Abstract Number: Esoc2026a1792 Impact of Age and Frailty on the Effect of a Multimodal Individualized Long-Term Intervention After Stroke: The Last-Long Trial

Abstract Background and aims The randomized controlled LAST-long trial evaluated a multidimensional, stroke coordinator–led, community-based intervention and found no overall improvement in functional outcome after stroke. As the study population mainly comprised patients with mild stroke symptoms, older and frailer patients may benefit more. This was examined in the present study. Methods LAST-long was a single-blinded randomized controlled trial conducted at four Norwegian hospitals (2019-2022). Participants were randomized (1:1) to monthly meetings with a stroke coordinator for 18 months, focusing on structured risk assessment and individualized goal setting, or to standard care. The primary outcome was modified Rankin Scale (mRS); secondary outcomes included mortality, cognition, activities of daily living, physical function, fatigue, and quality of life. Higher age was defined as ≥80 years. Frailty was assessed using the Frailty Index (≥0.2). Associations between age, frailty, intervention, and outcomes were analyzed using linear and logistic regression. Results Of 301 patients, 268 (89.0%) had valid mRS data at 18 months (136 intervention, 132 control). Mean (SD) age was 71.1 (12.4) years; 66 (24.6%) were aged ≥80, and 89 (33.2%) were frail. Frailty, but not age, was associated with worse mRS. Age appeared to modify the treatment effect. For secondary outcomes, frailty consistently predicted poorer outcomes, while higher age was associated with increased mortality. The intervention was associated with better TMT-B performance in younger participants, with a significant age–intervention interaction. Conclusions In this sub-study of LAST-long, frailty was associated with unfavorable outcomes. Age, but not frailty, appeared to modify the effect of the intervention. Conflict of interest Ihle-Hansen: nothing to disclose. Thommessen: nothing to disclose. Langlo: nothing to disclose. Bergh: nothing to disclose. Hokstad: nothing to disclose. Lydersen:nothing to disclose. Seljeseth: nothing to disclose. Saltvedt: nothing to disclose. Askim: nothing to disclose.