The week in health tech, explained in plain English.Get the free newsletter →
September 16, 2026

Takeaways from the most recent news in the technology and policies shaping healthcare.

Health IT

EHR Marker Flags Transplant Patients at Risk of Rejection

By Healthcare Technology News ·

An objective marker pulled from electronic health records can identify transplant patients at high risk of organ rejection driven by medication nonadherence, according to Becker's Hospital Review. Researchers from New York City-based Mount Sinai Health System and Houston-based Texas Children's evaluated the tool in a multicenter randomized controlled trial published June 15 in the American Journal of Transplantation.

The marker, called the Medication Level Variability Index, measures fluctuations in a patient's immunosuppressant drug levels over time. Wide swings signal that a patient is skipping or inconsistently taking the medications that keep a transplanted organ from being rejected. Because the index is derived directly from routine lab data already sitting in the EHR, clinicians can monitor adherence without relying on patient self-reports, which are often unreliable.

The approach points to a broader shift toward using existing EHR data to surface clinical risks earlier. For transplant programs, catching nonadherence before rejection sets in could preserve scarce donor organs and prevent costly hospitalizations and re-transplants.

Sources

More in Health IT

Health IT

Patient Identity Is Healthcare AI's Missing Control

Healthcare Dive argues that fragmented patient identity data is the overlooked control that determines whether clinical AI models can be trusted.

Why it matters: AI outputs are only as good as the patient records behind them, making identity resolution a prerequisite for safe clinical deployment.

Health IT

ARPA-H's $62.7M Autonomous AI Bet Raises a Liability Question

ARPA-H picked six organizations for a $62.7 million contract to build an autonomous AI agent that manages heart failure care between visits, raising unresolved questions about who is liable when it acts on its own.

Why it matters: Autonomous clinical AI moves beyond decision support to independent action, and the legal accountability for its decisions remains undefined.