Nurse-Delivered Telehealth Falls Short for Diabetes in Fee-for-Service Trial

A comprehensive, nurse-delivered telehealth program for patients with hard-to-control type 2 diabetes (T2D) and hypertension did not significantly lower hemoglobin A_1c (HbA_1c) compared with a simpler self-monitoring approach when delivered inside a fee-for-service (FFS) health system, according to a pragmatic randomized trial published in Annals of Internal Medicine.¹ The findings, released alongside an accompanying editorial,² complicate the case for exporting telehealth models that have worked well inside vertically integrated systems like the Veterans Health Administration (VA) into the FFS settings that make up most of US health care.

The EXTEND (Expanding Technology-Enabled, Nurse-Delivered Chronic Disease Care) trial enrolled 220 adults across 6 Duke University Health System (DUHS) primary care and endocrinology clinics, randomizing them evenly to either a self-monitoring control program or EXTEND Plus, a 12-month intervention combining mobile monitoring, self-management coaching, and pharmacist-supported medication management delivered by nurses.

How the Intervention Impacted Glycemic Control

At 12 months, mean HbA_1c fell by 1.1 percentage points in the comprehensive telehealth arm vs 0.7 percentage points with self-monitoring alone. This amounts to a between-group difference of −0.4 percentage points (95% CI, −1.0 to 0.3) that did not reach statistical significance. Among secondary outcomes, only diabetes self-care scores showed a statistically significant advantage for the telehealth group. Directional but nonsignificant improvements also appeared for systolic and diastolic blood pressure (BP) and weight.

The result stands in contrast to earlier VA-based work from the same investigators, in which a similar nurse-delivered comprehensive telehealth model produced meaningful HbA_1c reductions in veterans with persistently uncontrolled T2D. That VA success has since been folded into routine clinical practice at the agency, the study authors noted, underscoring how differently an intervention can perform once it leaves a system built around integrated, salaried care delivery.

Operational Barriers Drove a Fidelity Shortfall

The trial's implementation data may explain much of the gap. Nurses completed a median of just 9 of the targeted 12 encounters per participant, with only 20% reaching full fidelity. In exploratory analyses, participants who completed at least 7 encounters saw meaningfully larger HbA_1c improvements than those completing 6 or fewer, hinting that an adequately dosed version of the program might have performed better.

Semistructured interviews with patients and focus groups with nurses and clinical pharmacist practitioners (CPPs) pointed to concrete operational barriers: Nurses struggled to fit study encounters into already-full clinical schedules and to reach patients at scheduled times, while CPPs described difficulty balancing telehealth duties against competing clinical demands.

What This Means for Fee-for-Service Reimbursement Models

The trial's authors point toward payment structure as the main reason behind the results. Because FFS reimbursement is built around discrete, billable clinic encounters rather than continuous remote management, staff assigned to the telehealth program faced what the authors called "intractable encroachment" from competing clinical duties.

Furthermore, according to research published in The American Journal of Managed Care^® (AJMC^®), rural hospital chief financial officers have identified lack of payment parity between telehealth and in-person visits, along with broader uncertainty about the reimbursement environment, as key barriers preventing health systems from investing in and optimizing telehealth programs—concerns that closely parallel the FFS-driven workflow conflicts seen in the EXTEND trial.³ More recent AJMC coverage of Medicare claims data found that expanded telemedicine flexibilities have not driven runaway utilization or spending across FFS, Medicare Advantage, Medicaid, or commercial populations, suggesting that financial risk to payers from broader telehealth reimbursement may be smaller than once feared.⁴ Taken together, these findings suggest that the obstacle to comprehensive telehealth in FFS settings may lie less in payer-side cost concerns and more in how reimbursement structures shape staffing and workflow on the delivery side.

What's Next for Comprehensive Telehealth in FFS Systems

Future replication and implementation efforts need built-in mechanisms to monitor fidelity in real time, anticipate workflow conflicts before launch, and test reimbursement arrangements that protect dedicated staff time rather than asking clinicians to absorb new responsibilities on top of existing productivity targets. The study authors also point to emerging centralized virtual care support models as a possible infrastructure fix for systems lacking the technical scaffolding to integrate patient-generated data into the electronic health record efficiently.

References

1. Crowley MJ, Lewinski AA, Yang Q, et al. Expanding technology-enabled, nurse-delivered chronic disease care: a pragmatic, randomized, effectiveness–implementation trial. Ann Intern Med. Published online June 23, 2026. doi:10.7326/ANNALS-26-00132
2. Naylor MD, Hirschman KB. Pragmatic trials: the importance of foundational research, fidelity, implementation, and context. Ann Intern Med. Published online June 23, 2026. doi:10.7326/ANNALS-26-01922
3. Uscher-Pines L, Sousa JL, Zachrison KS, Schwamm L, Mehrotra A. Financial impact of telehealth: rural chief financial officer perspectives. Am J Manag Care. 2022;28(12):e436-e443. doi:10.37765/ajmc.2022.89279
4. Shaw ML, Hohmann E. Telehealth didn't break the bank—and the data prove it. AJMC. May 10, 2026. Accessed June 22, 2026. https://www.ajmc.com/view/telehealth-didn-t-break-the-bank-and-the-data-prove-it