How Wearable Tech Can Translate Into Practical Precision Medicine in Psychiatry

Today, Amae Health announced they are partnering with Google Health Enterprise to incorporate wearable-derived behavioral data into its precision psychiatry care model.^1,2 Wearable devices like the Fitbit are giving psychiatrists a new way to track 2 kinds of data, according to Scott Fears, MD, the chief medical officer at Amae Health and professor of psychiatry at UCLA.

The first is concrete and immediately interpretable: sleep, daily activity, heart rate, and heart rate variability. These are measures psychiatry has always cared about, but patient self-report is often unreliable. Fears noted that individuals with insomnia can overreport sleeplessness, describing themselves as not sleeping at all when objective data shows 3 to 4 disrupted hours a night. A wearable is not perfect either, he said, since lying still while awake can register as sleep, but it offers a complementary, longitudinal data source. At Amae Health, this data appears directly on a clinician's dashboard at the point of care, and tracking it over months or years reveals patterns neither patient nor provider could otherwise recall.

The second category is less visible: patterns embedded in wearable data that require deeper analysis to interpret. Fears pointed to research showing that patients with bipolar disorder do not necessarily show different total sleep or activity numbers than those without the diagnosis, but do show greater inter-day variability, with activity clustering into bursts rather than remaining consistent.

That hidden signal is expanding quickly. Fears cited Google's newly published Sensor FM, a foundation model trained on wearable sensor data from roughly 5 million people, encompassing on the order of a trillion minutes of recordings. Despite Fitbits not measuring blood chemistry, the model was able to predict metabolic markers such as insulin resistance, trained in part on a subset of roughly 5000 to 6000 users with known insulin levels.

Amae Health is pursuing a parallel approach with vocal biomarkers pulled from recorded therapy sessions. Patients and clinicians independently rate anxiety, depression, thought disorder, and suicidality after each session, and Fears said vocal features extracted from the transcripts correlate strongly with both sets of ratings.

"[We want] to use the biomarkers from both the Fitbit and voice diaries to identify very early evidence of, for example, depression decompensation earlier than the individual is able to identify it, and earlier than [when] a clinician sitting down with [them] may kind of start to see something, but isn't sure," Fears said.

The goal, he explained, is to combine both data streams to flag subtle shifts early enough to prompt a simple, low-risk intervention before a patient fully decompensates.

Dr Fears is the chief medical officer at Amae Health and a professor of psychiatry at UCLA.

References

1. Amae Health partners with Google Health Enterprise to bring wearable data integration to the treatment of complex mental health conditions. News release. August 4, 2026. Accessed August 4, 2026. https://www.globenewswire.com/news-release/2026/08/04/3338378/0/en/amae-health-partners-with-google-health-enterprise-to-bring-wearable-data-integration-to-the-treatment-of-complex-mental-health-conditions.html

2. Kuntz L. Bringing wearable data into psychiatric care: Amae Health partners with Google Health Enterprise to improve outcomes. Psychiatric Times. August 4, 2026. https://www.psychiatrictimes.com/view/bringing-wearable-data-into-psychiatric-care-amae-health-partners-with-google-health-enterprise-to-improve-outcomes