# The Effects of Digital Health Interventions for Family Members in Intensive Care Units: Systematic Review and Meta-Analysis of Randomized Controlled Trials

## TL;DR

This study reviews whether digital health tools help reduce stress and improve communication for families of ICU patients, but finds mixed and uncertain results.

## Contribution

This is the first comprehensive meta-analysis of randomized trials focusing specifically on digital health interventions for ICU family members.

## Key findings

- Digital health interventions showed no statistically significant reduction in anxiety, depression, or posttraumatic stress disorder among ICU family members.
- The certainty of evidence was low to moderate, with wide prediction intervals indicating substantial variability in intervention effects.
- The study highlights the need for future research on personalized and theory-informed digital health interventions for ICU families.

## Abstract

Family members of intensive care units (ICUs) often experience psychological difficulties like anxiety and depression, which affect family functioning and their ability to support patient recovery. Digital health interventions (DHIs) offer accessible and timely support; yet, evidence regarding their effectiveness for ICU families demonstrates inconsistent.

This systematic review and meta-analysis aimed to explore the impact of DHIs on psychological outcomes, quality of life (QoL), and quality of communication (QOC) among ICU family members.

A total of 9 databases (CNKI [China Knowledge Resource Integrated Database], Wanfang, VIP [Weipu Database], SinoMed, Cochrane Library, PubMed, Embase, CINAHL, and Web of Science) were searched for randomized controlled trials (RCTs) published up to October 11, 2025. Eligible studies evaluated DHIs targeting adult ICU patients’ family members and reported psychological outcomes, QoL, or QOC; interventions used solely for monitoring or tracking were excluded. Furthermore, 2 reviewers independently screened studies, extracted data, and assessed the risk of bias using the Cochrane Risk of Bias Tool. Random-effects meta-analyses were performed using the Hartung-Knapp-Sidik-Jonkman method. Prediction intervals (PIs) were calculated to reflect between-study heterogeneity. Subgroup analyses were performed by patients’ primary diagnosis, relationship with patients, and type of digital health technology. Evidence quality was assessed using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach.

In total, 17 RCTs involving 1864 participants were included. Interrater agreement was high across study selection and assessment processes (κ=0.55-1.00). Most studies were judged to have some concerns about the risk of bias. Meta-analyses showed no statistically significant pooled effects of DHIs on anxiety (standardized mean difference [SMD] –0.34, 95% CI –0.68 to 0.00; 95% PI –1.46 to 0.79; P=.05), depression (SMD –0.26, 95% CI –0.52 to 0.01; 95% PI –1.11 to 0.60; P=.06), posttraumatic stress disorder (SMD –0.21, 95% CI –0.49 to 0.06; 95% PI –1.03 to 0.60; P=.11), QoL (SMD 0.09, 95% CI –0.10 to 0.28; 95% PI –0.41 to 0.60; P=.36), or QOC (SMD 0.14, 95% CI –0.03 to 0.31; 95% PI –0.26 to 0.55; P=.10). Wide PIs indicated substantial variability in intervention effects across settings. The certainty of evidence ranged from very low to moderate.

This review emphasizes significant uncertainty in estimated effects and provides the first thorough synthesis of RCT evidence on DHIs for ICU family members. The inherently high heterogeneity and low certainty of the underlying evidence limit the conclusions of this review, despite its methodological rigor and use of PIs. Compared with previous reviews, this review concentrates on the ICU family members and includes more recent research. Future studies should focus on high-risk subgroups, use mixed methods designs, and develop theory-informed, personalized, and interactive DHIs. DHIs may enable proactive, customized communication techniques for families of ICU patients, even when there are few formal clinical recommendations.

PROSPERO CRD420251044704; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251044704