Diabetic retinopathy screening fails not because the test is difficult but because patients do not attend, and most interventions address this one patient at a time. This randomised stepped-wedge trial tested a system-level approach in 5,665 patients with type 2 diabetes overdue for screening.
A teleretinal screening bulk order, placed by a centralised provider for a defined cohort, combined a retinal fundus photograph order with an electronic patient education message. Receiving a bulk order increased the odds of completing screening (odds ratio 1.78, 95% CI 1.49-2.14). Overall completion at six months was 17.5%. Of patients sent the electronic message, 48.7% opened it, and doing so raised completion by 8.5 percentage points and appointments scheduled by 19.1 points.
The absolute completion rate is the sobering part. Even with the intervention, more than four in five overdue patients still had not been screened at six months, and fewer than half opened the message. This is a genuine improvement to an inadequate baseline, not a solution. Its value is that it requires no additional clinician time — the order is placed centrally for a cohort — which makes it the kind of intervention that can be applied at population scale, where per-patient outreach cannot.
- Bulk orders raised odds of screening completion (OR 1.78)
- Opening the message added 8.5 points to completion
- Overall completion still only 17.5% at six months
- Requires no per-patient clinician time — scalable to whole populations
The statistics, in plain English
An odds ratio of 1.78 sounds substantial and translates to a modest absolute gain because the baseline is so low — a useful reminder that relative measures flatter interventions applied to rare outcomes. The instrumental-variable analysis, using random assignment to estimate the effect of actually opening the message, is a genuine strength: it separates the effect of the intervention from the self-selection of patients who engage with messages anyway.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for ophthalmology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free