Expert Panel Consensus on the Effectiveness and Implementation of Models to Support Posthospitalization Care Transitions for People With Substance Use Disorders.

Abstract

Objectives: Hospitals are increasingly offering treatment for substance use disorders (SUDs) during medical admissions. However, there is a lack of consensus on the best approach to facilitating a successful transition to long-term medical and SUD care after hospitalization. We aimed to establish a hierarchy of existing SUD care transition models in 2 categories-effectiveness and implementation-using an expert consensus approach.

Methods: We conducted a modified online Delphi study that convened 25 interdisciplinary clinicians with experience facilitating posthospitalization care transitions for patients with SUD. Panelists rated 10 prespecified posthospitalization care transition models according to 6 criteria concerning each model's anticipated effectiveness (eg, linkage to care, treatment retention) and implementation (eg, feasibility, acceptability). Ratings were made on a 9-point bidirectional scale. Group consensus was determined using the interpercentile range adjusted for symmetry.

Results: After 3 rounds of the Delphi process (96% retention across all 3 rounds), consensus was reached on all 60 rating criteria. Interdisciplinary addiction consult teams (ACTs) and in-reach from partnering outpatient clinics were rated highest for effectiveness. Interdisciplinary ACTs and bridge clinics were rated highest for implementation. Screening, brief intervention, and referral to treatment; protocol implementation; and postdischarge outreach received the lowest ratings overall. Feasibility of implementation was perceived as the largest challenge for all highly rated models.

Conclusions: An expert consensus approach including diverse clinician stakeholders found that interdisciplinary ACT, in-reach from partnering outpatient clinics, and bridge clinics had the greatest potential to enhance posthospitalization care transitions for patients with SUD when considering both perceived effectiveness and implementation.