Medication Reminders and Adherence: What the Research Actually Shows

This page used to say that phone call reminders improve medication adherence more than texts, apps or pill organisers, under a heading promising evidence. There was no evidence, and the claim does not hold. Here is what the research actually reports, including the parts that are unflattering to a company selling reminder calls.

Do reminder calls beat texts for medication adherence?

There is no good evidence that they do — and the intervention with the strongest support is text messaging, not calling. A meta-analysis in JAMA Internal Medicine (Thakkar et al., 2016) pooled 16 studies covering 2,742 patients with chronic conditions and found text messaging roughly doubled the odds of adherence: OR 2.11 (95% CI 1.52–2.93), lifting adherence from 50% to 67.8%.

That is a real effect, and it belongs to SMS. No comparable body of work shows voice calls outperforming it. If someone quotes you a figure for calls beating texts, ask which study — the honest answer is usually that there isn't one.

The authors are careful about their own result, and so should anyone citing it be: the trials were short, and most measured adherence by asking people, which is not the same as knowing.

How much does any reminder actually help?

Less than the industry implies. The largest systematic review of the field — Cochrane, Nieuwlaat et al., 2014 — assessed 182 randomised controlled trials. Its conclusion is sobering: "even the most effective interventions did not lead to large improvements" in adherence or clinical outcomes.

Of the 17 trials at lowest risk of bias, "only five of these RCTs improved both medicine adherence and clinical outcomes, and no common characteristics for their success could be identified." The reviewers end by saying "it is uncertain how medicine adherence can consistently be improved so that the full health benefits of medicines can be realized."

Read that last line properly. After 182 trials, the field cannot reliably say what works. Any product claiming to have solved adherence is ahead of the evidence by a distance.

Why doesn't a reminder fix it?

Because a large share of non-adherence is not forgetting. It is deciding — quietly, and usually without telling anyone.

People stop taking medicines because of side effects they did not want to make a fuss about, because the cost is uncomfortable, because they feel fine and doubt they still need it, because the instructions were never clear, or because they believe something about the medicine that nobody has corrected. A reminder answers none of that. It arrives, is acknowledged, and changes nothing, because the tablet was not missed by accident.

This is the honest boundary of the product. A reminder addresses unintentional non-adherence — forgetting, a disrupted routine, a complicated schedule. It does nothing for the intentional kind, and treating a deliberate decision as a memory problem is how people end up feeling nagged rather than helped.

If someone has stopped taking something on purpose, the useful next step is a conversation with their prescriber or pharmacist about why, not a louder alarm.

When is a reminder the right tool, then?

In the narrow cases where the only obstacle really is memory:

• A new prescription, before it has attached itself to anything in the day. This is where routines are most fragile and a prompt does the most work.
• A schedule with an awkward shape — three times a day, with meals, or a course that ends on a specific date rather than running indefinitely.
• Someone without a smartphone, where the SMS evidence above simply does not apply because texts are not part of their day. This is the one case where a call has a clear mechanical advantage, and it is why the format exists.
• A change of routine — travel, a new job, a hospital discharge — where an established habit has just lost its cue.

Note what is not on that list: chronic long-term adherence in someone who has decided the medicine is not worth taking. The research does not support reminders fixing that, and neither will we.

Setting it up

The practical part is genuinely quick. With ReminderIt connected to Claude, describe the schedule in plain language — "I need my morning tablet at 8am and my evening one at 8pm, Europe/Paris time" — and it creates the reminders and reads the times back before saving. The read-back is the part that matters: an AM/PM slip is the most common setup error, and it is the only moment you will catch it.

For a fixed course rather than an ongoing prescription, say so — "three times a day with meals for seven days, then stop" — so the reminders end when the course does instead of ringing for months afterwards.

Times, doses and whether a medicine should be taken with food are your prescriber's call, not ours and not an AI's. Set the reminders to match what you were told; if the instructions are unclear, a pharmacist will clarify them for free in about two minutes, and that conversation is worth more than any scheduling feature.

Sources: Thakkar J, et al. Mobile Telephone Text Messaging for Medication Adherence in Chronic Disease: A Meta-analysis, JAMA Internal Medicine (2016). Nieuwlaat R, et al. Interventions for enhancing medication adherence, Cochrane Database of Systematic Reviews (2014), CD000011.