through such voluntary organizations [9]. The reason health care is preferred as an employee benefit is that employer-provided health insurance is purchased with pretax dollars (the employer deducts the cost as a business expense), but health insurance purchased by the employee is paid with posttax dollars, using income remaining after taxes are deducted from the salary. This tax policy results in a natural desire to obtain much more in the way of benefits (soon viewed as entitlements) than the very same consumer would desire if facing a level playing field from a tax perspective. In short, people "purchase" more health care than they would otherwise desire to avoid taxes.
This differential taxation policy explains why people want their health insurance paid for by their employer, unlike car insurance or home insurance. And it explains why people are quite willing to purchase (for themselves) car insurance with high deductibles and pay for car maintenance on their own rather than insist their insurance cover it, but want employer-paid low-deductible health insurance that covers everything, including routine annual physicals.
Some might wonder how this distortive tax system for health care developed historically [10]. The answer dates back to wage and price controls put into effect during World War II. Given the increased domestic demand in certain industries fueled by the war effort, strategic shortages developed because employers in those industries were prohibited from offering higher salaries to attract needed employees. The employers' solution: offer more benefits. Employer-paid health insurance was the main benefit so offered. At the time, the federal government did not think to tax this benefit. When it attempted to do so a decade later, strong business lobbying defeated the effort. Thus, the economic distortions that riddle our health care system date back to wage and price control decisions of a far-off war.
Back to Screening CT
Baker argues that interpreting radiologists must appreciate the issue of prevalence of disease and the implications for false-positive interpretations. Of course. He implies that such information should be conveyed, before the purchase, to potential customers. I agree: Ethical business conduct (to say nothing of the legal doctrine of informed consent) demands this. Should IV contrast material be offered as an option, explaining the counterbalancing issues of increased cost and risk [11] versus increased accuracy? No doubt. But none of this argues against screening CT per se, only against certain methods of offering screening CT.
To make an analogy: There is no question that, 15 years ago, many radiologists interpreted mammography suboptimally and analyzed mammographic images of suboptimal quality. The American College of Radiology did not use that as justification for opposing screening mammography. Instead, the organization used the potential value of screening mammography as justification for increased training efforts and the establishment of standards in obtaining and interpreting mammograms.
Unfortunately, this effort in mammography was done primarily through political rather than market mechanisms. The end result is that mammographers are now drowning in a flood of Mammography Quality Standards Act–mandated paperwork while demand overwhelms supply because of political pressures to limit charges, which creates the economic definition of a shortage (price artificially held below market demand). Radiologists must learn from this debacle. On the one hand, it would be foolish to push for universal insurance coverage for screening CT—a political mandate that would only increase insurance costs for everyone, even those who are not candidates for or desirous of screening CT. And on the other hand, it would be foolish to oppose screening CT for those who are fully informed and willing to pay the market rate. Physicians who see their efforts in an educational rather than paternal light should be gladdened by the number of people willing to take responsibility—financial and otherwise—for their own health.
Advertising and Entrepreneurial Radiology
Although Baker is correct that there are practitioners in the market who advertise screening CT and provide a product of limited quality, I am sure he would also agree that the problem of poor quality is not limited to entrepreneurial radiology. Baker has no doubt seen, for example, double-contrast barium enemas of shockingly poor quality, poorly interpreted. Those studies, properly ordered by referring physicians, were duly paid for by third-party payers, generating the same payments as high-quality examinations performed by expert gastrointestinal radiologists. I have never heard of an insurer refusing payment of an imaging study because the properly ordered examination was poorly performed.
In fact, one opportunity afforded by screening CT (and more generally by entrepreneurial radiology) can be stated simply: advertising directly to the public allows radiologists to educate the public both to potential health risks and to what imagers can do to help avoid them. An educated public is a more discriminating public.
I am not attempting to defend all aspects of medical advertising. That would be foolish. I am merely attempting to disabuse radiologists of the faulty assumption that advertising is necessarily inappropriate and unprofessional. Already many radiology groups have Web sites acting as de facto advertisements.
