Background: Previous research has shown that mobile health (mHealth) interventions are effective in reminding older adults with chronic conditions about health care appointments and promoting adherence to medication schedules. However, the evidence is limited by the short duration and poor quality of the interventions.

Objective: We evaluated the effectiveness of the Support, Monitoring and Reminder Technology for Mild Dementia (SMART4MD) tablet app in improving the quality of life (QoL) of people with mild cognitive impairment (PwMCI) and their caregivers through medication reminders and health care appointment alerts.

Methods: An 18-month pragmatic randomized controlled trial was conducted in Spain and Sweden from December 2017 to September 2020 and included 1078 PwMCI and their informal caregivers.

Results: For PwMCI, the intervention improved the primary outcome, composite Quality of Life in Alzheimer’s Disease (QoL-AD) scale, at 18 months (mean difference 0.75, 95% CI 0.07-1.42; P=.03), and a similar effect was observed at 6 months (mean difference 0.73, 95% CI 0.09-1.36; P=.02). Confirmatory secondary outcomes for PwMCI, including medication adherence (P=.12) and Mini-Mental State Examination (MMSE) score (P=.45), did not differ significantly between groups at the 18-month follow-up. Exploratory analyses showed higher total accumulated quality-adjusted life years (QALYs) at 6 months for PwMCI (mean difference 0.035, 95% CI 0.02-0.05; nominal P<.001), informal caregivers (mean difference 0.050, 95% CI 0.04-0.07; nominal P<.001), and dyads (mean difference 0.085, 95% CI 0.06-0.11; nominal P<.001). A dropout rate of approximately 40% (429/1083, 39.61%) was observed in both the intervention and control groups.

Conclusions: The SMART4MD intervention was associated with a modest improvement in PwMCI’s QoL as measured by the composite QoL-AD score. Evidence for benefits among informal caregivers and dyads was less consistent and should be interpreted as exploratory. Nominally significant findings for exploratory outcomes should be interpreted with caution, as they were not adjusted for multiplicity.