Wednesday, June 11, 2014

OVER/EARLY-DIAGNOSIS


Currently there is an intriguing concept in medicine, especially in oncology, which suggests that physicians have a tendency to over-diagnose. The argument goes that because of the amplified diagnostic abilities we as physicians are able to make diagnosis of a disease such as cancer earlier. And that is projected to be bad thing? A correlated argument forced as causality states that over-utilization of the diagnostic capabilities leads to “harm.”  What would be considered a laudable form of care provided to the patient in terms of early diagnosis, intervention and potential cure from the malady (cancer in this case), the experts go to extreme length, to define that as harm (1).  This “over-diagnosis” in their minds is the same as early diagnosis and therefore by using their statistical correlates it is considered a travesty. Usually the authors of these articles are policy-wonks or involved in policy making. They cite psychological harm as well as physical harm behind the reasoning for their premise. The “psychological” harm is a subjective methodology steeped in questions whose answers are created to evoke emotions. The physical harm cited however is real and are the complications related to the surgical procedure, which unfortunately is the unintended consequence of any surgical intervention.

The question arises whether we as oncologists think that Early diagnosis is the same as Over diagnosis?
The consequence of such a thought process if followed through brings us to the conclusion that we should wait till potential cancer related symptoms become apparent and then intervene, even though as the saying goes, “The cat is out of the bag.” Is that ethical reasoning? What is even more puzzling is the use of lead-time-bias that is being used to further the argument that if intervention is done early and survival is shown to increase that is nothing more than the biology of a slow growing tumor and that the lead time in diagnosis gave the false representation of increased survival. This purported argument has been very strong in the field of uro-oncology especially with the use of the PSA screening methodology. The problem that arises here is that we are nowhere close to determining the aggressiveness of the cancer except by obtaining multiple biopsies and evaluating the Gleason’s score. Arriving at this juncture then, the argument suggests that “Watchful waiting’ would be a good measure rather than subjecting patients to a radical surgical prostatectomy or a brachytherapy +/- external beam radiation therapy as a curative intent. Yet recent studies have shown that upfront intervention saves lives over the “Watchful waiting!” (2)

Given the controversy that surrounds prostate cancer, is watchful waiting the right approach in oncology care for a patient with prostate cancer?

The current data-driven analytical mind-set of correlations assumed as causations is behind a lot of these illogical thought processes and is borne of the meaningful use objectives. Unfortunately some of these data analytics give enormous weight to the cost factor in their analysis and tend to forget the individual patient. Scientific literature is replete with such cost-to-care articles especially since the global economic downturn of 2000. A more reasoned approach would be to use hard empiricism borne of a well thought out hypothesis and validated through repeatability before using the power of media blitz.

The Jeremy Bentham utilitarian concept, through the eyes of the epidemiologists has created a tall mountain to climb for the individual with cancer. The defining art of oncology care remains that each patient is an individual who has specific needs and comes with his or her set of co-morbidities that confound the unified pluralistic ideology driven motive of “one size fits all.”

Kant’s “Reason” then must be applied to the medical care we deliver to each individual patient rather than the broad brush strokes of the multitudes that have been sampled into the Bell curve and bounded confines of the 95% Confidence Intervals to yield the biased p-value paradigm of the “truth.”

There is a fine line between menace and utility. Where upon we as physicians must decide how to shape the destiny of oncological care. Soon there will be algorithms that will drive medical care and through the rigors of data crunching a holographic representation of a “healthcare provider” will render the physician moot. When that day comes, let us hope not too soon, age and cost hopefully do not drive utility and need to become the technological menace for humanity.


2a.  Radical Prostatectomy or Watchful Waiting in Early Prostate Cancer — NEJM http://nej.md/1jWzJ02

No comments:

Post a Comment