Withings Hypertension Notification Retrospective Study (HTNF)

Validation of the Withings Hypertension Notification Feature (HTNF): A Retrospective Study Comparing Diagnostic Performance Against the FDA-Authorized Predicate Device (Apple Watch HTNF, K250507)

This retrospective, non-interventional clinical investigation evaluates the diagnostic performance of the Withings Hypertension Notification Feature (HTNF), a software-only medical device (SaMD) that analyzes demographic, anthropometric, and physiological parameters collected by Withings connected smart scales to identify patterns suggestive of hypertension. Ground-truth blood pressure measurements are obtained concurrently from Withings connected blood pressure monitors. The study tests whether the Withings HTNF achieves diagnostic performance substantially equivalent to the FDA-authorized predicate device (Apple Watch HTNF, K250507) for the purpose of 510(k) clearance.

Study Overview

• Status: Active, not recruiting
• Conditions: Hypertension
• Intervention / Treatment: Device: Withings Blood Pressure Monitor (reference); Device: Withings Hypertension Notification Feature (HTNF)

Detailed Description

Hypertension affects approximately 1.4 billion adults worldwide and is the leading modifiable risk factor for cardiovascular disease. In September 2025, the FDA granted 510(k) authorization (K250507) to Apple Inc. for a Hypertension Notification Feature on Apple Watch, establishing the first regulatory precedent for consumer wearable-based hypertension screening.

The Withings Hypertension Notification Feature (HTNF) is a software-only medical device (Software as a Medical Device, SaMD) that analyzes demographic, anthropometric, and physiological parameter data opportunistically collected by Withings connected smart scales to identify patterns suggestive of hypertension and provide a notification to the user. Ground-truth hypertension status is defined at the user-week level as a weekly average systolic blood pressure (SBP) >= 130 mmHg or diastolic blood pressure (DBP) >= 80 mmHg (2017 ACC/AHA guideline), based on a minimum of 7 weekly blood pressure measurements from Withings BPM devices.

The co-primary endpoints are the stage-adjusted sensitivity and stage-adjusted specificity of the Withings HTNF algorithm on a held-out validation set, evaluated via a conjunctive intersection-union test (IUT) at one-sided alpha = 0.025. Stage adjustment is performed via Direct Standardization to the predicate's stage distribution. Confidence intervals are derived using BCa bootstrap resampling with at least 10,000 replicates.

Secondary endpoints include stage-specific Se/Sp, PPV/NPV, AUC-ROC, subgroup performance (age, sex, BMI, race/ethnicity, device model, prior HTN diagnosis, antihypertensive medication), and time-to-first-false-positive analyses. For the US cohort, race, ethnicity, diagnosis, and medication history are obtained via linkage to the HealthVerity claims database.

Data were collected retrospectively between 2019 and 2026 from consumer users of commercially available Withings smart scales who received an information letter and did not opt out. No prospective measurements or interventions are performed.

• Study Type: Observational
• Enrollment (Estimated): 2500

Contacts and Locations

Study Locations

• France
  • Île-de-France Region
    • Issy-les-Moulineaux, Île-de-France Region, France, 92130
      • Withings SA

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

Adult consumer users (>= 18 years, >= 22 years in the US) of Withings connected smart scales (Body Cardio, Body Scan, Body Comp) with concurrent Withings connected blood pressure monitor data, who received the study information letter and did not opt out. The retrospective validation cohort is drawn from a HealthVerity-linked US subset to support subgroup analyses by race, ethnicity, prior hypertension diagnosis, and antihypertensive medication use.

Description

Inclusion Criteria:

• Users of commercially purchased, consumer-grade Withings smart scales (models Body Cardio, Body Comp, Body Scan) in a standard home environment
• Concurrent ground-truth blood pressure data collected via Withings connected blood pressure monitors (BPM, BPM+, BPM Connect, or BPM Core)
• At least 7 blood pressure measurements within a given week to establish reliable weekly hypertension status classification
• Age >= 18 years (all regions) or >= 22 years (United States), consistent with the predicate device age requirement
• Receipt of the study information letter without exercising the right to opt out (for applicable regulatory jurisdictions)

Exclusion Criteria:

• Users lacking valid weekly average systolic and diastolic blood pressure measurements (missing ground truth)
• Users with physiological measurements falling outside plausible clinical ranges (BMI < 10 or > 100 kg/m2, age < 18/22 or > 100 years)
• Users who exercised their right to opt out of the retrospective study
• Users without an individual Withings account (shared accounts are excluded to prevent data contamination)

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Intervention / Treatment

Withings Smart Scale Users; Retrospective cohort of adult consumer users of Withings connected smart scales (Body Cardio, Body Scan, Body Comp) with concurrent Withings blood pressure monitor measurements between 2019 and 2026. The held-out validation subset (~2,500 users) is drawn from the HealthVerity-linked US cohort to support stage-adjusted Sensitivity and Specificity primary endpoints and subgroup analyses.

Device: Withings Blood Pressure Monitor (reference); Ground-truth weekly average systolic and diastolic blood pressure measurements obtained from Withings connected blood pressure monitors (BPM [WPM02], BPM+ [WPM03], BPM Core [WPM04], BPM Connect [WPM05]), validated as non-invasive oscillometric blood pressure monitors per ISO 81060-2:2018. Minimum 7 measurements per user-week required.; Device: Withings Hypertension Notification Feature (HTNF); The Withings Hypertension Notification Feature (HTNF) is a software-only medical device (Software as a Medical Device, SaMD) that analyzes demographic, anthropometric, and physiological parameter data opportunistically collected by Withings connected smart scales to identify patterns suggestive of hypertension and provide a notification to the user.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Stage-Adjusted Sensitivity (Se) of the Withings HTNF Algorithm	Percentage of hypertensive user-weeks correctly flagged	up to 7 years
Stage-Adjusted Specificity (Sp) of the Withings HTNF Algorithm	Percentage of non-hypertensive user-weeks correctly not flagged	up to 7 years

Collaborators and Investigators

• Sponsor: Withings
• Investigators: Principal Investigator: Pierre Escourrou, MD, PhD, Centre Interdisciplinaire du Sommeil, Paris, France

Study record dates

Study Major Dates

• Study Start (Actual): March 1, 2026
• Primary Completion (Estimated): July 1, 2026
• Study Completion (Estimated): July 1, 2026

Study Registration Dates

• First Submitted: May 26, 2026
• First Submitted That Met QC Criteria: June 3, 2026
• First Posted (Actual): June 8, 2026

Study Record Updates

• Last Update Posted (Actual): June 8, 2026
• Last Update Submitted That Met QC Criteria: June 3, 2026
• Last Verified: June 1, 2026

More Information

Terms related to this study

• Keywords: Hypertension; Pulse Wave Velocity; Blood Pressure; mhealth; Connected Health; Software as a Medical Device; Smart Scale; 510(k)
• Additional Relevant MeSH Terms: Vascular Diseases; Cardiovascular Diseases; Hypertension
• Other Study ID Numbers: HTNF

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: Yes
• product manufactured in and exported from the U.S.: No