Objective:

To assess the available evidence about home-based primary care (HBPC) interventions for adults with serious or disabling chronic conditions.

Data sources:

Articles from January 1998 through May 2015 were identified using Ovid MEDLINE^®, CINAHL^®, ClinicalTrials.gov, Cochrane Database of Systematic Reviews, reference lists, and gray literature databases.

Review methods:

We included randomized controlled trials (RCTs) and observational studies of HBPC, including home visits by a primary care provider, longitudinal management, and comprehensive care. Study quality was assessed, data extracted, and results summarized qualitatively.

Results:

We identified 4,406 citations and reviewed 221 full-text articles; 19 studies were included. Two were RCTs, while 17 were observational studies.

The strongest evidence (moderate) was that HBPC reduces hospitalizations and hospital days. Reductions in emergency and specialty visits and in costs were supported by less strong evidence, while no or unclear effects were identified on hospital readmissions and nursing home days. Evidence about clinical outcomes was limited to studies that reported no significant differences in function or mortality. HBPC had a positive impact on patient and caregiver experience, including satisfaction, quality of life, and caregiver needs, but the strength of evidence for these outcomes was low.

In studies that reported on the impact of patient characteristics, moderate evidence indicated that frail or sicker patients are more likely than others to benefit from HBPC. No identified studies assessed the impact of organizational characteristics. No adverse events were reported. Only one study examined the potential for a negative impact; none was found.

The services included in the HBPC interventions varied widely, and no identifiable combination was related to more positive outcomes. We identified four studies that evaluated the addition of specific services. Combining palliative care and primary care home visits increased the likelihood of death at home (2 studies; low strength of evidence), while studies on adding caregiver support (1 study) or transitional care (1 study) to HBPC were rated as having insufficient evidence.

Conclusions:

Current research evidence is generally positive, providing moderate-strength evidence that HBPC reduces use of inpatient care and providing low-strength evidence about its impact on use of other health services, costs, and patient and caregiver experience. Future research should focus on the content and organizational context of HBPC interventions so that experiences can be replicated or improved on by others. Additional research is also needed about which patients benefit most from HBPC and how HBPC can best be used in the continuum of care.