Design: Pragmatic, two-stage sequential multiple-assignment randomised trial
Population: 20,604 US adults with ASCVD or high risk and recent statin non-adherence
Primary outcome: Statin refill within 14 days of each randomisation
Effect: 14.4% vs 12.0%; adjusted risk difference 2.3 points (95% CI 1.3 to 3.4)

ADHERE-ASCVD was a pragmatic, health-system-embedded randomised trial at a large US integrated provider. It enrolled 20,604 adults with atherosclerotic disease or high risk who had recently lapsed on statin refills. They were randomised to portal message, text message, email or usual communication; those who did not refill within 14 days were randomised again to repeat, switch channel or stop.

Any digital message raised 14-day refills from 12.0% to 14.4% (adjusted risk difference 2.3 points, 95% CI 1.3 to 3.4). A second message to non-responders added a further 2.5 points. Switching to a different channel did no better than repeating the same one. By 28 days, cumulative refill was 24.9% versus 21.2%.

The absolute effect is small and the outcome is a refill, not an event. But the intervention is nearly free, scales to a whole practice list, and the finding that a repeat reminder works is directly usable. Most people still did not refill, so outreach complements a conversation rather than replacing it.

Send a reminder to patients who have lapsed on statin refills; a second one is worth sending to non-responders.
Use whichever channel the patient already reads; switching channel added nothing.
Ask about side effects and cost when a patient stops, since three-quarters still did not refill.
Build refill-gap alerts into clinic software where possible.

Why it matters

Lapsed adherence is one of the cheapest risk factors to address, and a repeated message is about the cheapest tool there is.

The statistics, in plain English

A 2.3-point absolute rise means about one extra refill for every 43 people messaged. The relative increase (risk ratio 1.20) sounds larger than it is because baseline refill was low. Refill is a surrogate for taking the drug, and the trial did not measure heart attacks or strokes.