NIP-101h Health and Fitness Data Framework: HealthNote SDK, and API

The world of health and fitness data is booming. Users are tracking more aspects of their well-being than ever before, from daily steps and workout intensity to sleep patterns and caloric intake. But for developers looking to build innovative applications on this data, significant hurdles remain: ensuring user privacy, achieving interoperability between different services, and simply managing the complexity of diverse health metrics.

Enter the NIP-101h Health Profile Framework and its companion tools: the HealthNote SDK and the HealthNote API. This ecosystem is designed to empower developers to create next-generation health and fitness applications that are both powerful and privacy-preserving, built on the decentralized and user-centric principles of Nostr.

NIP-101h: A Standardized Language for Health Metrics

At the core of this ecosystem is NIP-101h. It's a Nostr Improvement Proposal that defines a standardized way to represent, store, and share granular health and fitness data. Instead of proprietary data silos, NIP-101h introduces specific Nostr event kinds for individual metrics like weight (kind 1351), height (kind 1352), step count (kind 1359), and many more.

Key features of NIP-101h:

• Granularity: Each piece of health information (e.g., weight, caloric intake) is a distinct Nostr event, allowing for fine-grained control and access.
• User Control: Built on Nostr, the data remains under user control. Users decide what to share, with whom, and on which relays.
• Standardization: Defines common structures for units, timestamps, and metadata, promoting interoperability.
• Extensibility: New metrics can be added as new NIP-101h.X specifications, allowing the framework to evolve.
• Privacy by Design: Encourages the use of NIP-04/NIP-44 for encryption and includes a consent tag for users to specify data-sharing preferences.

You can explore the full NIP-101h specification and its metric directory in the main project repository.

The HealthNote SDK: Simplifying Client-Side Integration

While NIP-101h provides the "what," the HealthNote SDK provides the "how" for client-side applications. This (currently draft) TypeScript SDK aims to make it trivial for developers to:

• Create & Validate NIP-101h Events: Easily construct well-formed Nostr events for any supported health metric, ensuring they conform to the NIP-101h specification.
• Handle Encryption: Seamlessly integrate with NIP-44 to encrypt sensitive health data before publication.
• Manage Consent: Automatically include appropriate consent tags (e.g., defaulting to aggregate-only) to respect user preferences.
• Publish to Relays: Interact with Nostr relays to publish the user's health data.
• Prepare Data for Analytics: Extract minimal, privacy-preserving "stat-blobs" for use with the HealthNote API.

The SDK's goal is to abstract away the low-level details of Nostr event creation and NIP-101h formatting, letting developers focus on their application's unique features.

The HealthNote API: Powerful Insights, Zero Raw Data Exposure

This is where things get really exciting for developers wanting to build data-driven features. The HealthNote API (detailed in HealthNote-API.md) is a server-side component designed to provide powerful analytics over aggregated NIP-101h data without ever accessing or exposing individual users' raw, unencrypted metrics.

Here's how it achieves this:

1. Privacy-Preserving Ingestion: The SDK sends only "stat-blobs" to the API. These blobs contain the numeric value, unit, timestamp, and metric kind, but not the original encrypted content or sensitive user identifiers beyond what's necessary for aggregation.
2. Aggregation at its Core: The API's endpoints are designed to return only aggregated data.
   • GET /trend: Provides time-series data (e.g., average daily step count over the last month).
   • GET /correlate: Computes statistical correlations between two metrics (e.g., does increased activity duration correlate with changes in workout intensity?).
   • GET /distribution: Shows how values for a metric are distributed across the user base.
3. Built-in Privacy Techniques:
   • k-Anonymity: Ensures that each data point in an aggregated response represents at least 'k' (e.g., 5) distinct users, preventing re-identification.
   • Differential Privacy (Optional): Can add statistical noise to query results, further protecting individual data points while preserving overall trends.
4. No Raw Data Access for Developers: Developers querying the API receive only these aggregated, anonymized results, perfect for powering charts, dashboards, and trend analysis in their applications.

A Typical Workflow

1. A user records a workout in their NIP-101h-compatible fitness app.
2. The app uses the HealthNote SDK to create NIP-101h events for metrics like distance, duration, and calories burned. Sensitive data is encrypted.
3. The SDK publishes these events to the user's configured Nostr relays.
4. The SDK also extracts stat-blobs (e.g., { kind: 1363, value: 5, unit: 'km', ... }) and sends them to the HealthNote API for ingestion, tagged with an aggregate-only consent.
5. Later, the app (or an authorized third-party service) queries the HealthNote API: GET /trend?kind=1363&bucket=week&stat=sum.
6. The API returns a JSON object like: {"series": [{"date": "2024-W20", "value": 15000}, ...]} showing the total distance run by all consenting users, week by week. This data can directly populate a trend chart.

Benefits for the Ecosystem

• For Users:
  • Greater control and ownership of their health data.
  • Ability to use a diverse range of interoperable health and fitness apps.
  • Confidence that their data can contribute to insights without sacrificing personal privacy.
• For Developers:
  • Easier to build sophisticated health and fitness applications without becoming privacy experts or building complex data aggregation pipelines.
  • Access to rich, aggregated data for creating compelling user-facing features (trends, benchmarks, correlations).
  • Reduced burden of storing and securing sensitive raw health data for analytical purposes.
  • Opportunity to participate in an open, interoperable ecosystem.

The Road Ahead

The NIP-101h framework, the HealthNote SDK, and the HealthNote API are foundational pieces for a new generation of health and fitness applications. As these tools mature and gain adoption, we envision a vibrant ecosystem where users can seamlessly move their data between services, and developers can innovate rapidly, all while upholding the highest standards of privacy and user control.

We encourage developers to explore the NIP-101h specifications, experiment with the (upcoming) SDK, and review the HealthNote API design. Your feedback and contributions will be invaluable as we build this privacy-first future for health data.

https://github.com/HealthNoteLabs