An anonymous reader and Ernie G attacked me yesterday. I suspect that neither has read this blog for a long enough time to understand my well stated position on performance measures. They also question the “anger” of my posts. The title of this blog is Medical Rants. I am the author and entitled to hyperbole. I hope to make readers think and respond. I guess I am succeeding.
Performance measures can be useful if used to improve care. I do not mind some feedback on some measures. However, too often performance measures are renamed quality measures and used to establish rankings and incentives.
The problems with performance measures are numerous. First, many of the performance measures are constructed in an arbitrary and debatable fashion. Some performance measures have caused harm – the 4 hour rule is the best example. The sore throat example was cited at a talk I gave. These family physicians were being told that they should test all children with pharyngitis. I know the guidelines well, and this performance measure is not consistent with the guidelines. As I am writing this rant, I have just reread the IDSA guideline, which makes explicit that we should reserve testing to patients who have signs and symptoms suggestive of streptococcal pharyngitis. Thus, this performance measure will encourage over testing.
Look at several other measures. The HgbA1c measure is clearly too aggressive. That measure indirectly encourages physicians to add a 3rd medication and likely tip the harm benefit ratio over to the harm side.
But even if the measures were perfect, we have another problem. The measures only evaluate a subset of patient care. We have measures for some common problems, but we care for more than common problems. We do more than manage disease. We make diagnoses (or miss them); we counsel patients; we comfort patients. We order inexpensive tests and medications; we order expensive tests and medications.
If used properly performance measures might help us refine part of our care. However, too often performance measures take too much importance and influence our priorities.
I believe it my opportunity to shout about this issue. This is my blog; I pay for it; and I will display my anger when I want to display my anger.
To say it simply – performance measures do not define quality because quality is multidimensional. To suggest that performance measures are quality measures is, in my opinion, disingenuous and potentially dangerous. They encourage physicians to focus on a subset of measures – taking time away from other concerns. They encourage physicians to “cherry pick” patients who are more likely to meet performance measures. I worry greatly about the unintended consequences of the performance measurement movement.
To Ernie G – you misread my point about clinical care. I was talking about academic medicine. When I write about clinical conditions, I am seeing patients. While I do not see as many patients as private physicians, I do see more than most researchers. Your attack on academic clinician educators is unseemly. I really do go to the bedside and make diagnoses, consider therapy and teach. I believe you underestimate the clinical acumen of academic clinician educators.
The point that I was making yesterday is that making decisions about performance measures and research when totally divorced from clinical practice is fraught with hazard. I wonder why Ernie G and embarrassed are so mad at me.
{ 3 comments… read them below or add one }
The current iteration of “performance measures” has more to do with the Maoist “re-education” of physicians that any real interest in quality per se.
I’m a nurse I think performance measures are similar to protocols. Overreliance on either or as a substitution for critical thinking is just plain “dumb”.
Measuring medical performance is a joke. The government and insurance companies have no reliable method to measure medical quality, so they rely instead on easily measurable events, that have little meaning. These are poor surrogates. Just because medical quality can’t be easily assesed doesn’t mean we should use check-lists and other measurable quantifiable events. How do we know if a sculpture is work of art? If we don’t have a good method to do so, would we weigh the sculpture just because it is easy to do so?
{ 1 trackback }