Evangelists and Snails Redux: The Case of Cholesterol Screening

Controversy commonly surrounds the detection of disease. Some 20 years ago, Sackett and Holland [1] suggested that this controversy stems largely from the ideologic differences between “advocates” versus “methodologists,” or, as they later called them, “evangelists” and “snails.” Thus, evangelists hold that “the pre-existing evidence plus commonsense—in the face of the ongoing toll of disability and untimely death—demand massive screening programs for the detection of citizens with risk factors for these disorders now, even in the absence of experiments to determine whether the alteration of many risk factors will, in fact, alter risk.” Snails, in contrast, are convinced that “screening, like any other untested health maneuver, may do more harm than good and must meet scientific as well as political criteria before it is implemented.”

This issue of Annals contains a guideline from the American College of Physicians on screening for cholesterol [2] and a perspective on that guideline [3], classic snail and evangelist documents, respectively. Nearly everyone now agrees that secondary prevention through cholesterol screening works. Thus, adequate, high-quality data on the effectiveness (and cost-effectiveness) of screening can and do bring people together. The controversy over cholesterol lies in the domain of primary prevention, or true screening [1]. This is shadowy territory, where the hard data trail off into uncertainty. Here, key critical trials are absent; we don't really know how much the alteration of risk factors will, in fact, alter risk (assuming such alteration exists); and any such risk reduction will be less efficient than it would be in secondary prevention.

Decision making under uncertainty is hard on everyone. Sackett and Holland [1] suggested that under conditions of uncertainty a divide opens between evangelists and snails along the lines of two entirely different value questions. Evangelists ask “Isn't screening good?” whereas snails ask “Isn't there a scientific method for making clinical and health care decisions?” It may be more accurate, however, to view the two positions as flowing from the two opposite ends of a single moral question. Thus, evangelists ask “Isn't it wrong to withhold from patients an intervention with potential benefits and undemonstrated harms?” whereas snails ask “Isn't it wrong to impose on patients an intervention with undemonstrated benefits and potential harms?”—a form of the “glass is half full, glass is half empty” debate [4].

As Sackett and Holland [1] pointed out, however, evangelists also assume that patients' interests are best served by providing any and all services with a reasonable promise of benefit, whether patients seek out that benefit or not. Snails, in contrast, distinguish between benefit sought by patients—in which clinicians only promise to provide care as good as, or better than, that available elsewhere, with no guarantee of the patient's health outcome—and benefit implied when screeners seek out citizens and ask them to have tests and, possibly, extensive and prolonged treatment. On these grounds, snails require a greater certainty of efficacy, amounting to a kind of “truth in advertising” standard, when recommending physician-initiated rather than patient-initiated interventions, particularly when they are applied across broad populations of healthy persons.

More and better data would no doubt sharpen the focus of the cholesterol debate. Unfortunately, even definitive, quantitative data help little when a debate involves benefits that, however real, are infrequent relative to an intervention's aggregate costs and, possibly, risks, as may be true for cholesterol screening in the primary prevention mode [5]. A good many medical practices fall into this gray zone, raising the melancholy thought that controversy over such practices is inevitable.

However, there's nothing here that we can't do something about. First, we can understand these controversies for what they are: moral debates about scientific issues. Second, we can recognize that neither position is morally superior: Both evangelists and snails advocate what they see as the maximum public good. Third, we can continue to bring the issues out into the open—as through the publication of the two papers in this issue—where they can be soberly examined and better understood. And, finally, we can accept the ideas that there are no right answers here; that there are, consequently, no winners and losers; that what's required is principled rather than positional negotiation [6]; and that, ultimately, rapprochement must take place on newly defined common ground.

An enormous amount is at stake in cholesterol screening—millions of lives and billions of dollars, not to mention important matters of principle. And anyone who needs convincing that medicine, even at its scientific best, is always a social act need look no further than the current debate on cholesterol screening.

Frank Davidoff, MD.

Editor.

• Copyright ©2004 by the American College of Physicians