Post-Surgical Remote Monitoring Voice UI Dialogue Script
Draft a patient-centric, conversational Voice User Interface (VUI) script for post-operative remote symptom check-ins and automated triage escalation.
Use this template when designing voice-first conversational flows for post-discharge patient care. It structures branching dialogues, error handling, and clinical escalation triggers suited to recovering patients.
Role: Senior Conversational UX Designer specializing in clinical voice user interfaces and patient communication safety.
Context
- Surgical procedure type: {{surgical_procedure_type}}
- Patient demographic profile: {{patient_demographic}}
- Core vital / symptom metric: {{primary_symptom_metric}}
- Clinical escalation threshold: {{escalation_threshold}}
- Delivery voice channel: {{vui_platform_channel}}
- Accessibility / cognitive constraint: {{accessibility_requirement}}
Task
Design a multi-turn, accessible Voice User Interface (VUI) dialogue script that conducts daily symptom triage for patients recovering from surgery, maintaining patient reassurance while detecting red-flag clinical anomalies.
Method
- Define voice persona attributes (tone, pacing, phonetic clarity) adapted for {{patient_demographic}} and {{accessibility_requirement}}.
- Write an empathetic, low-cognitive-load opening exchange that authenticates the patient securely over {{vui_platform_channel}}.
- Script primary branching logic to measure and categorize {{primary_symptom_metric}} using standardized conversational grading scales.
- Build graceful error-handling loops for silence, ambiguous responses, and multi-intent patient inputs.
- Craft immediate emergency branch scripts triggered when patient self-reporting crosses {{escalation_threshold}}.
- Formulate non-alarmist closing instructions for stable patients with clear timeline markers for the next automated interaction.
- Annotate all conversational turns with speech-synthesis markup cues (SSML pauses, pitch, emphasis).
Constraints
- MUST keep individual voice prompts under 25 words to prevent cognitive fatigue.
- MUST NOT provide medical diagnostic advice; focus solely on data collection and care-team triage.
- System must include a deterministic fallback to live clinical staff at any direct escalation trigger.
- Language must be free of clinical jargon, adhering strictly to plain-language health literacy guidelines.
Output format
- VUI Persona & SSML Delivery Specs (100 words)
- Main Happy Path Dialogue Script (Speaker labels, Spoken lines, SSML tags, Intent rules)
- Edge Path 1: High-Severity Escalation Flow (Spoken lines, Care team handoff protocol)
- Edge Path 2: Ambiguous & No-Input Repair Loops (3-tier reprompt ladder)
- Concluding Confirmation & Wrap-Up Script (Spoken dialogue)
Self-review
- Check that no prompt exceeds the strict 25-word cognitive ceiling.
- Verify escalation thresholds trigger clear human-in-the-loop protocols.
- Ensure SSML tags explicitly indicate pacing suitable for post-operative recovery states.
Explicit role, a named task, and discrete steps the model can follow.
Background, inputs and variables the model needs before it starts.
Hard boundaries — what the model must and must not do.
A named, field-level shape for the response.
Ordered work items that force analysis before an answer.
Length and structure that travel across frontier models.
Signal density — instruction weight without padding.
Documented variables so the scaffold adapts to new inputs.
Quality bar, assumptions and behaviour when inputs are thin.
How much real usage the template has behind it.