Umm…of these three possibilities, which is best? Which is worst?
1. Guidelines that are as perfect as possible: free from conflicts of interest in their conceptions, and taking into account all necessary clinical nuances (and by extension, having clear limits on their application to patient groups for which data are limited)
2. Guidelines developed under cost constraints: these will be imperfect, of course.
3. No guidelines. Wild West! (and we’ll never know just how wild it is, because we’ll have nothing to measure).
I love straw man arguments and hate multiple choice tests. Curious apparently is missing my point, so I must not be clear.
I favor limited guidelines. These guidelines can help physicians practice. Examples include:
- Patients with decreased systolic function should take either an ACE inhibitor or ARB unless they have adverse affects
- Patients with known CAD or equivalents should take a statin
- Patients with oxygen saturation less than 88% (secondary to their pulmonary disease) should be offered home oxygen
- Physicians should strive to help diabetic patients lower their HgbA1c, with a goal of at least < 8, depending on the number of drugs needed and the patient’s other co-morbidities.
I favor limited guidelines, but not measurement. Measurement has too many unintended consequences.
Why should we only focus on conditions with guidelines? Why should we ignore diagnostic errors? Why should we ignore bedside manner?
Curious, like many in our society, wants to measure something. He/she sets up a straw man that not measuring creates the Wild, Wild West. We should not do something for the sake of doing something.
So I refuse to answer the question directly. We need well proven guidelines, but we do not need performance measures.
{ 3 comments… read them below or add one }
At the Paris peace conference in 1919, David Lloyd George made an observation about the idealistic Woodrow Wilson. The exact words escape me, but it was very close to this: “He found that his rigid yardstick could not take an accurate measure of timber that had been gnarled and twisted by the storms of centuries.”
The context is different, but perhaps the metaphor applies. Measuring sticks are rarely flexible enough to measure accurately the contours of a multidimensional surface whose boundaries may change from one day to another.
Does that sound apt?
“We should not do something for the sake of doing something.”
But what about not letting the perfect be the enemy of the good?
But what about moving forward because the status quo is untenable?
oops, I forgot about the R. W. Emerson quote over to the right. something about foolish consistency?
(sorry, DB. couldn’t resist.)
If you cannot measure a thing (even in theory), then the thing in question lacks practical meaning. The definition of any concept implies its measures.
The problem is never measurement. It is only the use of measurement that can be problematic…especially in unintelligent, conflicted, or malicious hands. Or the use of measurement that does not acknowledge the limitations of existing measures.
There’s nothing wrong with measuring the average HbA1c in any given population of diabetics. What’s wrong is attaching powerful incentives to an A1c threshold that is so low that patients are, on average, harmed.
And who says we can’t measure bedside manner? There is a growing number of patient experience survey instruments designed to do just that. I’m sure you already know that multiple teams are currently working on measures of diagnostic error.
{ 1 trackback }