Higher VA Telehealth Hub Use Linked to Fewer Community Care Referrals

Key Clinical Summary

• US Department of Veterans Affairs (VA) mental health clinics with high Clinical Resource Hub (CRH) utilization had 20 fewer community care (CC) referrals per 1000 patients per month.
• Clinics with any CRH use showed a small increase in CC referrals, suggesting resource constraints drive dual use.
• Findings indicate effective CRH implementation may reduce reliance on VA-purchased community mental health care.

A large VA cohort study found that greater use of CRHs—a telehealth-based staffing model—was associated with reduced referrals to CC for mental health services. The longitudinal analysis evaluated whether CRH expansion improves internal care access or complements external referrals across VA clinics.

Study Findings

The study analyzed data from 1 120 250 veterans receiving outpatient mental health care across 1149 VA clinics between 2018 and 2019, with follow-up through September 2023. Patients had a mean age of 60 years, and 16% were female.

Using difference-in-differences (DID) analyses, investigators compared CC referral rates before and after CRH implementation. Clinics were categorized by CRH use, including high-penetration (top 25%) and low-penetration (bottom 25%) groups.

Clinics with any CRH utilization experienced a small but statistically significant increase in CC referrals compared with clinics without CRH use (DID, 0.525; 95% CI, 0.181-0.868; P = .003). Researchers suggest these sites may be under-resourced and rely on both CRH and CC to meet demand.

In contrast, clinics with high CRH penetration demonstrated substantially fewer CC referrals than low-penetration clinics by the end of the study period. Specifically, high-use clinics had a mean of 20 fewer CC referrals per 1000 patients per month (DID, −20.00; 95% CI, −21.90 to −18.20; P < .001).

High-penetration sites were more often smaller, rural community-based outpatient clinics, where CRH services may integrate more easily into workflows.

Clinical Implications

These findings suggest that tele-mental health hub models like CRHs can reduce dependence on external care networks when fully implemented. For VA clinicians and administrators, the distinction between partial and high CRH adoption is critical: simply introducing CRHs does not immediately decrease CC referrals, but deep integration into clinical workflows does.

Reduced reliance on community care may help mitigate care fragmentation, administrative complexity, and higher costs associated with external referrals. Maintaining care within the VA system may also improve continuity and coordination, particularly for patients with complex psychiatric needs.

The results are especially relevant for rural and underserved settings, where staffing shortages often limit access to mental health services. Telehealth hubs can expand capacity without requiring local workforce expansion.

Beyond the VA, similar hub-based telehealth models are increasingly used in other health systems. This study supports their role in retaining patients within integrated care networks, which may enhance quality and patient satisfaction while reducing system-level inefficiencies.

Study authors noted that “successful implementation of CRHs may be associated with decreased reliance on CC,” emphasizing that high-performing sites had fully integrated CRH services into routine care delivery. They also highlighted that clinics with only partial CRH use likely represent under-resourced facilities relying on multiple pathways to meet patient demand.

Conclusion

In VA mental health clinics, high utilization of CRHs was associated with significantly fewer community care referrals, underscoring the importance of robust telehealth infrastructure. Future efforts should focus on optimizing CRH implementation to improve access while reducing fragmentation in mental health care delivery.

Reference

Connolly SL, Jaske EL, Wheat C, et al. Veterans affairs clinical resource hubs and rates of mental health community care referrals. JAMA Netw Open. 2026;9(2):e2560084. doi:10.1001/jamanetworkopen.2025.60084