There’s been great hope that the patient-centered medical home model could help alleviate rising health care costs and improve the quality of care. But results from a large pilot study published in JAMA today found no cost savings and only a modest improvement in quality.
The concept of the patient-centered medical home model is that it redesigns care around the patient’s needs. It offers patients one-stop shopping, with a primary care clinician and interdisciplinary physicians, specialists, nurses, and care managers working together all in one place. For patients with complex or chronic diseases, this may decrease the need for multiple visits to various clinicians, improve coordination of care between clinicians, and give such patients a regular point of contact, such as a care manager who can help them navigate the day-to-day challenges of their condition. To encourage physicians to adopt the model, payers have offered financial incentives.
Numerous pilot studies of the patient-centered medical home model are under way. So far, the data suggest only modest quality improvement over usual care and offer little evidence that this model reigns in costs. Mark Friedberg, MD, MPP, a researcher at RAND, and his colleagues analyzed data from a large multipayer pilot of primary care practices in southeastern Pennsylvania featuring a medical home based on standards set by the National Committee for Quality Assurance. The analysis included 64 243 patients in the pilot practices and 55 959 control patients. The researchers found a modest improvement in the quality of care for only 1 of 11 measures, and no reductions in care utilization or costs.
Dr Friedberg discussed his findings with news@JAMA.
news@JAMA: What do you think is driving the enthusiasm behind medical homes?
Dr Friedberg: Two things have been big drivers of the enthusiasm: a large body of evidence over decades that having a strong primary care system leads to lower costs and higher-quality care, and some hope that medical care homes would help revitalize the field of primary care by attracting students and improving care.
news@JAMA: Prior to your study, had the research backed that up?
Dr Friedberg: It’s too early to tell. The research is very limited and not much is known about how to best implement the model. There is a theoretical model, but taking a real-life practice and transforming it into a medical home hasn’t been well studied.
A few studies of small and medium pilots have been consistent with our results. But some early studies of the medical home model didn’t include things like the National Committee for Quality Assurance criteria. Some, including the ones in large systems, did show some improvement in quality and cost of care.
news@JAMA: What does your study add?
Dr Friedberg: The main addition is that we evaluated a pilot that was larger, ran longer, and had more payers and greater financial incentives than most pilots before it. We found improvement on 1 quality measure [kidney disease screening for patients with diabetes], and a [not statistically significant] trend toward improvement on diabetes care. But we didn’t see overall improvements on screening, utilization, or costs of care. This was despite the practices hiring nonphysician staff to manage patients with chronic conditions.
news@JAMA: What do you think explains the lack of quality improvement or cost reduction?
Dr Friedberg: It’s really hard to say. I don’t think this study can answer that. These pilots are really complex interventions. To pinpoint 1 or a number of things, you would need a large number of studies that change 1 or 2 things. There are dozens of medical home trials under way. Hopefully, in a couple of years we will be in a position to say x, y, or z are necessary for a successful medical home.
news@JAMA: In light of these disappointing results for the medical home model, what do you think are the next steps for research in this area?
Dr Friedberg: We need continued experimentation, to try different combinations of medical home ingredients and to evaluate them. Then we will be in a position to refine this model and improve its success rate. I think it would be a mistake to say, based on this study, that the model can’t work.