Effects of Home-Based Primary Care on Hospital Use for High-Need Medicare Patients: an Observational Study.

Background: High-need, high-cost Medicare patients can have difficulties accessing office-based primary care. Home-based primary care (HBPC) can reduce access barriers and allow a clinician to obtain valuable information not obtained during office visit, possibly leading to reductions in hospital use. Objective: To determine whether HBPC for high-need, high-cost patients reduces hospitalizations and Medicare inpatient expenditures. Design: We conducted a matched retrospective cohort study using a difference-in-differences analysis to examine patients 2 years before and 2 years after their first home visit (HBPC group). Participants: The study included high-need, high-cost fee-for-service Medicare patients without prior HBPC use, of which 55,303 were new HBPC recipients and 156,142 were matched comparison patients. Intervention: Receipt of at least two HBPC visits and, within 6 months of the index HBPC visit, a majority of a patient's primary care visits in the home. Main Measures: Total and potentially avoidable hospitalizations and Medicare inpatient expenditures. Key Results: HBPC reduced total hospitalization rates, but the marginal effects were not statistically significant: a reduction of 11 total hospitalizations per 1000 patients in the first year (− 0.6%, p = 0.19) and 14 in the second year (− 0.7%, p = 0.16). However, HBPC reduced potentially avoidable hospitalization rates in the second year. The estimated marginal effect was a reduction of 6 potentially avoidable hospitalizations per 1000 patients in the first year (− 1.6%, p = 0.16) and 11 in the second (− 3.1%, p = 0.01). The estimated effect of HBPC was a small decrease in inpatient expenditures of $24 per patient per month (− 1.1%, p = 0.10) in the first year and $0 (0.0%, p = 0.99) in the second. Conclusions: After high-need, high-cost patients started receiving HBPC, they did not experience fewer total hospitalizations or lower inpatient spending but may have had lower rates of potentially avoidable hospitalizations after 2 years. [ABSTRACT FROM AUTHOR]