Key points
- A wai Profile can hold readings from Apple Health, Oura, Fitbit and glucose monitors alongside clinic records and hospital letters, on one timeline.1
- Wrist heart rate is close in the lab, with six of seven devices under 5% median error while cycling, and no device estimated calories burned within 20%.2
- Consumer sleep trackers detect sleep well, with sensitivity of at least 0.93 against a sleep laboratory, and detect wake less well, with specificity of 0.18 to 0.54.3
- Among 419,297 smartwatch users, 0.52% received an irregular pulse notification, and 84% of notifications received while wearing an ECG patch matched atrial fibrillation on it.4
- Treat device numbers as trends to discuss with a clinician, and clinic results and letters as the clinical record.
Summary
A personal health profile can take readings from phones, rings, watches and glucose monitors, alongside clinic records and hospital letters. Wearables are most reliable for trends such as resting heart rate and sleep timing, and approximate for calories burned and sleep stages. Clinic results and letters remain the clinical record, and any reading that worries you is a reason to speak to a clinician.
Your wearable readings and your clinic letters sit on one timeline, ready for you and your clinician to read together. Knowing what each source measures well, and where it is approximate, lets you both use all of it with confidence.
See it in your clinicThirty minutes on your own patients and doctors.
Request a demoConnect the devices you already wear and see their readings beside your clinic results and letters on one timeline.
| Source | What it adds | Measured by | What the research says | Best read as |
|---|---|---|---|---|
| Apple Health | Heart, sleep, activity | Phone and watch sensors | Wrist heart rate within 5% median error while cycling for most devices tested, calories over 20% error | Heart rate trends, activity trends |
| Oura | Sleep, readiness, temperature | Ring sensors | Consumer sleep trackers detect sleep well and wake less well, stages inconsistent | Sleep timing and duration trends |
| Fitbit | Steps, exercise, heart | Wrist sensors | Heart rate close in the lab, error higher when walking and for darker skin tones | Activity and resting heart rate trends |
| Glucose monitors | Continuous glucose readings | Sensor under the skin | 9.2% mean absolute relative difference in adults with diabetes, about 2.4 minutes lag | Daily glucose patterns to discuss with a clinician |
| Clinic records | Notes, results, prescriptions | Clinicians and laboratories | The signed clinical record | The reference for decisions about care |
| Hospital letters | Referrals and discharge summaries | Hospital clinicians | 8% of US family physicians report an ideal experience receiving outside hospital results | Context every new clinician should see |
What a Profile can hold
A wai Profile accepts readings from Apple Health, Oura, Fitbit and glucose monitors, along with clinic records and hospital letters. Connect what you have and add more later, or never.1
A resting heart rate that drifts up over a fortnight means more beside a new prescription, and a sleep change means more beside a letter about a procedure. The wearables page lists the devices in detail.
Heart rate and activity
Stanford researchers tested seven wrist-worn devices on 60 adults of diverse age, weight, skin tone and fitness, against continuous telemetry and indirect calorimetry, while sitting, walking, running and cycling.2 Six of the seven devices had a median heart rate error below 5% during cycling.2
No device in the same study estimated calories burned with an error below 20%.2 Error was also higher during walking, for men, at higher body mass index and for darker skin tones, so read activity and calorie figures as rough trends.
Sleep
In a laboratory study of 34 healthy young adults over three nights, consumer sleep trackers were compared with polysomnography, the clinical standard, and sensitivity for sleep was at least 0.93 for every device.3
Detecting wake was harder, with specificity between 0.18 and 0.54, sleep stage results were inconsistent, and devices performed worse on nights of disrupted sleep.3 Total sleep and timing are useful trends to share with a clinician, and stage-by-stage breakdowns are approximate.
