Multilevel Modelling of The Predictors of Post-Stroke Depression in South-West Nigeria

Abstract

Recent decline in stroke mortality engendered by improved treatment has led to an increase in post-stroke morbidity and related 
healthcare cost. Most of post-stroke morbidity is due to depression. Few studies have examined the association of contextual 
factors with post-stroke depression (PSD) using a multilevel framework. This study used multilevel modelling to examine both 
individual and contextual predictors of PSD in southwest, Nigeria. The study used secondary data from a comparative crosssectional study of one hundred and thirty (130) stroke survivors. Participants were consenting adult (aged ≥18 years) residing in 
southwest Nigeria who have survived stroke in the last 3 to 24 months preceding the time of the study. Data were preliminarily 
analyzed using descriptive statistics including percentages and frequency tables. Bivariate association tests were carried out using 
Chi-square test. Multivariate analyses were performed using multilevel logit modeling and results were presented as odds ratio 
and their 95% Confidence Intervals (95% CI). All analyses were performed at 5% significance level. The mean (±standard 
deviation) age of participants was 59.54 ± 11.08 and most participants (53.6%) were female. Exactly half of the participants were 
retirees while most of them were currently married (82.3%), resided in urban (76.7%), and never used alcohol (73.8%). 
Prevalence of PSD in this study was 41.5%. Post-stroke depression was more likely among younger stroke survivors (OR=1.47; 
95% CI: 1.16, 1.85), female (OR=1.37; 95% CI: 1.16, 1.60), alcohol users (OR=1.21; 95% CI: 1.04, 1.47), and the retired 
(OR=1.23; 95% CI: 1.04, 1.45). PSD was significantly associated with history of stressful life events and all depressed stroke 
survivors had considered suicide. Post stroke depression was common in younger survivors. Female survivors and alcohol users 
were more depressed. History of stressful life event is predictive of post stroke depression in this sample. Interventions focusing 
on women, alcohol users and people with history of stressful life events that had survived a stroke may reduce the burden of post 
stroke depression and possibilities of suicide.