# Effectiveness of Adaptive Digital Interventions Triggered by Passive Sensing for Sleep Improvement in Adults: A Systematic Review and Meta-Analysis

## TL;DR

This study reviews how adaptive digital tools, using real-time sensor data, can improve sleep and health in adults, showing benefits for sleep quality and mood.

## Contribution

The novelty lies in evaluating adaptive digital interventions triggered by passive sensing for sleep improvement through a systematic review and meta-analysis.

## Key findings

- Adaptive interventions significantly improved wakefulness after sleep onset with moderate heterogeneity.
- Secondary outcomes like quality of life and depressive symptoms showed favorable improvements.
- Greater benefits were observed in populations with cognitive impairments using advanced sensing tools.

## Abstract

Adaptive digital interventions that respond to real-time physiological data from passive sensors are emerging as personalized tools for sleep improvement. The aim of this systematic review and meta-analysis was to evaluate the effectiveness of these interventions in improving sleep outcomes and overall health indicators in adults. A comprehensive literature search was conducted across PubMed, Embase, Cochrane CENTRAL, and ScienceDirect for studies published from January 2015 to July 2025. The included studies were randomized controlled trials (RCTs) involving adults (≥18 years), with or without diagnosed sleep disorders, evaluating adaptive digital interventions triggered by passive sensing technologies (actigraphy, wearables, smartphones), compared to static digital tools, usual care, or waitlist controls. The outcomes had to include at least one sleep-related or secondary health metric. Two reviewers independently screened the studies, extracted data, and assessed the risk of bias using the Cochrane RoB 2 tool. Meta-analyses were conducted using random effects models. Effect sizes were expressed as standardized mean differences (SMDs) with 95% confidence intervals (CIs). Heterogeneity was evaluated using the I² statistic. Twelve RCTs (n = 798 participants) were included. Adaptive interventions significantly improved wakefulness after sleep onset (WASO, SMD = 3.22; 95% CI: 3.02-3.41), with moderate heterogeneity (I² = 70.7%). The effects on the Pittsburgh Sleep Quality Index (PSQI) score, sleep efficiency, and latency were small and not significant. However, the secondary outcomes, including improvements in quality of life (SMD = 1.36), depressive symptoms (SMD = 0.53), sleep duration (SMD = 0.41), and neuropsychiatric inventory scores (SMD = -1.21), were favorable. Subgroup analyses revealed greater benefits in populations with cognitive impairments and interventions using advanced sensing tools (MotionWatch8). Adaptive digital interventions triggered by passive sensing show promise for reducing night-time awakenings and enhancing mood and quality of life. Their utility may be greatest in cognitively vulnerable populations. Further research is needed to optimize adaptivity algorithms, ensure sustained engagement, and assess long-term outcomes in real-world settings.