by Elizabeth Short, Staff Writer, MedPage Today
A comprehensive, smartphone app-based program failed to reduce healthcare utilization among asthma patients with private insurance, but did benefit those on Medicaid, a large randomized controlled trial found.
Emergency department (ED) visits over 12 months were no different for commercially insured participants randomized to the active arm of the study (where the app contained asthma self-management features) and those assigned to a control arm (where the app did not), reported Bronwyn Harris, MD, of Stanford Medicine Children’s Health in Palo Alto, California.
But Medicaid patients in the study had a 43% reduction in risk (95% CI 17-69, P<0.01), she detailed here at the American College of Allergy, Asthma & Immunology annual meeting.
All patients in the study were considered at-risk for asthma-related ED visits and had to own an iPhone to participate. They received a free Apple Watch, a Beddit Sleep Monitor, and access to the study app.
Only patients in the active arm had access to the app’s asthma self-management features, which allowed them to log their symptoms each day and any associated triggers. Of note, the fully digital intervention involved no educators or providers.
Harris explained that to date, limited evidence exists about the impact of digital tools on healthcare utilization, and not just when it comes to asthma control. Furthermore, current home management for the condition is hampered by poor engagement with symptom tracking and an inability to record passive measures.
On top of those challenges, she said, significant disparities exist in asthma care between commercially insured and Medicaid patients.
In the study, passive data collection captured by the Apple Watch and sleep monitor — heart and respiratory rates, oxygen saturation, nighttime awakenings, step count activity, and local Air Quality Index — also prompted users to log symptoms. Patients were also asked every 2 weeks to log medication usage and adherence.
“If someone is missing more than three doses in the past 2 weeks, we’ll ask about what barriers they’re having to taking their medications — we can then give targeted education based on those barriers,” said Harris.
In cases where symptoms indicated a user’s asthma was uncontrolled, the app provided nudges for the patient to utilize corresponding educational materials.
After about 3 months, participants in the active arm could access a detailed PDF of their usage history — symptoms, exacerbations, quality of life, and more — that they could save and bring to medical appointments. Though self-reported, this feature could be used to help patients advocate for their own healthcare, said Harris, particularly if they feel their concerns are being minimized or ignored by providers.
“It just gives such a clear picture,” she told MedPage Today. “I think that will be very valuable to help even the playing field.”
Data analysis for the study included 899 participants across 41 states. Mean age was 37 years, 71% were women, 76% had commercial insurance, and 24% were insured through Medicaid. Just under two-thirds were white, 22% were Black, and 14% were Hispanic.
In terms of disease parameters, 61% had uncontrolled asthma (based on Asthma Control Test scores) and 39% had an asthma-related ED visit in the past 12 months. Sponsored by Apple and Elevance Health (formerly Anthem), the two-year study’s primary outcome measures are the impact of the program on costs for unplanned care and Asthma Control Test scores, which will be reported at a later date.
After enrollment, patients received a $25 gift card if they set up the devices within the first few weeks and could keep the Apple Watch and sleep monitor if they participated in surveys and used the devices regularly. There were no other incentives, by design, said Harris. “None of these incentives were actually tied to logging symptoms, because we wanted that symptom logging and engagement for the tools to be organic.”
Over 12 months, commercially insured patients had a similar ED visit rate whether they were assigned to the active arm or to the control group (0.33 vs 0.31 per patient-year). In the Medicaid subgroup, these rates were 0.33 versus 0.58 per patient-year, respectively, with a number-needed-to-treat of 4.1 to prevent one ED visit.
During the first 3 months, participants with Medicaid logged symptoms in the app a median 12.7 times per month while the commercially insured did so 13.7 times. App opens were a median 24 and 27.7 times per month, respectively.