All use cases

Engage / Lifecycle

Hormone-symptom titration check-in

A few weeks into hormone therapy, before the dose and route are dialed in, a patient is quietly wondering whether it's even working — and there's no one checking in to find out.

Close the loopGet aheadTime-based triggerBehavioral triggerIn-appPushChatGLP-1 & PeptidesHealth & TelehealthSubscription

What it does

The first weeks on a new hormone prescription are a trial-and-error window: the clinician is still finding the right dose and the right route — gel, injection, patch, or pellet — and the patient is judging the whole thing on how they feel day to day. Hot flashes, mood, energy, and libido move slowly and unevenly, so a patient who expected fast relief can decide it isn't working and drift off before the dial-in is done. The agent runs scheduled symptom check-ins through that window, logs the trend rather than a single bad day, and when symptoms lag the expected timeline or side-effects spike, surfaces it to the clinician for a dose-or-route review. It never adjusts a dose or switches a route — it captures the picture the clinician needs to dial it in.

How it works

  1. 1
    Trigger. a patient is in the early titration window of a new hormone prescription (the dial-in weeks), or reports a symptom change between scheduled check-ins.
  2. 2
    Decision. the agent runs the scheduled symptom check-in (hot flashes, mood, energy, libido), compares the trend against the expected response timeline and the patient's own baseline, and weighs whether this is normal dial-in variation or a lagging response or side-effect spike that warrants clinician review.
  3. 3
    Action. it logs the trend and reassures within the expected timeline where things are tracking; where symptoms lag or side-effects spike, it surfaces the trend to the clinician for a dose-or-route review with the history pre-summarized.
  4. 4
    Follow-through. the clinician's dose-or-route decision is what reaches the patient; the trend writes back to memory so the next check-in reads against the full arc; a symptom that suggests a safety concern rather than slow response is escalated rather than smoothed over.
Example
AgentYou're three weeks in — how have the hot flashes and your energy been this week compared to when you started?
CustomerEnergy's a bit better but the hot flashes are barely budging. I'm starting to think this isn't doing anything.
AgentThanks — that's useful, and three weeks is still early for the full effect. I've logged the trend and flagged it for your clinician to review whether your dose or route needs a tweak. They'll be in touch; nothing changes until they decide.

Configuration

How the agent is wired for this use case.

Triggera scheduled early-titration check-in milestone from the scheduling system, or a patient-reported symptom change during the dial-in window.
Tools & actions
  • Messaging channel · run the symptom check-in (hot flashes, mood, energy, libido) and deliver only the clinician's confirmed decision back.
  • Clinical/EHR system · read current dose, route, and start date to read the trend against the expected timeline.
  • Knowledge base · apply the expected hormone-response timeline and the normal-variation-versus-review thresholds.
  • Clinical queue · surface a lagging-response or side-effect-spike trend for a clinician dose-or-route review with the history pre-summarized.
  • Memory store · write the symptom trend back so each check-in reads against the full arc.
Autonomythe check-ins, trend logging, and timeline comparison run unattended, judge-gated on every patient-facing message; under a clinical-safety policy the agent never adjusts a dose or changes a route — a dose-or-route change requires explicit clinician sign-off, and anything uncertain is posed as a question.
Channelsin-app · push · chat
Escalationa lagging response or side-effect spike is surfaced for a clinician dose-or-route review; a symptom that suggests a safety concern is escalated immediately; every dose-or-route decision is the clinician's.

What you need

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

Signals

scheduled early-titration check-in milestones, patient-reported symptom changes, side-effect reports.

Data

current dose and route and start date, the symptom trend log and the patient's baseline, the expected hormone-response timeline, persistent patient memory, consent state.

Guardrails

clinical-safety policy class — the agent tracks and surfaces but never changes a dose or route; explicit clinician sign-off on any change; uncertain cases posed as questions, not assertions; judge gating on every patient-facing message; safety-concern symptoms escalated immediately; full audit trail of every flagged trend and decision.

Metrics it moves

  • adherenceup, by keeping patients engaged through the dial-in weeks instead of drifting off before the dose and route are right.
  • save-rateup, because a lagging response is caught and routed to a clinician review rather than ending in a silent quit.
  • time-to-resolutiondown, as the clinician gets a pre-summarized symptom trend to act on instead of reconstructing the dial-in from scratch.

Related use cases

Prevent / Care

Titration-readiness clinician handoff

the GLP-1 step-up readiness handoff this parallels for hormone dial-in

Retain / Save

Early-discontinuation save for new hormone starters

the save that catches the same early-weeks drop-off from the churn side

Retain / Save

Adherence guardian for medication programs

the general adherence-from-signals pattern this specializes for hormone titration

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