Pre-existing mental health disorders in patients admitted to the intensive care unit: A systematic review and meta-analysis of prevalence.

Abstract

To determine the reported prevalence rate of pre-existing mental health disorders in patients admitted to adult ICUs and identify the most commonly occurring types of these disorders. Systematic review and meta-analysis. Five electronic databases were searched from 1 January 2000 -15 April 2020. Google Scholar was used to perform forwards citation searching. This review was conducted in line with the PRISMA guidelines and protocol registered with PROSPERO CRD42020181818. Meta-analyses were performed using the quality effects model to calculate weighted pooled prevalence estimates and heterogeneity was tested using the I2 statistic. Seven articles were included in the final review and meta-analysis (143,179 participants). Identified prevalence rates varied considerably, ranging from 6.2-28.0%, reflecting variation in each study's clinical context, as well as different patient selection and identification methodologies. The pooled prevalence rate of all pre-existing mental health disorders was 19.4% (95% CI 8.9-32.6%). Depression was the most common subtype, accounting for an estimated 60.5% (95% CI 54.4-66.5%) of identified mental health disorders. All analyses showed significant heterogeneity with I2 >95%. Approximately 19% of adult ICU patients have a history of a mental health disorder, most commonly depression. Further research is needed to improve the accuracy of this estimate as well as determine the best identification method. This study has demonstrated that patients with pre-existing mental health disorders, particularly depression, constitute a significant subgroup in ICU. Given that the presence of a pre-existing mental health disorder appears to confer an increased mortality risk following ICU discharge, clinicians need to be made aware of this group of patients to provide additional support. Further research is needed to more accurately quantify this vulnerable group and establish methods to enable clinicians to readily identify and refer these patients for appropriate follow-up treatment.