We Keep Trying to "Fix" Medication Adherence With More Text Messages and Apps
A major randomized pragmatic trial suggests that even when texts are thoughtfully designed, texting isn't the lever we want it to be. Here's what the evidence shows — and what to do instead.
Comments on "Personalized Patient Data and Behavioral Nudges to Improve Adherence to Chronic Cardiovascular Medications: A Randomized Pragmatic Trial" — January 7, 2025
We keep trying to "fix" medication adherence with more messages. But this major randomized pragmatic trial suggests that — even when texts are thoughtfully designed — texting isn't the lever we want it to be.
What the Study Found (in Plain English)
Researchers enrolled 9,501 adults across three U.S. health systems who were taking chronic cardiovascular medications and showed early signs of nonadherence (a refill gap). Patients were randomized to one of three texting approaches or usual care:
- Generic refill reminder texts
- "Behavioral nudge" refill reminder texts
- Behavioral nudge texts plus a fixed-message chatbot
- Usual care
The primary outcome was refill adherence at 12 months using "proportion of days covered" (PDC) from pharmacy data, with secondary outcomes like ED visits, hospitalizations, and mortality.
Study Result
Average PDC at 12 months was roughly 60–63% across all groups — and there were no differences in clinical events.
The paper's conclusion is blunt: these text reminders did not improve adherence or outcomes at 12 months.
Why Texting Often Fails: It Asks Patients to Do Extra Work
The study is a good reminder of a simple behavioral truth: adherence fails in the "last mile" of everyday life — when people are busy, stressed, traveling, juggling kids, or simply exhausted.
Texting interventions often break down because they add steps:
- Notice the text
- Open it
- Decide what to do
- Act (refill, call, log in, reply, navigate a chatbot)
- Repeat… indefinitely
That's not support — it's a recurring nuisance. Over time, texts become background noise, feel like nagging, and create "behavioral sludge" (friction that pushes people toward doing nothing). The more an intervention depends on extra attention, extra clicks, or extra willpower, the more it selects for the people who were already most likely to adhere.
RedCap Takes a Different Approach: Nudge the Environment, Not the Patient
Nobel prize winner and behavioral economist Richard Thaler (with Cass Sunstein) popularized the idea that the best behavior change often comes from improving choice architecture — designing the environment so the right action becomes the easy, obvious default. They define a "nudge" as changing choice architecture in a predictable way without forbidding options, and importantly: it should be easy and cheap to avoid.
That's exactly where RedCap fits.
Texting is "Prompting"
- ✗ Reply to something
- ✗ Click something
- ✗ Log something
- ✗ Remember something new
RedCap is "Choice Architecture"
- ✓ "Did I take it?" becomes visually obvious
- ✓ "Which day am I on?" disappears
- ✓ Routine becomes automatic
- ✓ Less dependent on motivation
"Adherence isn't primarily an information problem. It's a friction problem. When an intervention requires patients to do extra work, it competes with real life — and real life usually wins."
The Takeaway
The trial didn't show that patients don't care. It showed something more useful: adherence isn't primarily an information problem. It's a friction problem.
RedCap works because it does the opposite of texting: it removes steps, removes ambiguity, and turns adherence into the easiest next move.
Or said differently: if we want medications to work, we can't just message patients more. We have to make taking the medication effortless.