OBJECTIVES : Seventeen medical homes (MH) were established in the Local Health Authority (LHA) of Parma, Emilia-Romagna, Italy, between 2011 and 2016. This cohort study estimates changes in healthcare utilization associated with MH implementation.

METHODS : Residents were identified in the Parma LHA administrative healthcare database for the period 01/2011-12/2017 who lived in Parma ≥1 year, were ≥14 years of age, and had a documented general practitioner (GP) in Parma LHA. Exposure to MH care was time-varying. Rates of healthcare utilization were compared using Cox proportional hazards or logistic regression. Additional analyses explored potential heterogeneity in the exposure-outcome relationships for 1) MH implemented early (2011-2012) versus late (after 2012), 2) the first year versus later years after MH implementation, and 3) MH having low versus medium-high complexity, in terms of breadth of their on-site services.

RESULTS : Over 7 years, the study followed 431,378 residents; 22% of their total person-time was under exposure to MH. Changes in admissions and ambulatory care-sensitive conditions admissions were not generally associated with MH implementation. MH implementation was associated with a 7% decrease in emergency department (ED) utilization (HR=0.93; 95% CI: 0.92-0.94). Modestly greater decreases in certain utilization measures were observed among patients receiving care from GP’s in a MH implemented after 2012, a MH that had been open longer than 1 year, or a MH having medium-high complexity.

CONCLUSIONS : Accounting for pre-existing differences in utilization, this MH initiative was associated with some modest improvements in utilization measures, some of which depended on when the MH was implemented, how long it had been open, or its complexity. Most notably, exposure to MH care was associated with significantly lower rates of ED utilization. Understanding these associations may provide important information for planning and refining MH efforts.