AHA Scientific Sessions 2026 Abstract · Nov 2, 2026
Population-Level Prevalence of Nocturnal Blood Pressure Dipping Phenotypes: Insights from 90,465 Oura Ring Users
By Valkonen M, Timisjarvi V, Kazemi K et al.
Background: Nocturnal blood pressure (BP) non-dipping is an emerging independent predictor of cardiovascular events and mortality. However, population-level prevalence remains poorly characterized, as ambulatory blood pressure monitoring (ABPM) — the gold standard method for out-of-clinic blood pressure assessment — is rarely conducted at scale. Here, we applied a cuffless PPG-based blood pressure method to a large consumer wearable cohort to estimate nocturnal dipping prevalence and its clinical correlates. Methods: We retrospectively analysed deidentified data from 90,465 adult Oura Ring users with PPG monitoring across a 30-day window (mean age 42.7 ± 12.8 years; 70.5% female; BMI 27.2 ± 5.7 kg/m ² ). Estimated MAP (eMAP) was derived from a proprietary cuffless PPG algorithm. For each night, sleep-vs-daytime eMAP percentage change was assigned to one of four classes — extreme (<–20%), normal (–20 to –10%), non-dipper (–10 to 0%), reverse (>0%). Each user's modal class across available nights served as their dipping phenotype, with non-dipper and reverse classified as abnormal. Associations with self-reported cardiovascular and metabolic diagnoses were tested with χ ² , Cochran-Armitage trend, and multivariable logistic regression adjusted for age, sex, and BMI. The eMAP algorithm was previously validated against simultaneous 48-hour Oscar2 ABPM in 134 adults, with dipper-vs-non-dipper AUC 0.87 at the 10% threshold and mean absolute error 5.41 mmHg for nocturnal eMAP change. Results: Overall, 15.2% of users were extreme, 66.7% normal, 17.7% non-dipper, and 0.4% reverse dippers (18.1% abnormal). Abnormal dipping rose monotonically with age, from 13.4% at <30 years to 48.5% at ≥70 years (Cochran-Armitage z = 51.2, p < 0.001). Abnormal dipping was more common in hypertensives (29.1% vs 15.7%; χ ² = 1584, p < 0.001). Adjusted odds of abnormal dipping were elevated for hypertension diagnosis (aOR 1.33, 95% CI 1.24-1.42), type 2 diabetes (1.45, 1.32-1.58), prior cardiovascular events (1.31, 1.22-1.41), female sex (1.34, 1.28-1.40), each year of age (1.02, 1.01-1.02), and each kg/m ² BMI (1.05, 1.04-1.05). Obstructive sleep apnea was associated in unadjusted analyses (OSA+ 24.3% vs OSA– 17.4% abnormal) but attenuated to near-null after adjustment (aOR 0.94, 0.88-0.99), indicating the unadjusted association was largely explained by confounding with age, BMI, and hypertension. Conclusions: Cuffless wearable PPG identifies abnormal nocturnal BP dipping in 18.1% of the sample, rising with age and tracking established cardiometabolic risk factors. Modal-class phenotyping over many nights provides a scalable approach for population-level surveillance and risk stratification, motivating prospective validation against ABPM in outcome-linked cohorts.