Cataract Surgery Quality, Streetlamps, and Car Keys

Abstract

a i q d a m t a t i o g s o fl o t Measuring—and ultimately paying for—quality medical outcomes is a fundamental component of many efforts to reign in the ongoing escalation of costs of medical care. Currently, physicians and health care systems are paid for providing quantity of services without correlation to the quality of the services. Certainly there are secondary quality systems in place, but typically these systems rely on a set of standards that define minimal required systematic behaviors rather than outcomes. In contrast, for many endeavors in a capitalist economy, outcomes are inherent in the compensation, whether you are a professional athlete, financial analyst, business owner, or manufacturer. Quality outcome metrics in professions are much harder to craft, however, than baseball batting averages, automobile safety rankings, or gasoline mileage. Controversies rage over teacher evaluation and compensation based on standardized testing or other student performance measures, for example. In the case of physicians, the many variables in patient care are impossible to take fully into account, which has lead to the culturally accepted concept of the “art” of medicine in addition to the science of medicine. Critics of efforts to tie medical reimbursement to outcomes point out that physicians will rapidly figure out the types of diseases that have poorer average outcomes (in cataract surgery, think of denser nuclei, pseudoexfoliation, corneal guttae, etc.) and refer those patients elsewhere, presumably to hapless consultants who don’t know how to dodge these reimbursement bullets. In the absence of an unimaginably complex system of tracking comorbidities, it strains credulity to deny the strong potential for disruptive distortions in patient care that actually increase total cost and worsen the patient’s medical care experience. Nevertheless, current U.S. law includes the progressive introduction of the first step of tying reimbursement to reporting of certain metrics, known as the Physician Quality Reimbursement Initiative (PQRI). This program is often referred to as “pay for performance” although it is more accurate to call it “pay for reporting performance.” Performance is a process metric, one that hopefully correlates with outcome. But despite inserting “quality” into the acronym, “pay for performance” is not the same as “pay for outcome,” because performing a predetermined preferred set of behaviors does not necessarily result in better quality outcome, given the numerous other factors that determine final outcome. Some health policy experts believe that the step from performance to quality outcome is relatively small. Is it? Despite the intense pressure for implementation of PQRI, studies showing improved quality of care are sparse. Quality of care may even decline when it is not tied to a specific incentive. In this issue, Hahn et al report a large and carefully structured multicenter study of 7 German cataract surgery centers, where all surgeons included were highly experih