Early Medical Home Demo Evidence Not So Great

The Annals of Family Medicine has several articles discussing the National Demonstration Project testing the patient-centered medical home.  (Annals Article) The project involved 36 family practices which were selected to implement a medical home model; half the practices were pretty much on their own and half had help from a facilitator.  In 2002, seven family medicine organizations met to discuss a new model of primary care.  As  as a result of this initiative, the American Academy of Family Physicians formed a subsidiary called TransforMED to test the new model of practice.  The National Demonstration Project is being directed by TransforMED.

Several of the Annals papers are just descriptive of the project, the evaluation methods and the process of implementation.  One lesson appears to be that flexibility is needed as implementation goes forward.  Basically, self-implementing practices appeared to get as many components of the medical home in place as did facilitated practices, although the facilitated ones had a greater ability to make and sustain change.  Neither group got more than 70% of the model components in place over two years.  Interestingly, many practices found several of the recommended model components to have little value or utility, including e-visits, group visits and team-based care.

Patients’ perceptions of the practices actually declined slightly over the course of the project, and declined in both the self-implementing and facilitated groups.  Process-of-care measures improved modestly over the course of the project, but were not compared to groups not participating in the project so it is hard to know if the improvement is due to the implementation of the medical home or other factors.  Overall, a great deal of effort seems to have been expended for uncertain benefits.

One of the most discouraging things about the NDP and the current analysis of its status is that there is no economic analysis of its effect on either administrative or health costs.  While the medical home’s effect on quality of care and health outcomes is critical, understanding its overall effect on health spending is also very important.  Perhaps future analyses will fill this gap.  And more work is needed on actual health outcomes; very little was presented in the Annals papers.  The medical home seems like a good, common-sense approach but data is necessary to see how it actually functions in everyday practice.