Goals of care discussions in hospitalised patients: systematic review of randomised trials

Structured communication tools and decision aids are recommended to facilitate goals of care discussions (GOCDs) in hospitalised patients, but their impact remains unclear.

To determine the effectiveness of communication tools and decision aids on GOCDs between healthcare providers, patients and substitute decision makers.

PubMed, Web of Science, CINAHL and ClinicalTrials.gov were searched from January 2015 to July 2025, with backward citation tracking to identify additional trials.

Randomised trials of communication tools or decision aids for hospitalised adults or substitute decision makers were eligible, regardless of primary outcomes.

Trial characteristics, primary outcomes and methodological quality were extracted and assessed using the Core Grading of Recommendations, Assessment, Development and Evaluation framework. Random-effects meta-analyses were conducted when appropriate.

22 trials were included. Communication tools increased GOCD documentation (OR 1.46, 95% CI 1.04 to 2.05). Effects on resuscitation preferences were inconsistent (OR 1.84, 95% CI 0.95 to 3.60). Mixed effects were observed for concordant care, psychosocial outcomes and healthcare utilisation. Evidence certainty was moderate for most outcomes; several small pilot trials were rated low due to bias, imprecision or heterogeneity. Interventions were most effective when combining structured patient-facing or clinician-facing prompts such as videos, guides or electronic health record alerts.

Communication tools and decision aids improve GOCD documentation in hospitalised patients but show uncertain effects on other patient-centred outcomes or healthcare utilisation. Interprofessional training and system-level support may enhance impact, and future large trials should evaluate the effectiveness of facilitated GOCDs on important patient and health system outcomes.