Summary: REMIND-DM is the rationale, design, and baseline-characteristics report of a pragmatic, prospective, cluster-randomised quality improvement trial. Across 23 percutaneous coronary intervention (PCI) sites caring for 7,045 patients with type 2 diabetes within the Blue Cross Blue Shield of Michigan (BMC2) collaborative, sites were randomised to a templated PCI-report reminder of medication eligibility linked to a decision-support tool versus usual care. The primary outcome, new prescription of an SGLT2 inhibitor or GLP-1 receptor agonist at discharge, has not yet been reported, so no efficacy estimates are available from this publication.
PICO Summary
| Element | Detail |
|---|---|
| Population | 7,045 patients with type 2 diabetes undergoing coronary angiography/PCI across 23 PCI sites in the Blue Cross Blue Shield of Michigan (BMC2) quality improvement collaborative, United States. Cluster-randomised at the site level. |
| Intervention | A low-touch quality improvement strategy: a templated PCI-report “reminder” flagging eligibility for cardioprotective glucose-lowering therapy, linked to a decision-support tool. Applied at intervention-arm sites over a 6-month evaluation period after a 6-month baseline period. |
| Comparison | Usual care at control-arm sites, without reminder prompts (23 sites randomised in total between the two arms). |
| Outcome | Primary outcome: new prescription of an SGLT2 inhibitor or GLP-1 receptor agonist among eligible patients at discharge after PCI, analysed by a difference-in-difference design (baseline vs evaluation period, both arms). No effect estimate, 95% confidence interval, p value, ARR, or NNT is reported: this is a design and baseline paper and outcome analyses are prospective and not yet available. |
Expert Commentary
This publication should be read as a protocol and baseline-characteristics report, not as evidence of benefit. The verdict is therefore that no conclusion about effectiveness can yet be drawn. A genuine evidence gap is being addressed, because SGLT2 inhibitors and GLP-1 receptor agonists remain underprescribed in high-risk patients with type 2 diabetes despite strong cardiovascular evidence, and the post-PCI encounter is a plausible moment to intervene. The design is methodologically reasonable: cluster randomisation at the site level reduces contamination, and a difference-in-difference comparison across a defined baseline and evaluation period helps account for secular trends in prescribing. The principal limitation is that, by nature, a reminder embedded in a report is an open, unblinded quality improvement intervention, so any future effect could be inflated by co-interventions, secular uptake, or site-level confounding, and the difference-in-difference approach mitigates but does not eliminate this. Outcomes are also measured as prescription at discharge rather than sustained use or hard cardiovascular endpoints. Can I use this with my patients? Not yet, because no outcome data exist; for now it informs how a health system might design an implementation programme rather than what to prescribe at the bedside. The trial is investigator-led within a payer-supported collaborative, with no commercial manufacturer sponsorship described. The sensible course is to await the reported primary outcome before drawing any inference about whether these reminders change prescribing.
References
Chatur S, Seth M, Casey M, Thompson MP, Qureshi MI, Gupta V, et al. Reminders embedded in PCI reports to optimize discharge diabetes mellitus care (REMIND-DM): Rationale, design, and baseline characteristics. Am Heart J. 2025;285:12-20. doi:10.1016/j.ahj.2025.01.020