All of Us Program Links 14 Years of Wearable Data to Clinical Records

Integrating Continuous Digital Monitoring into Clinical Practice

Practicing physicians increasingly manage patients who utilize wearable sensors to monitor physiological parameters, yet integrating this high-frequency data into evidence-based care remains difficult due to technical bottlenecks and a lack of standardized clinical protocols [1]. Large-scale meta-analyses involving over 163,000 participants indicate that these devices effectively increase physical activity by 1,800 to 2,592 steps per day and improve body composition (standardized mean difference 0.7 to 2.0), though their impact on hard clinical endpoints like mortality is less established [2, 3]. In specialized populations such as cancer survivors, wearable-supported interventions have demonstrated significant increases in moderate-to-vigorous-intensity physical activity (standardized mean difference 0.66, 95% CI 0.47 to 0.86, p < 0.001) [4]. Although current guidelines for atrial fibrillation and heart failure emphasize risk factor modification, they often lack specific instructions for utilizing continuous digital health metrics [5, 6, 7]. Furthermore, the utility of these technologies depends on patient adherence and the ability of healthcare systems to process unstructured data using machine learning (a statistical method where computers identify patterns in data to predict clinical outcomes), which is currently applied to patient stratification in cardiovascular disease and oncology [8, 9].

Longitudinal Wearable Data Across Diverse Populations

The All of Us Research Program has recently characterized one of the largest and most demographically rich digital health technology datasets to date, providing a robust foundation for longitudinal clinical research. These digital health technologies (DHTs), defined as electronic tools that use sensors and software to capture health data directly from users, offer clinicians unique insights into disease detection through continuous, real-world data collection. By moving beyond the intermittent snapshots provided by traditional office visits, this dataset captures the daily physiological fluctuations of more than 59,000 participants. The longitudinal scope of this data is particularly extensive, spanning a period of 14 years, which allows researchers to observe long-term health trends and the gradual development of chronic conditions. To ensure these findings are applicable to the diverse patient populations seen in everyday clinical practice, the program utilized a nationwide device distribution effort to reach a broad range of participants. This initiative addressed a common limitation in digital health research by including individuals from varied socioeconomic and geographic backgrounds who are frequently underrepresented in clinical trials. The resulting dataset is massive in scale, containing more than 39 million step observations and more than 31 million sleep observations. For the practicing physician, this volume of data provides a high-resolution view of physical activity and rest patterns that can be directly linked to clinical outcomes, moving wearable devices closer to becoming validated diagnostic tools.

Multi-Modal Linkage to Clinical and Genomic Outcomes

The clinical utility of the All of Us dataset is significantly enhanced by its multi-modal data linkage, a research framework that integrates high-frequency wearable data with traditional medical markers to create a comprehensive patient profile. Specifically, nearly half (46%) of participants with Fitbit data also contributed electronic health records, providing a longitudinal view of their medical history alongside daily activity metrics. This linkage allows researchers to move beyond self-reported activity levels and instead utilize objective data to understand how daily behaviors correlate with documented diagnoses, laboratory results, and medication adherence. In addition to electronic health records, participants contributed physical measurements, such as height, weight, and blood pressure, which were recorded during in-person visits to establish a baseline of clinical health status. Beyond traditional clinical markers, the dataset incorporates genomics data linked directly to the wearable metrics. This integration of genetic information with continuous digital monitoring allows for the investigation of how specific genetic predispositions interact with lifestyle factors to influence disease progression. The resource also includes integrated survey data, which captures patient-reported outcomes, socioeconomic factors, and lifestyle habits that are often missing from standard clinical charts. By combining these diverse data streams, the resource enables the study of relationships between digital health metrics and clinical outcomes with a level of granularity previously unavailable in large-scale population studies. For the practicing physician, these findings point toward a future where wearable data is not merely a lifestyle metric but a validated clinical tool. By establishing robust connections between daily step counts or sleep patterns and objective clinical endpoints, researchers aim to improve real-world disease detection and refine the accuracy of longitudinal health monitoring. This comprehensive approach ensures that the insights derived from digital health technologies are grounded in the rigorous context of a patient's full clinical and genetic profile, ultimately paving the way for more personalized and proactive patient care.

References

1. Lu L, Zhang J, Xie Y, et al. Wearable Health Devices in Health Care: Narrative Systematic Review. JMIR mhealth and uhealth. 2020. doi:10.2196/18907

2. Ferguson T, Olds T, Curtis R, et al. Effectiveness of wearable activity trackers to increase physical activity and improve health: a systematic review of systematic reviews and meta-analyses. The Lancet Digital Health. 2022. doi:10.1016/s2589-7500(22)00111-x

3. Kirk MA, Amiri M, Pirbaglou M, Ritvo P. Wearable Technology and Physical Activity Behavior Change in Adults With Chronic Cardiometabolic Disease: A Systematic Review and Meta-Analysis. American Journal of Health Promotion. 2018. doi:10.1177/0890117118816278

4. Wang Z, Li Y, Wang Q, Su Y. The Effectiveness of Wearable Electronic Device System–Supported Physical Activity Programs for Cancer Survivors: Meta-Analysis of Randomized Controlled Trials. Journal of Medical Internet Research. 2025. doi:10.2196/74347

5. Hindricks G, Potpara T, Dagres N, et al. 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). European Heart Journal. 2020. doi:10.1093/eurheartj/ehaa612

6. McDonagh TA, Metra M, Adamo M, et al. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. European Heart Journal. 2021. doi:10.1093/eurheartj/ehab368

7. Joglar JA, Chung MK, Armbruster AL, et al. 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2023. doi:10.1161/cir.0000000000001193

8. Alhumaidi NH, Dermawan D, Kamaruzaman H, Alotaiq N. The Use of Machine Learning for Analyzing Real-World Data in Disease Prediction and Management: Systematic Review. JMIR Medical Informatics. 2024. doi:10.2196/68898

9. Jakob R, Harperink S, Rudolf AM, et al. Factors Influencing Adherence to mHealth Apps for Prevention or Management of Noncommunicable Diseases: Systematic Review. Journal of Medical Internet Research. 2022. doi:10.2196/35371