Continuous, AI-driven, long-term patient monitoring could lead to dramatic reductions in hospital stays, emergency department visits and overall costs, particularly for patients with the two most common pulmonary conditions, COPD and asthma.
That’s according to a two-year study conducted by researchers from Intermountain Health, a nonprofit healthcare system based in Salt Lake City, and CareCentra, a New York City-based company that developed an AI-driven respiratory care platform.
The Intermountain COPD and Adult Asthma Remote Evaluation (iCARE) study documented 1,200 patients at five Intermountain Health hospitals in Utah over a two-year period to observe what happens when a health system intervenes to prevent patients from deteriorating while also offering support for those patients at home.
Each patient in the study carried a digital spirometer (the MIR Smart One from Medical International Research) that measured lung function. Some patients also carried a combination of connected monitoring devices, including a sensor-equipped inhaler jacket that logged medication technique and adherence, a pulse oximeter and personal fitness and lifestyle trackers.
These devices generated more than 11.5 million data points across the program — an average of about 24 signals per patient per day. Those signals were fed into CareCentra’s AI platform — a continuous stream of data that the program used to evaluate each reading against individualized baselines and guideline-based protocols.
Peter Crossno, MD, principal investigator and senior medical director of respiratory care at Intermountain Health, said a critical component of the system was that it did not wait for symptoms to become emergencies.
“It watched for early signals of deterioration: a drop in forced expiratory volume in one second (FEV1), consecutive missed doses, changes in breathing and sleep patterns,” said Dr. Crossno in a press release. “When signals converged into risk, the AI acted — coaching patients, adjusting care plans or escalating to a pulmonary disease navigator before a crisis required emergency care.”
Intermountain Health
The study operated on a tired intervention logic. CareCentra’s AI acted as the first line of response — detecting digital signals of risk and guiding patient behavior before the situation escalates. When FEV1 dropped beyond a certain threshold, the system engaged inhaler technique and ensured adherence with real-time warnings and alerts sent directly to patients.
If the signals indicated an imminent exacerbation, the system escalated to a pulmonary disease navigator — a registered respiratory therapist trained to coordinate care across the patient’s entire clinical team.
“This is not call-center triage. It’s precision escalation,” Dr. Crossno said. “Every navigator contact is triggered by data, including by a patient’s ability to recognize and report their own deterioration. They get full context to enable preventive interception of risk that drives reduced hospitalizations and emergency department visits. The productivity gains are as striking as the clinical ones.”
Data from the study showed a 51% reduction in in-patient admissions, a 20% reduction in emergency visits, a 57% reduction in per-patient costs and a 67% increase in medication adherence. Prior to the study, each pulmonary disease navigator managed approximately 20 patients. With AI assistance in routing monitoring and prioritizing, a single navigator was able to monitor nearly 220 patients.
"The human element of care can never be replaced, but we can make it significantly more efficient by bringing care directly to the patient when they need it most,” said Kim Bennion, director of respiratory care research at Intermountain Health and a member of the research team. “… This will allow us to provide higher levels of care for more of our patients.”
Dr. Crossno said that patient engagement was higher than expected, especially among the older population. Daily active use of the program was 54% and increased with age. Patients in their 80s were at 62.5% for daily active use.
Intermountain Health is considering ways to scale the iCARE technology across its network of 34 hospitals and 400 clinics.
