Tools to improve referrals from primary care to specialty care.

Abstract

OBJECTIVES: Referrals from primary to specialty care are a critical first step in coordination of specialty care, but shortcomings in the appropriateness, clarity, or completeness of referrals are common. We examined (1) whether 3 tools to coordinate specialty care are associated with better referral characteristics and (2) whether greater perceived helpfulness of these tools is associated with better referral characteristics among specialists who use all 3 of them. STUDY DESIGN: National online survey about care coordination among medical specialists receiving referrals in the Veterans Health Administration. METHODS: Adjusted odds ratios (ORs) for associations between use and helpfulness of 3 coordination tools (service agreements, referral templates, and e-consults) and perceived frequency of 3 referral characteristics (appropriateness, clarity, and completeness). RESULTS: Among specialists (N = 497), use of referral templates was associated with perceptions that referrals were more frequently appropriate (adjusted OR, 1.5; 95% CI, 1.0-2.4), clear (adjusted OR, 1.6; 95% CI, 1.0-2.5), and complete (adjusted OR, 1.9; 95% CI, 1.1-3.2). Use of e-consults was associated with more frequent referral clarity (adjusted OR, 1.7; 95% CI, 1.0-3.0). Among specialists using all 3 tools, those reporting that templates were very helpful also perceived more frequent referral clarity (adjusted OR, 3.1; 95% CI, 1.1-8.5) and completeness (adjusted OR, 3.6; 95% CI, 1.5-8.7). Service agreements were not associated with any referral characteristic. CONCLUSIONS: Well-designed referral templates may help improve the clarity and completeness of primary care-specialty care referrals. Existing templates may provide models that can be adapted in collaboration with primary care and broadly applied to improve referrals. Work is needed to improve the impact of service agreements and e-consults on referrals.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 26 citations exist upstream.