All use cases

Engage / Lifecycle

Scheduled health-questionnaire check-in cadence

A short, friendly check-in arrives on schedule, asks how the treatment is really going, and turns the answers into something the care team can act on.

Get aheadTime-based triggerBehavioral triggerWhatsAppPushIn-appChatGLP-1 & PeptidesHealth & TelehealthSubscription

What it does

Between appointments, a GLP-1 program is mostly blind: it knows the patient was prescribed, but not whether the medication is working, whether side-effects are building, or whether the patient is quietly drifting off. Waiting for the next visit means small problems become cancellations and adherence dips go unseen. The agent runs a recurring check-in on a predefined interval, sends a structured health questionnaire on the channel the patient actually reads, guides them through it conversationally instead of dumping a form, and captures the medical feedback — treatment progress, severe side-effects — that the program needs to gauge adherence and catch issues early. WhatsApp-primary where that is the channel of record, which fits EU programs.

How it works

  1. 1
    Trigger. the patient's check-in interval comes due on the schedule, or an engagement gap since the last completed questionnaire crosses the program's threshold.
  2. 2
    Decision. the agent reads the patient's memory — program stage, last questionnaire, current dose context — picks the right question set for where they are, and a judge gates the outbound message for tone and safety before it sends.
  3. 3
    Action. the questionnaire arrives on WhatsApp, push, or in-app chat as a guided conversation; the agent walks the patient through each item and records the answers as structured patient-reported outcomes.
  4. 4
    Follow-through. answers are written back so adherence and side-effect trends are visible to the care team; a severe-side-effect answer escalates to a clinician with the captured context; if the patient completed a check-in on their own, the scheduled prompt is cancelled so it never double-sends.
Example
AgentQuick monthly check-in — takes about two minutes. First, since we last spoke, how would you say the treatment is going for you: better, about the same, or harder?
CustomerBetter mostly, but the nausea came back strong this week.
AgentThanks for telling me — I've logged that. A couple of follow-ups so your care team has the full picture: is it stopping you eating or drinking, and roughly how many days has it lasted? If it's severe I'll flag a clinician to look today.

Configuration

How the agent is wired for this use case.

Triggera checkin_due event on the patient's predefined interval from the scheduling system, or an engagement gap since the last completed questionnaire from the app backend
Tools & actions
  • Scheduling system · read the check-in interval and last-completed date, fire the due event
  • Patient memory store · read program stage and prior questionnaire answers; write the new answers back
  • Messaging channel · deliver the questionnaire as a guided conversation on WhatsApp, push, in-app, or chat
  • Clinical/EHR system · record structured patient-reported outcomes and adherence signals against the patient record
  • Clinician escalation queue · route a severe-side-effect answer to a clinician with the captured context
Autonomythe scheduled questionnaire and conversational guidance run unattended, judge-gated for tone and safety on every send, under a coaching-not-medical-advice clinical-safety policy; the agent captures and flags answers but never interprets results or advises a dose — any red-flag answer is detected and escalated to a clinician, never resolved in chat
Channelswhatsapp · push · in-app · chat
Escalationa severe or red-flag side-effect answer hands off to a clinician with the questionnaire context attached

What you need

The inputs this use case runs on. Your channels stay yours; the agent supplies the judgment.

Signals

checkin_due on the predefined interval, last-completed-questionnaire timestamp, engagement-gap events

Data

program stage, prior questionnaire answers and memory, channel of record and messaging consent, the structured question set

Guardrails

judge gating on every unprompted send; capture-and-flag only, never result interpretation or dose advice; red-flag answers escalate to a clinician; frequency caps; stale-prompt cancellation once a check-in is completed; regional channel and data-handling rules for the markets in scope

Metrics it moves

  • adherenceup, because a regular structured touch keeps the patient engaged and surfaces drift before it becomes a stop
  • dau-mauup, as the recurring check-in pulls quiet patients back into the conversation
  • churndown, by catching building side-effects and stalled progress while there is still time to act
  • csatup, since the patient feels monitored and supported between visits rather than left alone

Related use cases

Engage / Lifecycle

Commitment check-in

the same check-in muscle for self-set goals, where this card captures clinical patient-reported outcomes

Engage / Lifecycle

Between-doses lull check-in

the lighter weekly-trough touch, versus this card's structured questionnaire

Engage / Lifecycle

Enrollment-assigned continuous GLP-1 coach

the continuous coach the captured answers feed

See it on your own customer journey

Bring one drop-off, one churn cliff, or one silent segment. We will show you what a proactive agent with memory and judgment does with it.

Book a demo