Will Medical Homes Impact Health Care Services Utilization and Costs? Lessons from Family Medicine Groups (Canada)

Monday, June 23, 2014: 3:20 PM
Von KleinSmid 102 (Von KleinSmid Center)

Author(s): Mehdi Ammi

Discussant: Coady Wing

Context:

Patient-Centered Medical Homes are one of the prominent models promoted by the Affordable Care Act to strengthen primary care and improve health system-level outcomes. Knowledge is however particularly sparse regarding the impacts of these organizations, especially on health services utilization and health care costs. Canada’s experience may be informative in this respect. During the last decade, many Canadian provinces have implemented primary care reforms based on the key features of Medical Homes.

Objectives:

We investigate the impacts of a new form of primary care delivery: Quebec’s Family Medicine Groups (FMGs). More precisely, we examine the effects on health care services utilization and costs in outpatient (general practitioner, specialist and emergency department) and inpatient settings. Like Patient-Centered Medical Homes, FMGs include extended hours, patient enrolment and multidisciplinary teams, but they maintain the same remuneration scheme (fee-for-service) as outside FMGs.

Methods:

We built a panel of administrative data of the population of elderly and chronically ill patients, characterizing all individuals as FMG enrollees or not. This 7-year panel contains information on about 110,000 FMG patients and 700,000 non FMG patients. Participation in FMGs is voluntary and we address potential selection bias by combining propensity scores based on pre-treatment characteristics and inverse probability of treatment weighting with difference-in-differences and fixed effects models.

Results:

We find that FMGs significantly decrease patients’ health care services utilization and costs in outpatient settings relative to patients not in FMGs over five years of follow-up after enrollment. The number of primary care visits decreased by nearly 2% among FMG enrollees and declines in costs was of equal magnitude. The number of visits to specialists declined by 0.7% while costs decreased by 1%. We find a significant decrease in the number of emergency department visits, though associated costs remained stable. The impacts are much larger in the first year after enrollment, but do not persist over a longer period. No significant effect of FMGs was found in the inpatient setting.

Conclusion:

These results provide support for the idea that changing primary care organization and delivery can have impacts on the rest of the health care system. They suggest that Patient-Centered Medical Homes may impact health care services utilization and costs but raise concerns regarding the persistence of the effects over time.