Longitudinal studies of post-intensive care syndrome among ICU survivors: a scoping review

Background Some ICU survivors experience outcomes related to post-intensive care syndrome (PICS), including psychological distress, physical impairment, and cognitive dysfunction. Objective This review aimed to systematically summarize the assessment time points, measurement instruments, longitudinal trajectories, and associated factors of PICS-related outcomes to inform long-term follow-up and management strategies for ICU survivors. Methods PubMed, Web of Science, CINAHL, Embase, and PsycINFO were systematically searched from inception through June 10, 2026. Two reviewers independently conducted study selection and data extraction, and the findings were synthesized narratively. Results 72 studies were included: 50 were derived from distinct cohorts, and 22 belonged to eight groups of studies with overlapping parent cohorts. Psychological, physical, and cognitive outcomes were assessed in 51, 34, and 27 studies, respectively. Ten studies assessed all three domains and 20 assessed two; only six of these 30 reported patient-level co-occurrence of impairments. Assessments most commonly occurred at approximately 3, 6, and 12 months, although follow-up index events and instruments varied. Within the first 3 months, improvement was reported for anxiety in five studies, depressive symptoms in six, physical function in three, and cognitive function in two. Between 3 and 6 months, six studies described improvement in physical function and four in cognitive function, while psychological findings varied. Findings beyond 6 months were heterogeneous, with some studies describing persistent or delayed symptoms. Associated factors were reported across the pre-ICU, ICU, and post-ICU periods, but evidence specifically addressing predictors of change over time remained limited. Conclusions The available evidence suggests that trajectories may differ across PICS domains, although findings during specific follow-up periods varied across studies. In some studies, long intervals between assessments limited the ability to characterize dynamic recovery trajectories. Further research is also needed to establish standardized assessment instruments and clinically meaningful thresholds. Evidence regarding several associated factors also remains insufficient and requires further validation.