Aligning Measures to Improve Quality

Feb 11, 2014, 4:51 PM, Posted by

Doctors go over a patient's charts in the emergency room.

The quest over the last decade and a half to define and quantify “quality” in health care in the United States has resulted in widespread use of quality measures. Unfortunately, the alignment of these measures among entities in both the private and public sectors has been secondary to the efforts to identify and use good measures. This failure has resulted in a tremendous lack of comparability between quality improvement efforts.

While not surprising, the near total lack of alignment has become a major obstacle in the effort to improve care for patients. It leads to significant burdens for those looking to improve, wastes valuable (and finite) resources and is a drag on overall quality improvement efforts. Additionally, it creates a considerable barrier to efforts encouraging value-based decision making by consumers and others.

Since 2011, the Buying Value initiative has worked to advance value-based purchasing in health care, working with federal and state agencies, quality collaboratives, providers, health plans, purchasers and consumer organizations. Our 2014 work will be focused exclusively on improving the alignment of measures while also encouraging innovation and respecting the priorities of different stakeholders.

Last fall, Buying Value thoroughly analyzed state and regional quality measures, shining a light on the enormous variation amongst the measures in use nationally. The study, which was conducted by Bailit Health Purchasing of Massachusetts, analyzed close to 1,400 measures used in a range of health programs sponsored by state agencies and regional health improvement collaboratives, such as medical homes, Medicaid, ACOs, value-based purchasing and system reform.

It found that, of the 509 unique or distinct measures in the universe, only 20 percent appeared in more than one of the sets, and only 4 percent appeared in a third of the sets. Compounding the problem, 23 percent of the standard measures used were modified in one or more ways, making alignment difficult (if not impossible).

Incredibly, of the national standard measures that were used, NOT ONE appeared in every set. Breast Cancer Screening was the most commonly used standard measure, but only 63 percent of the sets (30) used it.

Further aggravating the problem, states/regions frequently used non-standardized, “homegrown” measures, which made up 36 percent of the measures (198). While local innovation in measures ought to be encouraged in order to better fit the needs of the subpopulations for which they are chosen and to accommodate any local limitations on data collection and use, the study found that only 27 percent of the homegrown measures (53) could be classified as truly innovative.

That said, after nearly 20 years on the frontlines of quality improvement, I have hope that change is on the horizon. Following the study, Buying Value has launched an effort to increase alignment by providing technical assistance to states/regions on the selection of best measures of quality while emphasizing comparability. We are also developing and disseminating a consensus strategy to improve alignment while respecting the different needs of various parties and supporting innovation. This work is generously being supported by the Robert Wood Johnson Foundation with programmatic and administrative support from the National Quality Forum.

Gerry Shea directs The Buying Value Project, a private-purchaser-led effort started in 1999 to promote improved healthcare quality and lower costs through value purchasing. Learn more and follow the project at