Irregular pulse notifications
In a study of 419,297 smartwatch users monitored for a median of 117 days, 0.52% received an irregular pulse notification.4 Among those notified who returned an ECG patch with usable data, 34% had atrial fibrillation on the patch, and 84% of notifications that came while the patch was worn matched atrial fibrillation on it.4
A notification is a reason to talk to a clinician, who can arrange a proper assessment. With the notification on the same timeline as your clinic results, the clinician sees when it happened and what else changed around it. If you feel unwell at any point, call your local emergency number.
Continuous glucose monitors
In a study of a 14-day factory-calibrated sensor across seven US sites, the mean absolute relative difference against laboratory reference was 9.2% in 144 adults and 9.7% in 129 children, and 93.2% of adult readings fell within 20% or 20 mg/dL of the reference.5 The sensor lagged blood glucose by about 2.4 minutes in adults.5
Those figures come from people with type 1 or type 2 diabetes in supervised sessions. Bring patterns across the day to a clinician, and keep any change to treatment for that conversation.
Clinic records and hospital letters
Clinic notes, results, prescriptions, referrals and discharge letters are the clinical record, written and signed by clinicians. In a survey of 8,122 US family physicians, only 8% reported an ideal experience receiving test results from outside hospitals or health systems.6
Holding those letters yourself puts them in front of every new clinician you see. Our guide to exporting or deleting your health record covers your right to take or remove that record.
Where to start
First, open a Profile on the set up your Profile page and connect the one device you wear most. Second, add the last year of clinic results and letters you hold. Third, before your next appointment, note one or two trends you want to discuss, such as resting heart rate or sleep timing.
Fourth, ask your clinic to send future results into your Profile, using the message in our guide to asking your clinic. Fifth, review your sharing permissions every few months and switch off anything you no longer use, and your next appointment starts with the whole picture.
Bring your devices and your letters together, and give your next clinician the whole picture in one timeline.
Frequently asked questions
Which devices can I connect to a wai Profile?
Apple Health, Oura, Fitbit and glucose monitors, alongside clinic records and hospital letters.1 You choose what to connect.
Are wearable readings accurate enough for my clinician to use?
My watch flagged an irregular pulse. What should I do?
Speak to a clinician. In a large study, a third of notified people who wore an ECG patch later had atrial fibrillation on it, so a notification needs a proper assessment.4 If you feel unwell now, call your local emergency number.
Does the Profile interpret my readings for me?
It gives cited answers about your own results, grounded in your record, and makes clear when it is time to see a clinician.1 It does not diagnose.
Can I disconnect a device later?
Yes. Every permission is explicit and revocable, and you can export or remove the whole record.1
Is my wearable data sold?
No. Readings in a wai Profile are never sold, and they are not used to train external models.1
Sources
- 1wai, 2026wai Profiles, a personal health record for every patientRead 14 October 2026Read source
- 2Journal of Personalized Medicine, Shcherbina and colleagues, Stanford University, 2017Accuracy in Wrist-Worn, Sensor-Based Measurements of Heart Rate and Energy Expenditure in a Diverse CohortRead 14 October 2026Read source
- 3Sleep, Chinoy and colleagues, 2021Performance of seven consumer sleep-tracking devices compared with polysomnographyRead 14 October 2026Read source
- 4New England Journal of Medicine, Perez and colleagues, 2019Large-Scale Assessment of a Smartwatch to Identify Atrial FibrillationRead 14 October 2026Read source
- 5Journal of Diabetes Science and Technology, Alva and colleagues, 2022Accuracy of a 14-Day Factory-Calibrated Continuous Glucose Monitoring System With Advanced Algorithm in Pediatric and Adult Population With DiabetesRead 14 October 2026Read source
- 6JAMA Network Open, Everson and colleagues, 2025EHR Interoperability Experiences Reported by Family PhysiciansRead 14 October 2026Read source
Your devices and your records on one timeline your clinician can read.
Set up your wai Profile, connect what you wear and add your letters, with export and removal open to you at any time.
Partner with us
See it on your own patients, then build it with us
Thirty minutes on your own patients, doctors and rules, in your market. We reply within one working day, and the clinics that partner with us early shape what we build next.
Read how patient memory works in a clinic