Ethical Care | CareJournal Hub

Executive Summary

CareJournal Hub is built on four foundational principles of ethical caregiving technology: transparency, user control, data security, and human oversight. We believe AI can empower caregivers—not replace them or their judgment.

This document explains how we implement these principles across our platform, backed by HIPAA-aligned security standards, clear data practices, and user-centered design.

Our Four Pillars

1. Transparency & Explainability

Every AI-generated insight includes a clear explanation of why it was generated. You can click "💡" to understand the reasoning. We never hide the AI's logic from you.

2. User Control & Autonomy

You decide whether to use AI features. Toggle AI on/off anytime in Settings. Export your data or delete your account entirely. The platform is designed around your choices.

3. Data Privacy & Security

Your health information is protected with bank-level encryption, HIPAA-aligned access controls, and strict authentication. Your data is never shared without explicit permission.

4. Human Oversight

AI never replaces your doctor's judgment. Our insights are suggestions to discuss with healthcare providers, not medical directives. You remain in control of every care decision.

For Families & Caregivers

Caregiving is overwhelming. You're tracking medications, appointments, mood changes, and family coordination all at once. CareJournal Hub is built to reduce that cognitive load while keeping you in charge.

How AI Works in CareJournal Hub

When you write journal entries about your loved one's care, our AI reads patterns in:

Recent mood trends
Medication adherence
Appointment preparation
Care coordination challenges

Then it generates suggestions like "Consider asking the doctor about the recent mood dips we've noticed" or "You're doing great with medication adherence this week."

Proof Point: Transparency in Action

Write a journal entry about your loved one's week (mood, activities, concerns)
Open the Insights section → You'll see AI suggestions appear
Click the "💡" button → A dialog shows exactly why this insight was generated
Example: "This insight was generated because your journal entries mention 'anxious' or 'frustrated' in 3 of the last 5 days. We also noted you have an appointment with Dr. Singh next week—a good time to discuss this pattern."
Share with doctor → Copy the insights and bring them to the appointment
Report if wrong → See a suggestion that doesn't fit? Click "🚩" and tell us why

You Control Your Data

Opt Into or Out of AI

In Settings → Privacy & AI, you can toggle AI features on or off. If you disable AI:

No insights are generated
Your journal entries are still saved (for your eyes only)
You're not sharing data with OpenAI

Export Your Data

In Settings → Data Management, you can download:

PDF: A formatted summary of medical records (patients, medications, appointments)
JSON: Your complete data for backup or transfer to another service

Delete Your Account

If you decide CareJournal Hub isn't right for you, you can permanently delete your account and all associated data. This removes:

All patients and care records
Medication logs and appointments
Journal entries
All other personal data

How We Handle Your Health Information

Passwords: Encrypted with industry-standard bcrypt (12 rounds of hashing)

Sessions: Secure HTTP-only cookies that expire after 7 days

Access: Only you can see your data (we never sell or share it)

Ownership: You own everything. We're just the tool.

The Feedback Loop

We're building ethical AI together with you. When you:

Click "✕" (inaccurate)
Click "🚩" (report a problem)
Add comments to feedback

...you're helping us improve. Your feedback is reviewed by our team to refine how AI works.

For Healthcare Providers & GCM Partners

CareJournal Hub is a clinical care coordination platform designed to support—not replace—professional judgment. This section details our security practices, data handling, and ethical safeguards.

Compliance & Security Standards

HIPAA-Aligned Architecture

CareJournal Hub is built to HIPAA standards with the following controls:

Control	Implementation
Authentication	Email/password with 8+ character minimum + bcrypt hashing (12 salt rounds); optional Google OAuth
Session Management	HTTP-only cookies with 7-day TTL; secure flag in production; SameSite protection
Authorization	Role-based access (users can only view their own patients/data); middleware verification on all protected endpoints
Data Integrity	PostgreSQL database with transactional integrity; audit logs for sensitive operations
Transport Security	HTTPS-only in production; strict CORS policies

Formal Audit Status: Built to HIPAA standards; pursuing formal compliance audit and certification (roadmap: Q2 2026).

Data Architecture & Minimization

What Data We Collect

For caregivers: Name, email, password (hashed), role, referral information

For patients/clients: Name, date of birth, relationship, medical conditions, medications, appointments, journal entries, doctor visit summaries

What Data We Send to OpenAI for AI Insights

To generate care suggestions, we send:

Last 5 recent mood entries (date + mood only; full content only for 3 most concerning if mood is negative)
Medication list (name, dosage, frequency, purpose only; NO patient identifiers)
Recent appointment titles (for context; NO doctor's full notes)
Patient name only (for personalization; NO medical history)

We explicitly do NOT send to OpenAI: Patient ID, full journal entries, detailed medical history, family member information, financial data

Access Control & Ownership Verification

Every protected endpoint requires:

User authentication (via isAuthenticated middleware)
Patient ownership verification (users can only access their own patients)
Role-based authorization (GCMs see their clients; family members see shared records)

User Rights & Data Portability

Right to Access

Users can view all their data via the platform's standard interface and also through full export capability.

Right to Export

Implemented via /api/users/export/json and /api/users/export/pdf:

JSON export includes all structured data
PDF export provides formatted medical records summary
Data is exported in standard formats for portability

Right to Delete

Implemented via /api/users/delete-account:

Cascading deletion removes user and all related records
Deletion is permanent and cannot be undone
No data retention after deletion

Compliance Roadmap

Milestone	Target Date	Description
Formal security audit	Q2 2026	Third-party penetration testing & audit
BAA templates	Q2 2026	Business Associate Agreements for healthcare partners
SOC 2 Type II certification	Q3 2026	Service Organization Control compliance
HIPAA certification	Q4 2026	Formal HIPAA compliance certification

For GCM Partners Specifically

Client Management & Visibility

GCMs can view all clients they've added to their roster
Read-only access to client medication adherence and appointment data
Ability to generate reports on client engagement (in development)
Protected by the same access controls as individual users

Family Coordination

GCMs can see which family members are invited to client accounts
View-only access to shared insights and journal summaries
Ability to support families with platform training

Data Segregation

Each GCM's client data is isolated
No GCM can see another GCM's clients
Multi-tenant isolation at the database level

Questions?

For families & caregivers: support@carejournalhub.com

For healthcare providers & GCM partners: partners@carejournalhub.com

For privacy & security questions: privacy@carejournalhub.com

References

This document is informed by:

Ethical Care Tech: How AI Should (and Shouldn't) Be Used in Caregiving (Industry framework)
HIPAA Privacy & Security Rules: 45 CFR §§ 160, 164
HL7 Healthcare Data Governance Framework
AI Ethics Principles: Transparency, Fairness, Accountability, Human Oversight
Industry Best Practices: Healthcare security standards, caregiver-centered design

Ethical Care at CareJournal Hub