Friday, September 12, 2014

Of Risk Stratification, Health System Variation and "Stupid" Decision-Making

A fly in the ointment
Years ago, a middle-aged Population Health Blog patient came in for a routine follow-up appointment.  Since his last visit, he had developed iron deficiency anemia. Since slow blood loss can be a sign of an early and curable cancer in the gastrointestinal tract, the PHB recommended a series of unpleasant tests. After a rather routine explanation of the time, expense and inconvenience of those tests, the patient surprised the PHB with a one-word answer: "No."

He went on to live for decades.

Which brings the PHB to this JAMA article on individually-tailored screening for another type of cancer. While even screening for prostate cancer is controversial, it's possible to stratify a man's risk of the condition with some questions, examination data and test results.  That risk can be portrayed in lay terms (there is a "42-in-100 chance" that cancer is present, but doing a biopsy has a "4-in-100 chance of causing an infection..."). 

The points of the well-written article is that 1) risk-stratification can be used to identify persons at high vs. low risk, 2) the decisions to screen, perform additional testing and embark on treatment can be, based on that risk, "tailored" to maximize a good outcome and 3) patients can use their level of risk to ultimately decide how they want testing and treatment to achieve the outcome they want.

Bravo, says the PHB.  While we're on the cusp of understanding whether a more sophisticated approach to screening ultimately leads to better outcomes than the standard all-or-none guideline (USPSTF "recommends against prostate-specific antigen (PSA)-based screening for prostate cancer"), there is enough face-validity to believe that patients will ultimately benefit.

But there is a fly in the ointment and a monkey in this wrench.

The fly? Variation will not go away. While health system bureaucrats everywhere would prefer that 0% of men undergo prostate screening, that 100% women over 50 get mammograms, and that 0% of us have a body mass index in excess of 25, individuals - after looking as the risk-benefit here, here and here, may choose otherwise.  We don't know what the "right" screening rates are.  In fact, we may not be asking the right questions.

The monkey?  Some "bad" decisions will occur. Once persons truly understand the benefits, risks and alternatives (including not dying prematurely of a preventable illness and side-effect risks that are less than driving in a car), they are allowed to make "stupid" decisions.  Physicians and bureaucrats may not like it when anemic patients, like the one described above, refuse no-brainer recommendations, but in a free country that's the price we pay. Our challenge is to make sure that our patients have all the information they need (which is apparently not the case here) to make a truly informed decision.

Image from Wikipedia

No comments: