Current measurement of diabetes care quality is complex and cumbersome, says NCDP

The National Changing Diabetes® Program (NCDP) announced today results of a study that found extraordinary variety in the measurement of clinical processes critical to optimal diabetes management such as blood pressure and glucose levels.

In the study, Quality Measurement in Diabetes Care, published today in Population Health Management, researchers at Thomas Jefferson University's Jefferson School of Population Health, Philadelphia, reported 146 distinct measures for 31 medical outcomes - a quilt of measurements that healthcare providers say is complex, disjointed and cumbersome. In the study, the authors note that "although the broad array of existing measures creates valuable opportunities to quantify, benchmark, and improve a wide range of clinical processes and outcomes, providers and other stakeholders report that the broad scope of activities lacks clarity."

An increasingly common strategy for improving care and adherence to treatment guidelines has been periodic performance assessments of doctors, health systems, and health plans. Several organizations provide such assessments, and the Centers for Medicare and Medicaid Services includes diabetes measures in its doctor-incentive initiatives to encourage improved medical care and promote public reporting of performance data. However, according to the study authors, "currently, the environment is characterized by hundreds of measures that frequently do not conform completely with key facets of measure selection, data sources, and standards for defining high-quality care. Clinicians report dissatisfaction when they encounter mixed messages regarding testing and screening schedules or target goals for key outcome indicators."

For the study, the Jefferson research team searched the National Quality Measures Clearinghouse, a comprehensive database maintained by the Agency for Healthcare Research and Quality, and reviewed diabetes measures created by national organizations and institutions. The researchers categorized measures by medical procedure, such as managing medication and performing foot exams; health status indicators such as cholesterol and blood pressure levels; and other types of assessed care, such as patient self-management.

The team also conducted interviews between July and October, 2008, with leaders from the organizations who use and develop quality measures, along with other stakeholders in the quality measurement community.

"We found a measurement system that is both redundant and inconsistent, with many different measures assessing the same clinical indicators,'' says Dr. Nash, one of the study's authors and Dean of the Jefferson School of Population Health. "Methods and assessment goals vary among different organizations, as do, in some cases the sources of data (for example, patient or health plan level data) making uniform standards more difficult to achieve."

Doctors aren't alone in grappling with these challenges, researchers say. According to the study, large employers and policymakers are also struggling to understand which measures are most appropriate for their quality initiatives. Ideas proposed by the authors for reconciling different quality measures include: development of a "comprehensive'' measure set, which would focus on measuring indicators that are most significant for diabetes care, such as glucose, blood lipids, and blood pressure; and use of a "composite score,'' that would combine a range of indicators and generate a single score.

"This study demonstrates that current measurement of diabetes care quality is far too complex and disjointed, and at the same time lacking in a number of key areas, particularly at the population level," says Dana Haza, senior director of NDCP.

The study also found significant gaps in quality measures, including a lack of population-based or epidemiologic measures. In addition, much of the current focus is on patients who are employed and insured, which ignores "millions of Americans and may not accurately reflect the overall state of diabetes care,'' the researchers write. Patient perspectives, access to care, or efforts to identify and advise those with pre-diabetes, those at high risk of developing the disease, are also not commonly measured.

The research was sponsored by the National Changing Diabetes® Program, a diabetes leadership initiative established by Novo Nordisk to drive health systems change at the national and local level. About 24 million Americans have diabetes and health experts say that number could double in the next 20 years. With direct and indirect costs associated with diabetes and pre-diabetes estimated at $218 billion in the U.S. in 2007, much effort has gone into developing practice guidelines that improve care and patient outcomes.

"Streamlining and harmonizing existing measures is an important challenge that must be addressed to encourage wider use of performance assessment,'' the researchers conclude. Haza adds that a clearer understanding of the state of diabetes quality measurement is "an important step toward improving their usefulness, and addressing gaps in quality assessment.''