King's College London

Research portal

Predicting survival and functional outcomes after first-ever ischemic stroke in a multi-ethnic population: The South London Stroke Register

Research output: Contribution to journalConference paper

Original languageEnglish
Pages (from-to)53-53
Number of pages1
JournalInternational Journal Of Stroke
Volume10
Issue numberSupplement S2
DOIs
Publication statusPublished - 15 Apr 2015

King's Authors

Abstract

Background and purpose:
We aim to accurately predict mortality and functional outcomes at three months, six months and one year after first-ever ischemic stroke.

Methods:
3,461 ischemic stroke patients were registered in the South London Stroke Register (SLSR) between 1995 and 2013. Outcome measures are survival and independence in activities of daily living (Barthel Index ≥ 15) at 3 months, 6 months and 1 year after stroke onset. Patients registered up to 2005 were used to build and train the predictive models, while those registered after 2005 were used to test the models for predictive accuracy. Cox Proportional Hazards models were used in survival analysis to identify prognostic factors for mortality, while multivariate logistic regression analyses were used to predict functional outcomes. All the models were validated and assessed for predictive ability by using the area under the receiver operating characteristic curve (AUROC).

Results:
Five models were developed to predict survival and functional outcomes at three months, six months and one year after stroke onset. The final models comprised age, ethnicity, stroke subtypes, case mix, acute stroke care and prior-stroke risk factors. All predictive models showed satisfactory predictive power with AUROC of 0.85 to 0.90 for both training and validation datasets.

Conclusion:
Our population-based study gives important insights into prognostic factors after acute ischemic stroke and presents predictive models with high predictive accuracy for predicting survival and functional outcomes after first-ever ischemic stroke.

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454