Now that we no longer have to worry about the SGR, we have a new worry. The law consolidates several measurement tools into one big tool. CMS has declared that it wants to pay for value. The law provides a blueprint for paying for value. The underlying assumption of this approach is that we can define and measure value through measuring quality components.
Can we define health care value through measurement? Can we balance patient-centered care with performance measurement?
Readers of this blog know that I have frequently written about the pitfalls of performance measurement and why we should not call a conglomeration of those measures quality. As I type so often, quality is a multidimensional structure that differs for each patient. Should we treat type II diabetes the same in a newly diagnosed, otherwise healthy patient as we treat a patient with 20 years of diabetes who is now 85-years-old and has major co-morbidities?
When should we give empiric antibiotics to a patient with dyspnea and cough? Should we assume community acquired pneumonia for any infiltrate, despite lack of fever or productive cough? If we were measuring the sensitivity of rapid treatment of pneumonia, then we would treat more often accepting a lower specificity.
But what if we “cover” the patient with antibiotics for possible pneumonia and the patient develops a C. diff infection? Which performance measure should we use — treating possible pneumonia early to decrease complications and shorten length of stay or avoiding C. diff diarrhea?
What do we measure in our geriatric patients with six diseases, all of which have associated performance measures? And will we have a performance measure for avoiding drug-drug interactions or avoiding drug side effects?
How will we measure bedside manner or helping the patient understand what we are doing or shared decision-making? How will we measure prioritizing the patient’s most acute problem and addressing less dangerous problems until another time?
How will we measure making the proper diagnosis? Will we receive positive points for treating the wrong diagnosis perfectly?
How will we measure working with patients to start palliative care early enough to positively impact their quality of both living and dying?
Most important, should we not hold any performance measure to the same standard that we use to test new drugs? We should require a study showing that introducing a performance measure leads to better patient outcomes. This concept sets a high bar. If we treat one disease well does the patient benefit? Does focusing on one aspect of quality decrease our focus on other issues? Do we get distracted from really listening to the patient because our EHR tells us that we must focus on four performance measures relevant to the patient’s problem list?
When in doubt we often return to our hero Sir William Osler:
The good physician treats the disease; the great physician treats the patient who has the disease.
Could our quest for value actually harm some patients? Medicine is complex. Trying to measure a multidimensional concept like quality if fraught with hazards. Osler again:
Variability is the law of life, and as no two faces are the same, so no two bodies are alike, and no two individuals react alike and behave alike under the abnormal conditions which we know as disease.
Some measures are golden, but mostly those that we have tested. We have a responsibility as a profession to challenge this concept without seeing clear evidence that patients benefit from labeling some measures as value. Value and quality are fuzzy concepts. How can one oppose using value and quality? No one opposes the concept, but we all should demand that the implementation of measures does improve patient outcomes. We should all worry.
Robert Centor is an internal medicine physician who blogs at DB’s Medical Rants.