Summary: This is the published protocol and baseline description of a pragmatic, three-arm cluster-randomized trial in which 191 primary care providers were randomized to peer-champion support, peer-champion support plus insurance-coverage information, or usual care, with the planned primary outcome being the rate of new SGLT2i or GLP-1RA prescriptions from 3 days before to 28 days after a targeted visit. No efficacy, prescribing, cardiovascular, or renal outcome results are reported; the paper describes the design and a baseline cohort of 1,079 patients only.
PICO Summary
| Element | Detail |
|---|---|
| Population | Cluster-randomized at the clinician level: 191 primary care providers in the United States and their patients with type 2 diabetes who had sub-optimally controlled disease and an indication for one of these medication classes. Baseline cohort of 1,079 patients who attended a visit and will enter the primary analysis; mean age 66 years, 46% female, 61% white, 16% Hispanic. |
| Intervention | Two intervention arms. Arm 1 (66 providers, 484 patients): peer-champion support with pre-visit electronic health record messages containing prescribing tips. Arm 2 (63 providers, 446 patients): peer-champion support plus insurance-coverage information, with EHR messages identifying the most affordable in-class option and offering prior-authorization support. |
| Comparison | Usual care (no intervention): 62 providers, 459 patients. |
| Outcome | Planned primary outcome only: a new prescription for an SGLT2i or GLP-1RA medication, captured from 3 days before through 28 days after the targeted visit, in each intervention arm versus control. This is a protocol and design paper; no effect estimates, confidence intervals, p-values, ARR, or NNT are reported, and no cardiovascular or renal outcomes are measured. Registered as NCT05463705. |
Expert Commentary
This paper is a trial protocol, not a results paper, and it should be read as such. The verdict is that no clinical conclusion can yet be drawn, because nothing has been tested: the document describes the rationale, the three-arm pragmatic design, the randomization of 191 primary care providers, and a baseline cohort of 1,079 patients, with the primary outcome defined as a prescribing event rather than any cardiovascular or renal endpoint. The premise is sound, since underuse of SGLT2 inhibitors and GLP-1 receptor agonists in eligible patients is well documented and the listed barriers, including insurance coverage and diffusion of prescribing responsibility, are real. The most important limitation for interpretation is simply that the prescription rate is a process measure; an increase in prescribing would be encouraging but would not itself demonstrate that heart or kidney events are reduced, and the trial is not powered or designed to show that. Can I use this with my patients? Not yet, because there are no outcome data here to act on; the practical signal is only that peer-champion and affordability-focused nudges are being formally evaluated. Clinicians should watch for the forthcoming results before drawing any inference about whether these implementation strategies actually change prescribing, and should resist reading cardiorenal benefit into a design paper. The honest takeaway is to wait for the reported trial findings.
References
Haff N, Horn DM, Bhatkhande G, Sung M, Colling C, Wood W, et al. Encouraging the prescribing of SGLT2i and GLP-1RA medications to reduce cardiovascular and renal risk in patients with type 2 diabetes: Rationale and design of a randomized controlled trial. Am Heart J. 2025;285:39-51. doi:10.1016/j.ahj.2025.02.007
