APeticola has an interesting post at the always-challenging Health Care Renewal, on the tendency for diagnostic “thresholds” to decline over time - that is, the fact that certain conditions will tend to be seen, and treated, in more and more patients as those conditions acquire greater salience in the medical community.

In the news recently was someone or other’s recommendation that less severe degrees of obesity than morbid obesity be also treated by gastric bypass. This should surprise no one. There are constantly “discoveries” that it is better to be more and more aggressive with blood pressure, cholesterol, blood sugar targets; etc.; and thresholds for treatment are generally lowered with each new series of recommendations.

Diagnostic and treatment thresholds ratchet ever downward. It’s definitely good business. But is it good medicine?

Good question, certainly. And Peticola does a good job laying out some of the problems that this phenomenon may generate:

As treatment thresholds move downward, medical expenses move upward. And something else occurs: as thresholds are lowered for treating various risk factors, the benefit/risk ratio changes markedly.

But I wonder if there’s a point being missed here.

As to the phenomenon itself, I suspect its appearance is partly structural and partly mindset-driven. That is to say, it’s a reasonable expectation that diagnoses will increase as a given condition becomes better known and caregivers become sensitized to look for it. This effect is often cited as an explanation for purported “explosions” of conditions that simply were not well-recognized in the past, including autism, ADHD, prostate cancer, and others. In addition, there may be the related phenomenon of the gradual reduction in diagnostic thresholds, or the willingness to treat less-severe cases, but I suspect this too is at least to some degree understandable. Truly new diagnoses, with unfamiliar therapeutic regimens, or truly new therapies in previously-known conditions, will naturally be reserved for the worst and most obvious cases, even by clinicians who are not acting consciously under an overt diagnostic protocol. As caregivers become more comfortable distinguishing the condition among other, more-familiar ones, and more comfortable managing the treatment, they will naturally begin to recognize more-marginal cases, and to offer treatment in riskier scenarios. This impacts on the overall level of care offered at any given point as caregivers climb the learning curve, but it is understandable, probably inevitable, and not necessarily evidence of bad practice.

Subtracting out these “structural” causes of the accelerating diagnosis frequency, how much remains of dubious diagnosis predicated upon clinicians’ changing their own, subjective evaluation of appropriate risk/benefit (in ways not related to the learning-curve phenomenon)? To what extent are caregivers offering more- and more-questionable treatments just because they are popular or lucrative? And is it obviously a mistake, as the “structural” phenomena above are obviously understandable? The questions raised by Peticola seem to presume that there is an objective standard for the risk/benefit ratio, and that clinicians are the guardians of patient safety in that regard. This is certainly a popular traditional view, but one that famously conflicts with a strong patient autonomy approach to decisionmaking. And from that latter perspective, there may be reasons for declining risk/benefit thresholds for popular treatments that are not only rational but beneficial, and that explain much of the phenomenon Peticola identifies.

Consider that, in addition to the decline of the caregivers’ treatment threshold, there is a patients’ threshold as well. As treatments become more commonplace, the public begins to see them as normal, expected, and routine. Significantly invasive (in the broad sense) healthcare procedures can even become a fad - as with many forms of plastic surgery, bodily enhancements (monkey glands, anyone?), and even psychotherapy and mood-altering drugs. Patients hear about these things more and more commonly, until they begin to pressure their doctor for them - and doctors often acquiesce. There is a concern that patients may not be making informed assessments of risk - especially, they may assume that anything that is common is both safe and effective beyond question (the complacency problem with anti-virals for HIV is having terrible ramifications). But even so, patient demands for treatments can at least be defended as exercises of autonomy: if (informed) patients are willing to undergo the risks of surgery for larger breasts, flatter tummies, or straighter noses, why should they not? Injecting deadly poison into facial muscles? - if that’s what they want, OK. These questions have been asked and answered ad nauseum, and we no longer recognize a controversy in such procedures. Arguably, every new medical fad is just a repetition of the undergraduate-seminar debate about cosmetic surgery: yes, there are risks; yes, patients are being frivolous; yes, it’s their body to do with as they like; and yes, in the end it’s the patients’ value rankings that determine the risk/benefit assessment, not the clinician’s. So we must subtract even further from the category of suspiciously-easy diagnoses those that are caused by patients’ high risk tolerances and low thresholds for dissatisfaction, not the caregivers’. Now how many remain?

I’m not sure it’s going to be very much - and, to the extent that there are treatments being proposed simply because caregivers fail to care very much what they do to their patients, those are indeed questionable, but on the familiar grounds of professional obligation and competence. A caregiver who fails to discriminate among treatments, or who proposes risky treatments for questionable clinical conditions, is simply not practicing effective clinical critical thinking. We don’t need to categorize it further as an example of “low threshold” or anything else. All bad clinical decisions are evidence of the same basic failures of clinical competence (if mistaken) or dedication to patient welfare (if negligent). But the mere increase in popularity of a new treatment, or even its use in increasingly less urgent and more marginal cases, is not in itself a bad thing. It may be evidence of greater professional familiarity, or of the changing of the tides of patient enthusiasms - both natural and defensible phenomena in a patient-centered healthcare system at a time of great technological advance.