It’s the kind of dataset that makes a geneticist’s heart race — and for good reason. The National Institutes of Health (NIH) this week announced the launch of what it calls the world’s largest integrated health database, combining whole-genome sequences from more than half a million participants with their electronic health records, survey responses, and data from wearable devices like smartwatches and fitness trackers. The resource, built under the All of Us Research Program, gives researchers an unprecedented tool to untangle the genetic, environmental, and lifestyle factors that drive disease — and, just as critically, to understand why some treatments work in some people but fail in others.
“This is a game-changer for precision medicine,” said Dr. Josh Denny, CEO of the All of Us Research Program, in a press briefing. “We’re not just sequencing genomes; we’re linking them to real-world data on how people live, move, and get sick. That combination is what will finally let us move from one-size-fits-all medicine to care tailored to each individual.”
A Half-Million Genomes and Counting
The database, officially called the All of Us Researcher Workbench, currently holds whole-genome sequences from 510,000 participants, making it roughly 10 times larger than the next biggest repository, the UK Biobank’s exome-only data. But size isn’t the only story. What sets this resource apart is the depth of accompanying information. For each participant, the NIH has collected longitudinal electronic health records stretching back years — sometimes decades — along with self-reported data on diet, exercise, sleep, and mental health. And then there’s the wearable data.
More than 150,000 participants have shared continuous data from wearable devices, capturing step counts, heart rate variability, sleep stages, and even minute-by-minute activity patterns. That allows scientists to ask questions no one could before: How does a specific genetic variant affect your heart’s response to exercise? Does a particular drug interact with a sedentary lifestyle to worsen outcomes? “We’re moving from snapshots to movies of human health,” said Dr. Euan Ashley, a cardiologist and genomics researcher at Stanford University who is not involved with the program. “With this data, you can watch the trajectory of a disease unfold in real time, and see where a drug might intervene — or fail to.”
Early results are already emerging. In a preprint posted on medRxiv last month, a team from Vanderbilt University used the database to identify 16 new genetic loci associated with atrial fibrillation, a common heart rhythm disorder, by analyzing wearables that detected irregular pulses. Another group from the University of California, San Francisco found that a seemingly benign variant in the CYP2D6 gene — known to affect how people metabolize antidepressants — was tied to a 40% higher rate of treatment discontinuation, but only in participants who reported high stress levels. “That kind of interaction is invisible in a standard genome-wide association study,” Ashley noted. “You need the environmental context, and this database provides it.”
Wearables Add a New Dimension
The integration of wearable data is perhaps the most novel — and controversial — aspect of the NIH’s new resource. Participants in the All of Us program can choose to link their Fitbit, Apple Watch, or other devices, allowing researchers to access granular activity and physiological data. That opens the door to studying how lifestyle choices interact with genetics to shape health outcomes. It’s a theme that resonates with a growing body of research showing that your sedentary lifestyle is quietly breaking your cells’ power plants — the mitochondria. Pairing those insights with genomic data could reveal why some people stay healthy despite sitting all day while others develop metabolic disease.
But the wearable data also comes with privacy concerns. The NIH has built multiple layers of security: all participant data is de-identified, and researchers must apply for access, undergo training, and agree not to try to re-identify individuals. Still, some experts worry that as the dataset grows, the risk of re-identification increases. “There’s no such thing as perfect anonymity,” said Dr. Jennifer Wagner, a bioethicist at the University of Pennsylvania who studies genomic privacy. “The NIH has done a commendable job so far, but we need ongoing oversight, especially as artificial intelligence tools become more powerful at linking disparate data points.”
To address those concerns, the program has established a dedicated Privacy and Trust Advisory Board, and participants can withdraw at any time, with their data removed from future analyses. The NIH also recently launched a public dashboard showing how the data is being used — who is accessing it, for what studies, and what findings are emerging.
What This Means for You — and Your Doctor
For the average person, the immediate impact may not be obvious. But over the next five to ten years, the database is expected to reshape clinical guidelines and drug prescribing. “When your doctor chooses a blood pressure medication or an antidepressant, they’re often guessing based on population averages,” said Dr. Denny. “With this resource, we can start giving them evidence-based guidance: ‘This drug works best for patients with your genetic profile and activity level.’”
That kind of specificity could be a boon for conditions like diabetes, heart disease, and depression, where trial-and-error prescribing is common. It could also help uncover why certain drugs that work in clinical trials fail in real-world settings — a phenomenon known as the efficacy-effectiveness gap. “The people in clinical trials are often healthier, younger, and less diverse than the general population,” explained Dr. Lucila Ohno-Machado, a biomedical informatician at Yale University who is not part of the program. “This database represents the real world: people with multiple chronic conditions, different ancestries, and varied lifestyles. That’s exactly the data we need to close that gap.”
The diversity of the participants is a deliberate focus. Unlike earlier biobanks that skewed heavily White and European, the All of Us program has enrolled 49% of participants from racial and ethnic minority groups. Nearly 90% of participants have provided consent for their data to be used in research on any disease, not just the one that originally affected them. That broad consent, combined with the rich data, makes the resource uniquely powerful for studying health disparities. A team at Morehouse School of Medicine, for instance, is already using the database to investigate why Black Americans have higher rates of treatment-resistant hypertension, even after controlling for known risk factors.
Privacy, Trust, and the Long Game
Building a database of this scale is not just a scientific challenge — it’s a social one. The NIH has invested heavily in community engagement, holding listening sessions, partnering with churches and community health centers, and offering participants the chance to see their own genetic results if they choose. About 80% of participants have opted to receive clinically relevant findings, such as variants linked to breast cancer or high cholesterol. That feedback loop, researchers say, is crucial for maintaining trust.
But trust is fragile. In 2020, a security breach at a different NIH-funded repository exposed the genomic data of some 10,000 participants, leading to a wave of scrutiny. The All of Us program says it has learned from that incident, implementing end-to-end encryption and requiring researchers to undergo annual ethics training. “We can’t afford another breach,” Denny said. “This is about people’s most sensitive data. We have to earn their trust every day.”
Looking ahead, the NIH plans to double the database to one million genomes by 2028, and to include more types of data — from metabolomics and proteomics to continuous glucose monitors. The hope is that the resource will eventually become the backbone of a new, data-driven system of medicine, one where your genetic code, your daily step count, and your medical history are all woven together into a single, actionable portrait of your health. It’s an ambitious vision. But with half a million genomes already in hand, the foundation is laid.
Frequently Asked Questions
Who can access the NIH integrated health database?
Access is available to researchers worldwide who submit an application through the All of Us Researcher Workbench. Applicants must have an institutional affiliation, complete training on data use and ethics, and agree not to attempt to re-identify participants. The data is tiered: some is publicly available in summary form, while individual-level data requires a more stringent review.
How is my privacy protected if I participate?
The NIH removes direct identifiers (name, address, social security number) and applies statistical methods to prevent re-identification. Research data is encrypted, and access is logged and monitored. Participants can withdraw at any time, and previously shared data will be removed from future analyses. The program also has a Privacy and Trust Advisory Board that oversees data use.
Will this database help find cures for rare diseases?
Yes. The large, diverse dataset makes it easier to spot individuals with rare genetic variants that cause disease — even if they are geographically dispersed. Researchers can then study the natural history of those conditions and test potential treatments in cell or animal models. Several rare disease groups have already begun mining the data.
