ORCID

Zachary Bridgewater: 0000-0002-6830-0603

Joshua-Paul Miles: 0000-0002-6612-742X

Catherine Annis: 0000-0002-7311-1612

Julia Carboni: 0000-0001-6037-9276

Gilly Cantor: 0000-0001-8890-9259

Michelle Shumate: 0000-0002-4998-0463

Document Type

Poster

Date

4-9-2026

Keywords

Coordinated care, Referrals, Referral networks, Network effectiveness, Network functioning, Veterans

Campus Community

D'Aniello Institute for Veterans and Military Families

Language

English

Funder(s)

Army Research Office

Funding ID

W911NF-20-1-0202

Acknowledgements

This research was funded by Army Research Office (ARO) grant W911NF-20-1-0202.

Disciplines

Military and Veterans Studies | Public Affairs, Public Policy and Public Administration | Social and Behavioral Sciences

Description/Abstract

Coordinated care is an interorganizational network approach to care that connects individuals to diverse services (e.g., housing, income assistance, job support, social and spiritual support, etc.) via human navigators and shared referral platforms. This approach has become prevalent in the military-connected sphere to help service members, veterans, and their families (SMVF) access care. Yet, prior research suggests variation in the resolution of coordinated care referrals"”25% for one United Way 211 (Boyum et al., 2016), and 66% and 88% for two Unite Us networks (Drake et al., 2024; White et al., 2025). This research explores why some referrals resolve while others do not by examining how client demographics (i.e., age, sex, race, ethnicity), client military record (i.e., relationship to military, discharge character), and network characteristics (i.e., wait time, accuracy) influence request resolution.

This study uses a retrospective dataset of Unite Us requests for services across 11 networks following the AmericaServes care coordination model (i.e., an SMVF-focused approach) from October 2015 through September 2022. A generalized linear mixed model showed significant but weak effects of clients' demographics and military record on request resolution. Networks' typical accuracy (i.e., rejection volume) had a significant negative effect on request resolution. Typical wait time had no significant main effect; however, it interacted with accuracy such that longer wait times had higher odds of request resolution for networks with lower accuracy, and vice versa.

These findings offer two broader takeaways. First, the weak effects of demography suggest AmericaServes provides equitable treatment to clients of varying backgrounds. Second, the results related to rejection suggest that adequate resourcing and network knowledge play critical roles in promoting care access. Insufficiently resourced networks and poor network awareness are both mechanisms that could increase rejection rate. Pausing to build resources or knowledge offers promise as solutions to recover resolution rates.

Download

Share

COinS