Medical Homes Might Not Transform Healthcare

9 Mobile Health Apps Worth A Closer Look

Do medical practices recognized as patient-centered medical homes—in which personal physicians take responsibility for and coordinate all health care--provide better care than traditional medical practices? A new study in the Annals of Family Medicine (AFM) suggests that they don't, and that widespread expectations that the PCMH model will transform healthcare are overblown. Because health IT is central to that model, the questions raised about the PCMH also reflect on the ability of IT to improve the quality of care.

Not surprisingly, PCMH advocates have criticized the study by Leif Solberg and his colleagues, which was published online Nov. 14. Roland Goertz, MD, board chair of the American Academy of Family Physicians (AAFP), commented on the AFM website, "I disagree with Dr. Solberg's final assertion that 'expectations for large and rapid change are probably unrealistic.'" And Paul Grundy, MD, global director of healthcare transformation for IBM, objected to the article because, in his view, it could obstruct efforts to convert more physicians to the PCMH concept.

Solberg responded, "It is a little discouraging to have leaders respond to a scientific study that doesn't completely fit their view of what will help them to sell an idea."

[For more background on e-prescribing tools, see 6 E-Prescribing Vendors To Watch.]

The study that caused all this fuss focused on 21 primary care clinics that are part of the Health Partners Medical Group (HPMG) in Minneapolis. The National Committee on Quality Assurance (NCQA) have recognized all of these practices as level III patient-centered medical homes. The researchers compared the changes in patient satisfaction and quality indicators at these clinics between 2005 and 2009 to changes in similar metrics at non-HPMG practices that had not received NCQA recognition as medical homes.

The patient satisfaction scores at the HPMG clinics rose from 1 to 3 percent per year, depending on the indicator, during the study period. Quality scores for care of diabetes and coronary artery disease, preventive services, and the use of generic drugs increased by 2 to 7 percent per year.

Compared with the non-HPMG groups, the HPMG medical homes achieved slightly higher rates of improvement in patient satisfaction, but only caught up to the community average. The differences between the groups' improvements on the quality indicators were not statistically significant.

The authors concluded, "Achieving medical home recognition was associated with improvements in quality and patient satisfaction for these clinics, but the rate of improvement is slow and does not always exceed levels in the surrounding community in Minnesota (which are also improving). Expectations for large and rapid change are probably unrealistic."

This conclusion was in line with the results of a study of the AAFP's National Demonstration Project; that investigation found that while medical homes did produce small improvements in quality, there was a decline in patient satisfaction at practices transforming themselves into medical homes. On the other hand, some other studies have shown major quality improvements in PCMH clinics.

Goetz cited a North Dakota Blue Cross and Blue Shield program that led to a 64 percent increase in patients with diabetes who received "optimal diabetes care." Patients in North Dakota medical-home practices also had fewer ED visits and hospitalizations than those in non-medical-home practices, he noted.

The problem with the Solberg study, Goetz continued, is that "the Minnesota practices that were studied are located in an atypically advanced health care market." Consequently, all of the groups in the study—whether recognized as medical homes or not—"were very high performing to begin with."

Terry McGeeney, MD, president and CEO of TransforMED, an AAFP spinoff that helps practices become medical homes, criticized the study on another basis: The standards that the NCQA used to recognize practices as medical homes in 2008, he noted, were "woefully inadequate" in measuring patient engagement and "did very little to recognize care management and care coordination in a practice." For this and other reasons, these standards—which have improved somewhat, he added--are not a reliable measure of a real medical home's attributes.

The NCQA's current criteria require the use of health IT in many domains, including the identification and management of patient populations, care management, medication reconciliation, electronic prescribing, care coordination, and performance measurement and improvement.

While the PCMH model has become widely accepted, it is still not being broadly used in practices across the U.S. As of the end of 2010, 7,600 clinicians in 1,500 practices had earned NCQA PCMH recognition.

When are emerging technologies ready for clinical use? In the new issue of InformationWeek Healthcare, find out how three promising innovations--personalized medicine, clinical analytics, and natural language processing--show the trade-offs. Download the issue now. (Free registration required.)