Clinical, Disease & Dying Process Models
Relevant source files
The following files were used as context for generating this wiki page:
- artifacts/api-server/src/intelligence/hospice/catalog.ts
- artifacts/api-server/src/routes/anthropic/systemPrompt.ts
- artifacts/api-server/src/intelligence/hospice/planner.ts
- artifacts/mobile/data/resources.ts
- attached_assets/Pasted-Build-the-next-major-intelligence-architecture-for-Hosp_1774852812580.txt
- artifacts/api-server/src/intelligence/hospice/documentationCoach.ts
The Clinical, Disease, and Dying Process Models form the core intelligence layer of the Hospice Roadmap platform. These models translate complex medical trajectories and physiological changes into actionable guidance for caregivers and clinicians. By categorizing the end-of-life journey into specific disease types and physiological phases, the system provides high-fidelity, role-aware support that bridges the gap between clinical observation and practical caregiving.
This intelligence architecture is designed to support multiple stakeholders, including patients, caregivers, and healthcare professionals, by providing deep insights into symptom progression, expected versus urgent changes, and the specific mechanics of the dying process.
Sources: attached_assets/Pasted-Build-the-next-major-intelligence-architecture-for-Hosp_1774852812580.txt, artifacts/api-server/src/routes/anthropic/systemPrompt.ts:3-15
Intelligence Architecture & Data Flow
The system operates as a layered intelligence engine rather than a static knowledge base. It utilizes a deterministic retrieval system to map user input and patient context to specific clinical models and knowledge blocks.
The architecture ensures that every AI response is grounded in approved clinical data before being processed by the language model.
Sources: artifacts/api-server/src/intelligence/hospice/planner.ts:14-41, attached_assets/Pasted-Build-the-next-major-intelligence-architecture-for-Hosp_1774852812580.txt
Disease Trajectory Models
The platform implements dedicated models for major terminal illnesses, allowing the system to anticipate specific complications and define "good care" for each condition.
Major Disease Categories
| Disease Category | Key Clinical Indicators | Primary Focus |
|---|---|---|
| Advanced Cancer | Cachexia-anorexia, pathological fractures, hypercalcemia | Rapid decline management (final 4-8 weeks) |
| Heart Failure (CHF) | Dyspnea at rest, edema, ICD shocks | Roller coaster trajectory, sudden death risk |
| COPD / Lung Disease | Air hunger, CO2 retention (hypercapnia) | Breathlessness and airway management |
| Dementia | FAST Stage 7, dysphagia, aspiration pneumonia | Long plateau followed by steep terminal decline |
| ALS / MND | Respiratory failure, bulbar vs limb onset | Communication and respiratory support |
| Renal/Liver Failure | Uremia, Ascites, Hepatic Encephalopathy | Toxin buildup and fluid management |
Sources: artifacts/api-server/src/routes/anthropic/systemPrompt.ts:74-152, artifacts/mobile/data/resources.ts:1330-1410
Palliative Performance Scale (PPS)
The system uses the PPS to estimate prognosis and drive the urgency of the response plan.
- PPS 70-100%: Normal to reduced activity; focus on routine education.
- PPS 40-50%: Mainly bed-bound; requires considerable assistance.
- PPS 10-30%: Totally bed-bound; prognosis of days to hours.
Sources: artifacts/api-server/src/routes/anthropic/systemPrompt.ts:354-370
The Dying Process Model
The dying process is modeled as a sequence of physiological transitions, each requiring different support strategies and caregiver interventions.
Stages of Natural Decline
Clinical Signs and Interventions
- Weeks to Days: Characterized by "Natural Withdrawal." The focus is on presence and oral care rather than forcing nutrition.
- Active Dying (1-3 Days): Characterized by circulatory shutdown and "Terminal Secretions" (death rattle). Interventions include repositioning and anticholinergic medications.
- Final Moments (Hours): Characterized by extended apnea (45+ seconds) and cooling extremities. The model emphasizes that "Hearing is the last sense to go."
Sources: artifacts/api-server/src/intelligence/hospice/catalog.ts:48-110, artifacts/api-server/src/routes/anthropic/systemPrompt.ts:193-205
Symptom Management Model
The symptom model provides a structured approach to identifying, tracking, and treating common hospice symptoms.
Key Symptom Protocols
| Symptom | Identification / Cues | Interventions (Non-Pharmacological) | Pharmacological |
|---|---|---|---|
| Pain | Grimacing, moaning, PAINAD scale | Positioning, calm environment | Morphine, Hydromorphone, Fentanyl |
| Dyspnea | Labored breathing, cyanosis | Fan to the face, cool air, upright posture | Low-dose Opioids |
| Agitation | Picking at sheets, calling out | Reduced stimulation, familiar voices | Haloperidol, Midazolam |
| Secretions | Gurgling, rattling breath | Side-lying positioning | Atropine, Glycopyrrolate |
Sources: artifacts/api-server/src/intelligence/hospice/catalog.ts:117-210, artifacts/api-server/src/routes/anthropic/systemPrompt.ts:154-191
Time of Death and After-Death Model
This model provides critical guidance for the moments following physiological death, prioritizing emotional steadiness and practical logistics.
Immediate Post-Death Sequence
- Pronouncement: The hospice nurse (not 911) must be called to pronounce death.
- Wait Period: Families are encouraged to take 1-3 hours with the body; there is no rush to move the loved one.
- Medication Disposal: Controlled substances (Opioids) must be secured and disposed of per DEA guidelines, often supervised by the hospice nurse.
- Funeral Coordination: Contact the funeral home only after the hospice nurse has completed documentation.
Sources: artifacts/api-server/src/intelligence/hospice/catalog.ts:285-325, artifacts/api-server/src/intelligence/hospice/documentationCoach.ts:68-80
Clinical Documentation & Tracking
The architecture includes a "Documentation Coach" to ensure care quality and accountability. It provides templates for tracking symptoms, call logs, and care failures.
Summary
The Clinical, Disease & Dying Process Models provide a comprehensive framework for navigating end-of-life care. By integrating disease-specific trajectories, a nuanced dying process model, and rigorous symptom management protocols, the platform ensures that users receive expert-level guidance. This system prioritizes patient comfort and caregiver support, ensuring that the "living knowledge" of each family’s journey is preserved and used to drive personalized, compassionate responses.