The majority of those studies found overall cost reductions and decreases in ER visits, and 40 percent reported fewer hospital admissions. Smaller percentages of the studies reported improvements in other vital areas, including improved population health or increased use of preventive services (30 percent), improvements in access to care (25 percent), patient satisfaction (20 percent), and hospital readmissions (15 percent).
Although that report reinforced the concept of the PCMH as the future model of primary care and provided a five-year summary of evidence supporting the model(www.pcpcc.org), it didn't generate much media attention. Good news rarely does.
Fast forward one month. The Journal of the American Medical Association (JAMA) released one PCMH study(jama.jamanetwork.com) in late February that found "limited improvements in quality" and no association with reductions in use of hospital, ER or ambulatory care services or total costs. Based on that one study, reports in some daily newspapers and on medical news websites declared the PCMH a failure. If the PCMH could be likened to a horse, it would be fair to say that some particularly critical bloggers were ready to take it out back and shoot it.
Well, put your shovels away. There will be no burial today.
The PCMH is a work in progress. The AAFP has been developing the model for 10 years, a result of the original Future of Family Medicine project. Despite the findings of this one study, we still think the PCMH is a vital part of our specialty's future. The Academy and others will continue to study and refine it.
Mark Friedberg, M.D., the corresponding author of the JAMA article, said in an interview with AAFP News that physicians should not "attach too much importance to any study" and that more research on the PCMH is needed.
It's worth noting that the Pennsylvania Chronic Care Initiative, the program studied in the JAMA article, launched in 2008 under PCMH standards that were updated by the National Committee for Quality Assurance (NCQA) three years later. NCQA plans to update its standards again(www.ncqa.org) this month. Again, it's a work in progress. We know more now than we did in 2008 about what works, what's required and how to implement it.
The JAMA study has other limitations, too.
The Pennsylvania practices involved in the study were given incentives to earn PCMH recognition but no incentives to reduce costs. The project also failed to target chronically ill patients. With no incentive to control costs and no priority to help improve the health of the system's biggest users of care, should it be a surprise that costs did not decline?
Becoming recognized as a PCMH does not magically improve care and reduce costs. There is more to it than marking off the boxes on a checklist. For example, improving access to care is an element of the PCMH, but practices in the study didn't offer evening hours or weekend care because that step isn't required to earn PCMH recognition. But that kind of patient-centered effort can lead to reduced health care costs and less use of ERs and hospitalization.
Furthermore, only half the NCQA-recognized practices in the study achieved Level 3. According to NCQA, 75 percent of recognized practices in pilot projects typically achieve Level 3. "A higher (and more typical) concentration of high-capability medical homes would probably have produced better outcomes," NCQA said in its response to the study(www.ncqa.org).
The PCPCC also responded(www.pcpcc.org), saying the study failed to measure other key elements of the PCMH, including patient engagement and satisfaction, team-based care, and behavioral health integration.
The Commonwealth Fund, which helped fund the Pennsylvania study, also defended the PCMH(www.commonwealthfund.org), making some of the same points we have here: that we better understand how to implement the model than we did in 2008, and patients with complex conditions should be targeted to improve outcomes and reduce utilization and costs. It also points out that the three-year study may not have been long enough to see the long-term benefits of the model.
Since the JAMA study was initiated in 2008, six other regions of Pennsylvania have started their own pilots. Based on lessons learned from the first study, these subsequent projects have been updated accordingly. It will be interesting to see the findings when these more recent projects are completed.
The bottom line is that one study that questions the value of the PCMH is no more valuable than one study that supports it. No one study should drastically alter the way we practice medicine. It is the totality of evidence that matters, and right now, we have a mountain of evidence that shows that in the majority of studies, the PCMH has been shown to decrease costs and improve outcomes.
One study can get our attention, which this one surely did, and make us scrutinize what we are doing. The JAMA article could help us learn how to better implement and study this evolving model, but it should not lead us to abandon it.