Long-term outcomes of depression up to 10-years after stroke in the South London Stroke Register: a population-based study.

Background: Current evidence on the long-term outcomes of post-stroke depression (PSD) is limited, with most studies relying on short follow-ups and cross-sectional designs. We aim to examine (1) associations between depression at 3-months and long-term outcomes-including mortality, stroke recurrence, functional ability and quality of life (QoL)- up to 10-years; (2) the impact of depression recovery and timing of onset on these associations. Methods: Data were from the South London Stroke Register (1-January-1997-20-April-2023). Depression was defined as a score >7 on the Hospital Anxiety and Depression Scale. Physical disability was measured using Barthel Index; instrumental activity of daily living (IADL) using the Frenchay Activities Index; and QoL using the Short Form-12, which provides physical and mental health summary scores. Outcomes were assessed annually up to 10-years. Cox proportional hazards models estimated the associations between PSD and mortality and stroke recurrence, while generalized estimating equation was used for physical disability and IADL and linear mixed models for QoL, adjusting for covariates. Findings: Among 2581 stroke survivors assessed at 3-months, 918 (35.6%) exhibited depression symptom. PSD at 3-month was associated with higher mortality risk (aHR 1.18, 95% CI [1.03-1.36]), but not with stroke recurrence (0.85 [0.64-1.14]) over a 10-year follow-up. The number of patients in analysing the association with physical disability, IADL and QoL was 1388, 1167, and 1292 respectively. PSD was also linked to increased odds of physical disability (aOR 2.94, 95% CI [2.12-4.09]), IADL impairment (2.89 [2.13-3.92]) and lower physical (β = -5.93, 95% CI [-7.26 to -4.60]) and mental QoL (-7.56 [-8.99 to -6.13]) scores. Compared to patients with PSD at both 3-months and 1-year, those recovered by 1-year had similar mortality risk (0.95 [0.76-1.16]), but lower stroke recurrence (0.47 [0.25-0.92]), lower occurrence of physical disability (0.55 [0.36-0.85]) and IADL impairment (0.56 [0.36-0.89]), and improved physical (3.55 [1.30-5.80]) and mental (10.91 [8.56-13.25]) QoL. PSD at 1-year or 5-years was also associated with increased mortality (1-year: 1.33 [1.15-1.53], 5-year: 1.37 [1.10-1.71]), increased risks of physical disability (1-year: 2.20 [1.77-2.74], 5-year: 2.42 [1.39-4.22]) and IADL impairment (1-year: 3.00 [2.22-4.06]; 5-year: 2.69 [1.76-4.11]) and lower physical (1-year: -6.49 [-7.60 to -5.38]; 5-year: -6.78 [-8.30 to -1.24]) and mental QoL (1-year: -12.04 [-13.25 to -10.83]; 5-year: -6.76 [-8.81 to -4.72]) scores. Interpretation: PSD had lasting impact on stroke recovery, extending significantly beyond the acute phase. As recovery from depression within 1-year is associated with improved health outcomes, further research is needed to develop effective PSD interventions and enhance long-term stroke prognosis. Funding: National Institute for Health and Care Research (NIHR202339).