Discussing Wearable Data With a Clinician

Urgent first

If you have chest pain, severe breathlessness, fainting or signs of stroke, do not book a routine visit to show your data. In the US call 911 (or your local emergency number). See Symptoms That Need Assessment, Not Tracking. In the US, 988 is the Suicide & Crisis Lifeline. [1]

What "patient-generated health data" means

Clinicians call numbers you collect yourself, from apps, wearables or home devices, patient-generated health data. Studies of how it is used in care are mostly small and exploratory. Evidence label: insufficient evidence that bringing consumer data to visits improves outcomes.

What studies of clinicians and patients found

1. Mental health clinicians, 2020

   Patient-generated data were already part of the workflow but not well integrated. Clinicians supported using them but worried about reliability and accuracy. The authors said more research is needed on outcomes. [2]

   Who: Interviews with 12 psychiatrists and clinical psychologists at one academic hospital

2. Older adults and clinicians, 2021

   Themes included reliability of the data, trust between patient and provider, and privacy. Step count, sleep, blood glucose, blood pressure and stress level were rated most useful. The authors called for more research. [3]

   Who: Focus groups of 9 people (4 clinicians, 5 older adults) in Ontario, Canada

3. US Veterans Affairs pilot, 2022

   Support for use, with setup challenges. Clinicians expected to use the data to monitor progress toward behavior goals. Fitting the data into health records and clinic workflow still needs work. [4]

   Who: Interviews with 26 veterans and 16 clinicians about distributed Fitbits

These are opinion and experience studies with very small samples. They show attitudes and practical barriers, not that sharing data improves health. They also do not tell you what any one clinician will do with your numbers.

Practical points for the conversation

• Lead with the symptom or question. Say what you feel, since when, and what worries you. MedlinePlus advises writing down symptoms, including when they started and what makes them better or worse. [5]
• Say where the data come from. Name the device, what it directly senses and what it estimates. Scores such as "readiness" or "stress" are proprietary calculations. See Algorithm-Derived Scores Explained.
• Expect limits. Reliability and fitting data into clinic workflow and health records were concerns in the studies above, so a clinician may not be able to work with every app's format. [2][3]
• Be honest about what you do. MedlinePlus says to be honest during your appointment so the provider knows what is really going on. [5]
• Ask, do not self-adjust. Questions about medicines belong with your prescriber. Do not stop or change a medicine because of a device reading. See Talking With a Health Professional.
• Think about privacy. Privacy rules depend on who holds the data. See Health Data vs Consumer-App Data.

What a clinician may and may not do with it

A clinician may use a pattern as a prompt for questions, or may not use it at all. The NHS notes, for example, that palpitations may be investigated with an electrocardiogram (ECG), a standard test using pads on the skin. [6] See Device vs Clinical Reference Validation for why a device and a clinical test can differ.

If the visit feels dismissive

A clinician who does not review your app is not necessarily wrong. Ask what information would help most next time, and bring a short summary instead of raw exports. See Preparing a Useful Data Summary for a Clinician.