An AI medication adherence agent.
The OneDose Adherence Agent tracks whether a patient is actually taking their medication and intervenes when they are not. It reminds patients across SMS, voice and app on a schedule built from their own prescription, detects drift when doses are missed or refills slip, and escalates to a pharmacist when the pattern suggests the patient has stopped.
Adherence is a behaviour, not an event, so the agent is measured on the pattern across weeks rather than on whether a single reminder was delivered.
How does the Adherence Agent actually work?
What follows is the mechanism — what the agent does, in what order, and where a human stays in the loop.
The schedule comes from the prescription
Reminder timing is derived from the actual dosing regimen — including what to take with food, what not to take together, and which doses matter most if one has to be missed. A generic 9am ping for every patient is the version of this that stops working in week two.
It notices drift, not just a missed dose
One missed dose is noise. A pattern of missed evening doses, a refill collected four days late, and a patient who stopped answering is a signal. The agent is watching for the second thing.
It asks why
When adherence slips, the agent asks the patient what happened — side-effects, cost, confusion about the regimen, or the medication simply having run out. Those are four different problems with four different fixes, and a reminder solves none of them.
It escalates before the patient is lost
A patient reporting a side-effect they intend to solve by stopping is a pharmacist conversation, and the agent routes it as one — with the adherence history and the reported reason attached, so the pharmacist starts the call already knowing why.
What will the Adherence Agent not do?
The limits matter more than the capabilities when the subject is medication, so they are stated plainly rather than left to be discovered in a pilot.
- It does not change a regimen. Any suggestion that a dose, a drug or a schedule should change goes to a pharmacist or prescriber.
- It does not counsel on a side-effect. It captures the report, routes it to a human, and tells the patient a human is coming.
- It reports what the patient tells it. Self-reported adherence is self-reported, and where a deployment has dispensing or refill data the agent uses that too — but it does not claim to observe someone swallowing a tablet.
How does this agent fit with the others?
The Adherence Agent is one of six OneDose agents that share one patient context and one audit log. That shared context is the difference between a workforce and six tools: the voice agent already knows what the follow-up agent asked yesterday, so the patient is not asked to repeat themselves and the care team reads one history rather than six.
Frequently asked
- What is an AI medication adherence agent?
- An AI medication adherence agent is software that tracks whether a patient is taking their medication as prescribed and intervenes when they are not — sending reminders built from the specific regimen, detecting patterns of missed doses or late refills, asking the patient what went wrong, and escalating to a pharmacist when the patient appears to be stopping.
- How is this different from a reminder app?
- A reminder app fires a notification and has no idea what happened next. The OneDose Adherence Agent tracks the pattern over weeks, asks the patient why a dose was missed, distinguishes a side-effect from a cost problem from a refill that never got collected, and routes the ones that need a pharmacist to a pharmacist.
- How does OneDose know whether a patient actually took the dose?
- Primarily by asking, and by watching refill and dispensing behaviour where a deployment has access to it. Self-reported adherence has known limits, and OneDose treats it as a signal rather than as proof — which is why the escalation triggers on a pattern rather than on a single answer.
- What happens if a patient stops responding entirely?
- Non-response is itself a signal. A patient who stops answering after reporting a side-effect is escalated, because the most likely explanation is that they have stopped taking the medication rather than that they are busy.