Why Wearable Data Still Isn’t Used in Clinical Care (And What Needs to Change)

Jadumani Singh
11 hours ago
3 min read

Wearable devices are everywhere, from smartwatches to rings and fitness trackers. They continuously capture health data like heart rate, activity, sleep, and even ECG or oxygen saturation. But despite this explosion in data, very little of it is actually used in clinical care.

At JR Analytics, we analysed leading wearable platforms and their integration pathways within the Australian healthcare system.

Here’s what we found.

The Core Problem

Wearable technology has advanced rapidly, but healthcare systems have not kept pace.

There is no direct integration between wearable devices and clinical systems
Data sits within consumer ecosystems, not healthcare environments
Clinical systems are not designed to ingest continuous patient-generated data

👉 The result: Clinicians rarely see or use wearable data in real-world decision-making.

Why Integration Fails

Most wearable data never reaches clinical systems, and the reason is structural.

Platforms like Apple HealthKit and Google Health Connect act as data gateways, not clinical systems. While they store and organise data, they do not connect directly to hospital EMRs or My Health Record. Instead, integration requires additional layers, often complex, fragmented, and inconsistent.

How Wearable Data Actually Reaches Clinical Systems

In reality, wearable integration follows a multi-step pipeline:

Wearable Device: Sensors capture physiological data (heart rate, activity, ECG, SpO₂)
Smartphone Application: Data is transferred and stored in apps (Apple Health, Samsung Health, Garmin, etc.)
Middleware Layer: Data is aggregated, standardised (e.g. HL7 FHIR) and validated
Only then can data be reviewed and potentially used by clinicians

What Wearable Data Is Actually Good For

Not all wearable data is clinically useful, and context matters.

Trend Monitoring:

Heart rate (HR), Respiratory rate (RR) and activity patterns
These are relatively reliable and useful for longitudinal monitoring.

Acute Decision-Making:

SpO₂ (variable accuracy)
Temperature (peripheral, not core)
ECG (single-lead, user-triggered)
These metrics are variable, context-dependent and not suitable for real-time clinical decisions

Wearable data is most valuable when used for trends and patterns, and not isolated readings

The Australian Healthcare Context

Integration challenges are even more pronounced in Australia:

Fragmented EMR systems
Limited interoperability
My Health Record is document-based, not designed for continuous data

Important limitation:

Raw wearable data cannot be uploaded directly
Only summarised, clinically relevant information can be shared

This reinforces a key requirement that data must be transformed before it becomes clinically useful

The Real Barrier: It’s Not the Devices

The issue is not the wearable devices themselves. The real challenge is data volume, data variability, lack of clinical workflows, and governance and responsibility. Wearable data is patient-generated, non-clinical grade and variable in quality. Without proper validation and oversight, it cannot be safely used in healthcare settings.

What Needs to Change

If wearable data is to become clinically useful, the approach needs to shift.

1. Move from Raw Data → Clinical Summaries

Clinicians do not need continuous streams; they need trends, alerts and summarised insights

2. Use Middleware as the Bridge-Middleware platforms are currently the most practical solution:

Enable multi-device integration
Convert data into FHIR standards
Support scalable deployment

3. Keep Clinicians in the Loop: Wearable data must be reviewed, validated and clearly labelled as patient-generated

4. Start Small (Pilot First)

Controlled environments
Defined workflows
Measurable outcomes

What This Means for Digital Health

The opportunity is not in building more wearable devices. The opportunity is in making wearable data clinically usable. This means structuring data, filtering noise, and delivering meaningful summaries