Background: We recently demonstrated that late functional improvement between 3-12 months post-stroke occurs in about one-fourth of patients with ischaemic stroke. It is unknown whether this improvement is associated with better long-term clinical or health-economic outcomes. Methods: In a prospective, population-based cohort of 1-year ischaemic stroke survivors (Oxford Vascular Study;2002-2014), we determined changes in functional status (modified Rankin Scale[mRS], Rivermead Mobility Index[RMI], Barthel Index[BI]) from 3-12 months post-stroke. We examined the association of late improvement (by ≥1 mRS grades, ≥1 RMI points, and/or ≥2 BI points) with 5-year mortality, institutionalization (Cox regressions), and health/social-care costs (generalized linear models), adjusted for age/sex/3-month disability/stroke subtype. Results: Among 1,288 1-year survivors, 1,135 had 3-month mRS>0, with 319(28.1%) demonstrating late improvement. 1-year survivors with late mRS improvement had lower 5-year mortality (aHR:0.68,95%CI 0.51-0.91,p=0.009), institutionalization (aHR:0.48,0.33-0.72,p<0.001), and costs (margin -$17,369,-25,271 to -9,469,p<0.001). These associations remained on excluding patients with recurrent strokes during follow-up (e.g. 5-year death/institutionalization aHR:0.59,0.44-0.79,p<0.001) and on examining late improvement per RMI and/or BI (e.g. 5-year death/institutionalization aHR:0.67,0.53-0.84,p=0.001). Conclusions: Late functional improvement post-stroke is associated with lower 5-year mortality, institutionalization, and costs. These findings should motivate patients and clinicians to maximize late recovery and encourage payers to consider access to rehabilitative services for at least 1-year post-stroke.