Medication Adherence, Mental Disorders, Mood Disorders
Conditions
Keywords
psychiatry, medication adherence, wearable, smartwatch, randomized trial, artificial intelligence, digital health, young adult
Brief summary
Medication nonadherence remains difficult to measure objectively in psychiatric care. This completed, single-center exploratory randomized trial evaluated whether four weeks of AI-enabled smartwatch medication monitoring improved clinician-rated medication adherence compared with waitlist usual care in psychiatric outpatients aged 12 years or older, predominantly adolescents and young adults with mood disorders.
Detailed description
This was a single-center, prospective, exploratory randomized controlled trial conducted in an outpatient psychiatric clinic. Participants aged 12 years or older were randomized 1:1 to an AI-enabled smartwatch medication-monitoring system or waitlist usual care for four weeks. Participants in both groups continued their usual psychiatric care, and no additional medication-adherence counseling was provided as part of the intervention. Medication adherence was assessed at baseline and at Week 4 using the clinician-rated Korean Brief Adherence Rating Scale (K-BARS). The primary analysis compared the change in K-BARS score between the randomized groups. Control participants were offered the smartwatch for an additional four weeks after completing the randomized comparison period. Analyses of responder thresholds, device engagement, and the optional post-waitlist smartwatch period were exploratory. The study was registered retrospectively after enrollment and study completion.
Interventions
DEVICEAI-enabled smartwatch
The smartwatch was designed to detect medication-taking behavior using a built-in camera and electronic tags attached to medication containers. The camera remained inactive under ordinary conditions and captured a brief sequence only when container motion, ambient light, and close proximity between the watch and tag were detected concurrently. The captured sequence was processed by an artificial intelligence algorithm to classify whether a medication-taking event had occurred. Participants assigned to the intervention were instructed to wear the device for four weeks. The device provided monitoring only and delivered no reminders, alerts, counseling, or treatment recommendations.
Sponsors
Wonkwang University Hospital
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
SINGLE (Outcomes Assessor)
Masking description
The BARS outcome assessor was blinded to treatment allocation at the time of rating; ratings were based on structured interview and pill-count inputs without access to device-recorded data. Participant blinding was not feasible because the intervention consisted of a visible wearable device.
Intervention model description
Participants were randomized 1:1 to the AI-enabled smartwatch intervention or waitlist usual care. The randomized comparison period lasted four weeks. At baseline, participants assigned to the intervention received smartwatch training, and electronic tags were attached to their active medication containers. Participants in both groups continued usual psychiatric care, with no additional medication-adherence counseling provided as part of the study. Medication adherence was assessed using the clinician-rated Korean Brief Adherence Rating Scale at baseline and Week 4. After completing the randomized comparison period, control participants were offered optional smartwatch monitoring for an additional four weeks; data from this optional period were analyzed exploratorily and were not part of the primary randomized comparison.
Eligibility
Sex/Gender
ALL
Age
12 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Age 12 years or older * Receiving outpatient psychiatric care at Wonkwang University Hospital * Receiving at least one regularly prescribed psychotropic medication * Willing and able to wear the study smartwatch and complete the study assessments * Able to provide written informed consent; for minors, written assent and written consent from a parent or legal guardian were required
Exclusion criteria
* Diagnosed congenital genetic disorder * Clear history of acquired brain injury * Seizure disorder or another neurological condition that could interfere with study participation * Psychotic symptoms, aggression, or behavioral instability severe enough to prevent safe and stable smartwatch use or create a substantial risk of device damage
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Mean Change From Baseline in Korean Brief Adherence Rating Scale Score at Week 4 | Baseline and Week 4 | Medication adherence was assessed using the clinician-rated Korean Brief Adherence Rating Scale (K-BARS). The clinician assessed the number of prescribed doses, the number of doses reportedly taken, and the frequency of missed doses over the preceding 28 days and integrated this information into a score ranging from 0% to 100%, with higher scores indicating greater medication adherence. For each participant, the change score was calculated as the Week 4 score minus the baseline score. The primary outcome was the difference in mean change score between the randomized groups. |
Countries
South Korea
Outcome results
None listed