In Reply: We agree with Dr Finestone that the
performance by physicians in our study was far from ideal. We do not, however,
agree with his implications. He argues that improvements in physician performance
would best be accomplished by tying reimbursement directly to performance.
While we agree that this approach might work for some measures, we object
to it for theoretical as well as practical reasons.
We believe that physicians should strive to perform optimally, regardless
of monetary rewards or punishments. Furthermore, we believe that most physicians
do aim to deliver the best care they can. Medical practice is complex, involves
systems and not just individuals, and the measurement of quality is a scientific
field still in its infancy. Current quality indicators suffer from the "searching
for the keys under the lamp-post" phenomenon, in that they represent what
we know how to measure, not necessarily what would be most important to change.
We hope that our method creates an environment for improvement that has positive
effects on performance beyond those indicators we measured. We are testing
this effect in ongoing studies.