Though many radiologists believe it is unprofessional to advertise, advertising is simply informing potential patients of opportunities to improve their health, opportunities about which, were it not for the advertisement, they would be unaware. I'm told interventional radiologists get a surge of spontaneous requests from the public for uterine artery embolizations every time the subject is discussed on The Oprah Winfrey Show. Why is it appropriate to accept patients if they hear about your services from Oprah Winfrey, but not if they hear about you directly, through your own efforts?
Advertising is often frowned on because we have all seen screening CT ads that are misleading, if not downright mendacious. Yet none of us who believe that obstetricians perform less skilled fetal sonography than radiologists believe the presence of obstetricians in that market means we should not enter it at all. Similarly, that companies offering poor-quality screening CT engage in advertising should not dissuade those offering a better product from advertising that very fact. Over time, those providing the better product (as judged by the final consumer, evaluating all aspects of the service) can be expected to win. Health care consumers, after all, are not stupid. If they seem that way to some radiologists, it may be because, until recently, consumers had little say in how their health care would be provided and thus pursued a policy economists call "rational ignorance," another perverse consequence of third-party payments. When prospective patients are given a benefits package that involves limited choice of service providers, it is rational to remain ignorant of what else is available in the market. When consumers pay with their own money, they become appropriately cautious and compare both price and quality, as they do in every other competitive market.
Specific Concerns
My comments so far have been general. But Dr. Baker's specific arguments must also be carefully considered. For example, Baker hypothesizes, as an argument against screening for aortoiliac atherosclerosis, that patients at risk who find they do not have aortic or iliac artery calcification may then choose to ignore cholesterol-lowering recommendations given to them by their physicians. With an infinitude of personal preferences, this may, of course, occur. However, the only relevant report of which I am aware relates to a preliminary finding in the Early Lung Cancer Action Project trials, where Henschke noted that 23% of 134 active smokers retrospectively self-reported they had quit smoking after a screening examination with normal results [12].
Elsewhere, Baker notes that renal cell carcinoma is relatively rare, occurring in approximately one in 650 people who are 40–60 years old. The implication is this incidence is so low as to make screening a bad deal. Yet Kopans [13] indicates that the incidence of breast cancer in women in the same age range is about one in 510. Mammographic screening is performed annually. No one suggests abdominal CT screening need be done that often. Mammographic screening is associated with a relatively high false-positive rate; no one suggests that four of five solid renal masses found on screening abdominal CT will be benign disease. Baker suggests upward of one half of renal cancers will be missed on CT performed without contrast material. Yet everyone is concerned about the percentage of missed breast cancers seen in retrospect; this is a major impetus for double interpretation and the development of computer-aided diagnosis. If Baker favors screening mammography for breast cancer, why does he think he has made an argument against screening CT for renal cell carcinoma?
In addition, as Baker moves from one type of visceral abdominal cancer to another, showing each to have a relatively low chance of early detection, he neglects to mention that the probabilities are additive [14]. That is, the risk of developing renal cancer is low; the risk of developing pancreatic cancer is low; the risk of developing lymphoma is low; and the risk of developing liver cancer is low. But the risk of developing renal cancer or pancreatic cancer or liver cancer or lymphoma is not quite so low. Thus, Baker's "divide and conquer" analysis is noncontextual.
Summary
In summary, Baker's observation of poor quality in screening CT is an argument for offering better quality, not an argument for discarding screening CT altogether. His statistical observations are relevant to the interpretation of findings on screening CT but are less relevant if viewed as objective evidence that no one should undergo screening CT. Some of Baker's analysis is methodologically uncharitable in that he makes arguments that, if applied consistently, would oppose screening mammography. Finally, Baker's concern that poor-quality screening CT will make the population hostile to radiologists in general should be viewed as an appeal for more ethical and qualified physicians to offer screening CT, not as a call for politicians to ban this kind of imaging.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Levatter, D. J. Brenner, and C. D. Elliston Radiation Risk, Screening, and Standard CT: Further Reply to Brenner and Ellison * Dr Brenner and Mr Elliston respond: Radiology, October 1, 2005; 237(1): 376 - 377. [Full Text] [PDF]
|
|
|
|
 |
|
 S. R. Baker Screening CT and Modern Economics Am. J. Roentgenol., November 1, 2003; 181(5): 1195 - 1196. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
