Fatigue after ischaemic stroke: A five-year longitudinal study of prevalence and latent trajectory subgroups

Abstract

Objective To investigate the prevalence and longitudinal course of post-stroke fatigue and identify subgroups with distinct fatigue trajectories up to 5 years post-stroke. Design A prospective, longitudinal, multi-centre cohort study. Setting Hospital stroke units in the Central-Norway health region. Participants Incident ischaemic stroke patients were recruited during acute hospitalisation between June 2015 and November 2017. Data were collected during the hospital stay and at 3 months, 1 year, 3 years, and 5 years. Main measures Fatigue was assessed using the 7-item Fatigue Severity Scale with > 5 as cut-off to define clinical fatigue. Results A total of 693 participants were included. The prevalence of clinical fatigue was 24% (3 months), 28% (1 year), 30% (3 years), and 28% (5 years). A Sankey diagram illustrated substantial variability in fatigue status across time points. Using group-based trajectory modelling, four latent subgroups with distinct fatigue trajectories were identified: (1) low fatigue, (2) emerging fatigue, (3) improving fatigue, and (4) high fatigue. Conclusion One quarter to one third of the study population consistently exhibited fatigue up to 5 years post-stroke. High variability in fatigue status over time was evident, indicating individual-level heterogeneity in the longitudinal course. This was further confirmed by the identification of four latent subgroups with distinct post-stroke fatigue trajectories. Early detection of individuals at risk of an adverse fatigue trajectory and differentiated follow-up may improve outcomes for stroke survivors.