Frequency of Post-Stroke Anxiety: A Systematic Review and Meta-Analyses of Observational Studies

Abstract

ABSTRACT

Background: Psychological conditions can exist following a stroke, including anxiety. Anxiety disorders and significant anxiety symptoms are associated with negative consequences and reduced quality of life after stroke. Despite these concerns, the understanding of post- stroke anxiety’s presence remains inadequate and unclear. Therefore, this research aims to conduct an updated systematic review and meta-analysis of observational studies regarding the frequency of post-stroke anxiety.

Search Methods: A digital search was performed on Ovid Medline, Embase, PsycINFO, as well as Web of Science for pertinent and eligible studies published from 2013 to July 2024 that assessed anxiety in adult stroke survivors by screening for anxiety symptoms on a rating scale or diagnosing anxiety disorders to generate the outcome of interest, which is anxiety rate. Meta- analyses of eligible observational studies were conducted utilizing random-effects models to provide a general pooled estimate of the anxiety rate as well as anxiety rates based on assessment method, study setting, and time of assessment after stroke.

Results: Overall, 17 observational studies with 4,579 unique stroke participants were included in the review, revealing an overall anxiety rate of 23.2% (95% confidence interval: 17.8-29.9, I2: 93%, p-value<0.01). The rate of anxiety based on diagnostic interviews was 20.6% (95% confidence interval: 14.2-29, I2: 80%, p-value<0.01) and 24.2% (95% confidence interval:17.4-32.6, I2: 94%, p-value <0.01) by anxiety rating scales. Additionally, the rate of anxiety in hospital studies was 27.8% (95% confidence interval: 20.6-36.4, I2: 91%, p-value<0.01) and 5.4% (95% confidence interval: 1.2-21.1, I2: 93%, p-value<0.01) in rehabilitation settings. The frequencies of post-stroke anxiety in the acute phase were 35.2% (95% confidence interval: 14.3-63.8, I2: 95%, p-value<0.01), 19.9% (95% confidence interval: 14.3-27, I2: 93%, p-value<0.01) in the medium-term phase. Furthermore, anxiety was present in more chronic phases, showcasing frequencies of 23.8% (95% confidence interval: 16-33.8, I2 :94%, p-value<0.01) in the long-term phase and 30.4% (95% confidence interval: 21.7-40.8, I2: 83%, p-value<0.01) in the very long-term phase.

Conclusion: Post-stroke anxiety can show a frequent occurrence across various contexts and phases. The high and significant between-study heterogeneity and the potential methodological concerns within the included studies may have led to limiting the accuracy of the anxiety rates and their generalizability. However, the anxiety rate findings can still pose a concern regarding its burden among a population with a life-limiting condition such as stroke. This is critical given this psychological condition’s role as a predictor of negative outcomes in this population. Thus, anxiety screening using rating scales as well as anxiety diagnostic tools should be considered as a priority measure for patients following stroke onset not only in the initial phase but also consistently across their life span, in addition to considering interventions to alleviate its burden on this population.