It’s complicated: New payment models require clear goals, expectations

Payment reform does not have to hurt, and might actually help the bottom line. Alternative payment systems that tie compensation to quality measures should have clear, measurable goals, and it is crucial that the behaviors to be changed, and their connection to patient care, be clearly communicated. These findings were the results of a study by the American Medical Association and the Rand Corporation, released on March 19, 2015.

Focus and goals

The researchers examined the effects on physician practices of participation in alternative payment models such as pay-for-performance, medical homes, and bundled payments for episodes of care. The study focused on 34 physician practices in six markets. Many of the physician groups were participating in more than one payment reform project instituted by public and/or private payers and, at the same time, were implementing or upgrading their electronic health record (EHR) systems to do so.

Financial incentives, measuring productivity

One common feature of all new payment models is that they shift from fee-for-service (FFS) to another measure of productivity or value. The tools used to measure productivity have not yet caught up with these changes, however, because resource value units (RVUs) are based on the FFS system. Practice groups working with multiple payers may work simultaneously under FFS and alternative payment models. As they try to meet their goals, there may be pressure to raise the RVUs billed to the FFS payers while lowering costs under the alternative model.

Simultaneous changes, conflicting incentives

The transition to payment models and the adoption of EHRs to record and track quality measures both require intensive investment of capital. In addition to government incentive programs, physician practices are merging with other practices or aligning themselves with, or even being acquired by, hospitals or health systems. These changes give rise to an interdependence that may create conflicting incentives. For example, a physician practice that reduces hospitalizations may not be making the expected contribution to the financial health of the hospital system that owns it.

Physicians who are expected to reduce costs do not always have the information they need to do so. Specifically, they may not know the cost of the drugs they prescribe or the tests they order. The introduction of an expensive new drug may be out of their control and adversely affect their cost measures.

Effects on incomes

The practices shielded individual physicians, at least in part, from the effects of financial incentives to cut costs. The study found that the effect of participation on incomes was either neutral or positive.

Professional satisfaction

Doctors were willing to make changes that they perceived as related to patient care. For example, they willingly began documentation to highlight patients with diabetes. When they did not see a connection between a new documentation requirement and benefit to patients, however, they became dissatisfied. When the behaviors to be adopted and the reasons for doing so were made clear, doctors were willing to put in the extra work. When the leadership could not articulate clearly what doctors needed to do and why, the physicians were dissatisfied. Those who did not play leadership roles were less likely to find satisfaction in making the transition from FFS to payments with a value- or quality-based component.

Conflicts between the requirements of multiple incentive programs sometimes made it difficult for physicians to concentrate on the behavior they were to improve.

It’s all about the data

Implementation of quality measurement components requires both a lot of data and the ability to analyze it. Some physician practices needed to hire data experts in order to make use of the information they gathered. Data that was not current or complete could hamper efforts to introduce new behaviors or practices. Flaws in the integrity of the data and attribution errors also presented obstacles. The authors urged that any new data system be tested thoroughly and undergo a “dress rehearsal” or dry run before doctors are required to use it.