Jousting Between the New York Times and the Dartmouth Atlas Over Interpretations of Cost & Quality Variation.
Posted Jun 24 2010 5:25pm
The New York Times - Dartmouth Atlas dueling reminds the Disease Management Care Blog of academic-medical swordplay. Whether it was trapping your peers on hospital rounds with obscure eponyms (“Have you considered Pott’s Disease ?!”), battering the Morbidity and Mortality Conference speaker with a displays of expertise ( Pseudohypoparathyroidism ?! How about your pseudoknowledge?) or quoting contrary statistics at medical meetings (“Are you as confident of your findings knowing the t-test was only one sided?”), the DMCB has seen scholarly blood spill faster than medical school faculty trying to get out of taking emergency room call.
Which is why it’s not alarmed at the tenor of the discourse between the Times’ reporters Reed Abelson and Gardiner Harris and D-Atlasians Drs. Skinner and Fisher. Academicians keep things in perspective. After sharing stale coffee and fat-laden donuts, we generally get back taking care of the patients and publishing good research.
Unfortunately, this kind of jousting can be misinterpreted and used to fuel even partisan rancor . While such pugilism has its place, the DMCB thinks its ultimately OK to have disagreements over interpretations of data. Somewhere between the Times and the Dartmouth Atlas is where the truth lies. We only need to remind ourselves that seeking truth, not political advantage, is why the latest rejoinder to The Grey Lady is worth a close read.
After the DMCB's review, it thinks Drs. Skinner and Fisher are
1) distancing themselves from the intellectual drift a) from there is no apparent reason for the variation in U.S. health care costs, b) to the assumption that lower costs cause better quality. While health care costs are lower and quality is higher in sections of Minnesota and Colorado, we simply don't know if dropping a Mayo in Dade County or an Intermountain in McAllen Texas will accomplish either. Good for them.
2) recognizing that measures of health care quality used to statistically control for the assessment of observed variations are not only imperfect, but may have little correlation with patient outcomes - at least the ones that are important to patients themselves. More work is necessary and who knows, we may find that in some instances for some hospitals, more money is necessary to overcome the impact of factors like local poverty . Good point.
3) protesting too much about the fueling of their rock star status by potential misinterpretations of their data and how it could be leading to flawed public policy. It's not their fault, they say, if politicians are misusing their data. They're innocent bystanders. Really.
The DMCB will continue to ponder this in future posts and invites reader input. Hopefully as Medicare payment policy continues to evolve, the Feds will consider all the angles and exercise the kind of caution warranted when rushed policymaking and imperfect science intersect.