Summary: In adults 75 years and older with type 2 diabetes (HbA1c 8.0% or lower) treated with insulin and/or sulfonylureas, adding a patient previsit activation handout to physician academic detailing modestly increased diabetes medication deprescribing at 6 months compared with academic detailing alone (15.8% vs 9.0%; adjusted risk difference 7.5%, 95% CI 1.5% to 13.6%; P = .01). Severe self-reported hypoglycemia did not differ significantly between arms.
PICO Summary
| Element | Detail |
|---|---|
| Population | 450 patients aged 75 years and older with type 2 diabetes, HbA1c 8.0% or lower, on insulin and/or sulfonylureas (mean age 79.9 years, 49.6% female, mean HbA1c 7.5%), cared for by 211 primary care physicians; randomized clinical trial in one integrated health system in Northern California, United States (2020 to 2024). |
| Intervention | Physician academic detailing plus a patient previsit activation deprescribing handout delivered before the visit (n = 232). |
| Comparison | Physician academic detailing plus an attention-control healthy lifestyle handout (academic-detailing-only arm; n = 218). |
| Outcome | Diabetes medication deprescribing at 6 months: 15.8% (36/232) vs 9.0% (19/218); adjusted risk difference 7.5%, 95% CI 1.5% to 13.6%, P = .01 (absolute risk reduction 6.8%, approximate NNT 15). Effect persisted at 12 months: 22.8% vs 16.3%; adjusted risk difference 7.9%, 95% CI 0.4% to 15.5%, P = .04. Severe self-reported hypoglycemia at 6 months: 4.7% vs 6.5%; adjusted risk difference -2.3%, 95% CI -7.1% to 2.5% (no significant difference). |
Previsit Activation and Diabetes Deprescribing
RCT · type 2 diabetes · 75+ yrs · 6 months
A low-cost previsit activation handout modestly raised deprescribing of insulin/sulfonylureas in older adults (15.8% vs 9.0%) without increasing severe hypoglycemia.
Expert Commentary
This pragmatic randomized trial offers a measured verdict: a low-cost previsit handout that primes older patients to raise deprescribing with their physician produced a real but modest increase in stopping or reducing insulin and sulfonylureas, on a background where every prescriber had already received academic detailing. The comparison was therefore activation versus an attention-control handout, not against usual care, so the 7.5% absolute risk difference reflects the incremental value of patient engagement rather than the whole intervention. With deprescribing reached in only about one in six activated patients and roughly fifteen patients needing the handout for one additional deprescribing event, the signal is genuine yet small, and most regimens were left unchanged. The trial was conducted in a single well-integrated health system with embedded pharmacy and panel-management infrastructure, which is the main limitation on generalizability; results may not transfer to fragmented or under-resourced settings. Reassuringly, severe hypoglycemia did not increase, though the study was not powered to confirm a safety benefit and outcome ascertainment was unblinded by design. Can I use this with my patients? Yes, for older adults with tightly controlled diabetes on hypoglycemia-prone agents, a structured previsit prompt is a sensible, safe nudge. Clinicians should pair it with proactive prescriber-led review rather than relying on patient activation alone.
References
Grant RW, Peterson I, McCloskey JM, Lipska KJ, Nugent J, Karter AJ, Gilliam LK. Diabetes Deprescribing in Older Adults: A Randomized Clinical Trial. JAMA Intern Med. 2025;185(8):926-935. doi:10.1001/jamainternmed.2025.2015