2017 HSR&D/QUERI National Conference

4094 — Preventing Hospitalization with Home-Based Primary Care: Which Patients Benefit Most?

Lead/Presenter: Samuel Edwards, COIN - Portland
All Authors: Edwards ST (VA Portland) Saha S (VA Portland) Prentice JC (HSR&D QUERI Center for Access Policy Evaluation and Research (CAPER), VA Boston) Pizer SD (VA Boston)

Objectives:
Home-based primary care (HBPC) can reduce the rate of hospitalization for high-risk patients, but appropriate patient selection is critical to maximize effectiveness. We examined how medical complexity modified the relationship between enrollment in HBPC and hospitalizations for ambulatory care sensitive conditions (ACSC) for patients with diabetes, and whether the impact of HBPC on hospitalizations varied by clinical condition.

Methods:
We performed a retrospective cohort study of 364,972 VA patients over age 67, with diabetes mellitus and enrolled in Medicare. Our main exposure was enrollment in HBPC. Our outcomes were hospitalizations in VA or non-VA hospitals due to an ACSC as defined by the Agency for Healthcare Research and Quality's Prevention Quality Indicators. We used instrumental variables regression models to estimate the impact of HBPC enrollment on hospitalization for ACSCs overall, and in subgroups stratified by medical complexity. We also estimated models for each ACSC to determine which conditions were most sensitive to HBPC. We used distance from the Veteran's residence to the nearest HBPC site as our instrument. Patients were followed from 1/1/2006 to 12/31/2010.

Results:
Overall, HBPC was associated with decreased ACSC hospitalizations (OR 0.35 per person-month, 95% CI 0.30-0.42). For patients in the highest quartile of medical complexity, HBPC enrollment was associated with decreased ACSC hospitalizations (OR 0.43, 95% CI 0.19-0.93), while in the lowest quartile, HBPC was associated with increased ACSC hospitalizations (OR 33.2, 95% CI 4.6-240.1). HBPC enrollment was associated with decreased hospitalizations for a range of ACSCs including bacterial pneumonia (OR 0.21, 95% CI 0.14-0.31), dehydration (OR 0.28, 95% CI 0.15-0.51) and congestive heart failure (OR 0.46, 95% CI 0.35-0.60).

Implications:
HBPC enrollment is associated with fewer hospitalizations for a range of ACSCs in patients with diabetes, but only in the most medically complex patients. This demonstrates the importance of appropriate targeting, and suggests HBPC's impact is attributable to its comprehensive approach, rather than condition-specific interventions.

Impacts:
Enrollment in HBPC is associated with a decreased rate of hospitalization for ACSCs in the most medically complex patients.