Physician pay-for-performance concerns
raised in Senate hearing

During a March 26 hearing, the Senate Health and Human Services Committee listened to testimony on outcome-based reimbursement models in Texas’ Medicaid program and other health care coverage models. Overall, the testimony largely reflected the medical community’s concerns with programs such as pay for performance.

One of the physicians testifying on behalf of the medical community was John Holcomb, M.D., San Antonio pulmonologist and chair of the Texas Medical Association’s Select Committee on Medicaid.

Holcomb told the committee there is much interest in pay for performance and health plans and federal payment plans have begun implementing these programs with “almost no evidence” of their effectiveness. He voiced physicians’ concerns with the validity and reliability of measurements, costs to physicians and the sustainability of pay-for-performance programs.

Holcomb explained that two types of measurements are used to gauge quality, process measurements and outcome measurements, using the treatment of a diabetic patient as an example. “In the management of a diabetic patient, you would like to be able to tightly control the blood sugar—which is process—so that later on, 10, 15, 20 years later you have less blindness in that population, you have less loss of extremities, and you have less myocardial and stroke events,” he said. “Because the health plans can’t think in terms except in quarterly budgets, and the legislators, bless their hearts, only think in two-year cycles, it’s very hard for us to look 10 or 15 years out to get these outcomes.”

Holcomb said most health plans instituting pay-for-performance techniques are using claims systems to collect quality data and are only collecting meager process measures. For example, ordering an annual test makes a physician look like a “really good doctor” to the health plan, but in reality, the physician may not be helping the patient effectively manage his disease, he said. Outcomes measurements are much more complicated, not to mention expensive, and involve reporting what the patient actually experienced over the course of the illness.

Other hurdles that make implementing outcome-based reimbursement models in Medicaid difficult include patient compliance, an extremely geographically and ethnically diverse Medicaid population, insufficient data-collection systems, and low physician reimbursement.

On behalf of TMA, Holcomb recommended delaying statewide implementation of pay-for-performance programs in Medicaid until the committee could further investigate value, costs, impact and potential consequences. Instead, it would be beneficial to look to better investments to improve the quality of Medicaid services, he said. Some examples include looking more to medical home initiatives to improve health outcomes, reduce health care disparities and constrain health care costs; and continuing to streamline Medicaid administrative requirements in regards to paperwork, prior authorization, and billing and coding.

In his published testimony, Holcomb gave four recommendations: closely monitor and evaluate existing pay-for-performance programs using collaboration with physicians, health plans and experts to assess their effectiveness; adopt a core set of principles to guide the design, implementation and evaluation of future proposals; continue investing in physician reimbursement rates for pediatric and adult services; and invest in medical home initiatives for Medicaid patients.