Loading ...

Most of the quality measures that assess how well internists, group practices, and health plans deliver diabetes care are not scientifically sound or have not been tested in a way that supports their use, according to a new review.

The American College of Physicians (ACP) Performance Measurement Committee said it could only support five of 14 internist-related diabetes care measures currently in use. In its review in the Annals of Internal Medicine, the committee did not go so far as to say the measures do not have utility. Instead, the ACP seeks to spur developers to gather better evidence and do more testing of the measures.

“A lot of these were created out of a desire, a good desire, to measure the quality of care,” said Rebecca Andrews, MD, chair of the ACP Board of Regents and a member of the review committee. But without rigorous study or a strong evidence base, the measures are “basically very burdensome to physicians and practice and don’t really do what they’re meant to do,” Andrews told Medscape Medical News.

“Fourteen measures for one chronic condition is too much,” said Andrews, professor of medicine at the University of Connecticut School of Medicine in Farmington, Connecticut. “We should be focused on really impactful performance measures.”

The 13 reviewers rated measures on a 9-point scale and gave each an overall rating, and individual median ratings for importance, appropriate use, clinical evidence base, measure specifications, and feasibility/applicability. The measures were evaluated at the physician, group practice, and health plan levels. They were then deemed supported or not supported.

The metrics included in the review were Centers for Medicare & Medicaid Services (CMS), Merit-based Incentive Payment System (MIPS), consensus-based entity (CBE), and Healthcare Effectiveness Data and Information Set (HEDIS) measures.

The Need for Change

How to best measure quality is a decades-old debate.

Andrews said some studies have shown that measures often are not meaningful. A 2018 The New England Journal of Medicine paper by the ACP Performance Measurement Committee determined that 33 (37%) of 87 internal medicine measures were rated as valid, 30 (35%) as not valid, and 24 (28%) as of uncertain validity.

Diabetes-related quality measures have frequently missed the mark, in part because they are more focused on checking a box than improving care, said Patrick J. O’Connor, MD, MA, MPH, senior research investigator for diabetes at HealthPartners Institute, Bloomington, Minnesota. The HealthPartners Institute is the research and education arm of HealthPartners, a large, nonprofit, integrated health system.

A yearly eye exam for patients with diabetes — a measure that has existed for decades and is still in use — is a perfect example, O’Connor told Medscape Medical News.

“If you don’t want eye disease, eye complications, retinopathy in people with diabetes, the way to do that is to reduce the glucose and blood pressure, not necessarily to keep examining their eyes every year,” said O’Connor, who has published numerous studies delving into how to deliver high-quality diabetes care.

The ACP committee supported use of the diabetes eye exam measure MIPS #117 at the health plan level but not at the individual physician level because of “uncertain validity.” The measure requires documentation of an eye exam by an eye care professional, but many primary care doctors might not have access to the results for individual patients, the ACP said.

The panel backed two measures of glycemic control (CBE 0575 and MIPS Quality ID #1/ CBE 0059) for health plans but not for physicians or group practices because the first isn’t designed for them and the second hasn’t been tested at those levels.

Should More Be Done?

The committee should have gone further in some cases, O’Connor said. The eye exam measure should not be endorsed at all because it is related to detecting a complication, “rather than getting at the root causes of those complications,” he said.

O’Connor also knocked ACP for the same reasons in its endorsement of MIPS #488, which requires documentation of a kidney health exam in patients with diabetes.

Andrews noted that the committee did not support the diabetes foot exam measure CBE 0056, which looks at the percentage of patients who receive a visual inspection and a pulse exam. “There’s not high-quality evidence to support” the pulse exam, she said. That foot exam measure has not been used in any CMS accountability program.

It’s important that the group supported A1c control, said O’Connor. “But what about blood pressure and cholesterol and smoking?”

The committee didn’t support the diabetes care measure CBE 0729, a composite measure designed for group practices that includes tobacco use, because of its uncertain validity. Additionally, the measure takes an all-or-nothing approach that doesn’t differentiate between group practices that achieve most of its components from those who achieve none.

The ACP panel did not support the blood pressure control measure CBE 0061, saying it was not designed for use by physicians or group practices. “There are concerns about the numerator only counting the most recent BP [ blood pressure] reading,” which ignores any improvements, the paper noted. Averages should be allowed, the authors wrote. 

The committee also did not support a measure tracking the use of cholesterol-lowering statin therapy for patients with diabetes (HEDIS SPD). Clinicians should not be held accountable for statin adherence, the committee wrote. “That reflects logistics and social drivers of health, rather than quality of care,” it said.

That “doesn’t mean we would suggest not using a statin in a diabetic because that is guideline treatment,” Andrews said.

Seeking Collaboration, Not Division

The review, in line with all of ACP’s performance measure evaluations, is meant to encourage quality organizations to tweak their measures, Andrews said.

“When we write these papers, we try to say, ‘here’s the good about the measure, here’s the bad about the measure,’” she said. “This is our advice for what could change to make it better and actually an effective performance measure.”

New Measure Proposed

The ACP committee also suggested a new measure that “brings us into 2025 with where care for diabetes should be,” Andrews said.

The new measure would track the prescription of an SGLT2 inhibitor or GLP-1 agonist to go along with metformin and lifestyle modifications in adults with type 2 diabetes who have inadequate glycemic control.

Physicians would be measured on the percentage of patients who were prescribed metformin for 12 months before the encounter and were prescribed a GLP-1 or SGLT2.

Andrews said that ACP officials have met with CMS to discuss the measure as a potential addition to the MIPS program.

Andrews and O’Connor reported no relevant financial relationships.

Alicia Ault is a Saint Petersburg, Florida-based freelance journalist whose work has appeared in many health and science publications, including Smithsonian.com. You can find her on X @aliciaault and on Bluesky @aliciaault.bsky.social.