Health Affairs: NCQA's medical home assessment tool fails to deliver

The National Committee for Quality Assurance’s (NCQA) tool for assessing and accrediting patient-centered medical homes has been endorsed by the federal government, but research published in the February issue of Health Affairs suggested that there was no significant association between a healthcare organization’s ability to achieve a passing score on the assessment tool and the quality of care it delivers.

To investigate whether the Washington, D.C.-based advocacy organization’s tool truly measured performance, researchers from the David Geffen School of Medicine at the University of California, Los Angeles used a paper version of the 2008 patient-centered medical home tool to assess coordinated care for diabetes patients at 30 community health centers providing comprehensive primary care services and employing at least one physician.

The tool grades 30 features of healthcare organizations across nine domains according to a scale that awards several levels of recognition ranging from “not recognized” to levels one, two and three. Based on those grades, healthcare organizations are assigned a score of up to 100 and any organization receiving a score of more than 25 qualifies as a patient-centered medical home if it applies for accreditation through the NCQA.

Researchers examined the ability of organizations that qualified as medical homes to care for diabetes patients because they believed diabetes acts as an apt representation of chronic diseases requiring an array of services. To assess organizations’ processes of care, researchers looked at whether they conducted hemoglobin A1c, low-density lipoprotein cholesterol, blood pressure, urine protein and dilated eye examination tests for diabetes patients in the past year. To assess outcomes at organizations, researchers assessed intermediate outcomes of diabetes care, or test results for risk factors that precede a clinical outcome, such as blood pressure as a predecessor to heart attack. Researchers examined data on 1,455 total patients.

Each of the 30 community health centers assessed passed the NCQA assessment. The mean score that the organizations received was 67 and scores ranged from 33 to 90. “However, our analysis found no significant relationship between performance on the NCQA medical home tool and the quality of diabetes care, measured by either process or outcome,” according to the report’s lead author, Robin M.A. Clarke, MD, from UCLA's department of internal medicine, and her colleagues.

Researchers suggested that an updated, 2011 version of the tool may provide more accurate assessments, the tool focuses too heavily on technology and that tool may not be appropriately designed to assess the unique community health center model, which integrates nonmedical services into healthcare settings. The researchers also suggested that the Center for Medicare & Medicaid Innovation begin considering the tool’s effectiveness by asking itself two questions.

“First, does the NCQA tool accurately measure the core medical home domains (for example, care management or patient self-management) as delivered by community health centers?” Clarke et al wrote. “This question could be answered by sampling a number of the project’s participants and validating the NCQA results for these domains against the findings from a more thorough on-site investigation.”

“Second, does improving diabetes chronic care and providing a medical home for low-income patients require services beyond the traditional clinical processes outlined in the NCQA medical home definition?” the authors posed. “We propose that the demonstration project evaluate how effective enabling services are at overcoming barriers to care and improving chronic disease outcomes.”