Abstract
PurposeDisability is a potential risk for stroke survivors. This study aims to identify disability risk factors associated with stroke and their relative importance and relationships from a national behavioral risk factor dataset.MethodsData of post-stroke individuals in the U.S (n=19,603) including 397 variables were extracted from a publically available national dataset and analyzed. Data mining algorithms including C4.5 and linear regression with M5s methods were applied to build association models for post-stroke disability using Weka software. The relative importance and relationship of 70 variables associated with disability were presented in infographics for clinicians to understand easily.ResultsFifty-five percent of post-stroke patients experience disability. Exercise, employment and satisfaction of life were relatively important factors associated with disability among stroke patients. Modifiable behavior factors strongly associated with disability include exercise (OR: 0.46, P<0.01) and good rest (OR 0.37, P<0.01).ConclusionsData mining is promising to discover factors associated with post-stroke disability from a large population dataset. The findings can be potentially valuable for establishing the priorities for clinicians and researchers and for stroke patient education. The methods may generalize to other health conditions.
Highlights
In the United States, seven million stroke patients live beyond the acute stroke phase with significant disability and impairment [1]
Employment and satisfaction of life were relatively important factors associated with disability among stroke patients
Data mining is promising to discover factors associated with post-stroke disability from a large population dataset
Summary
In the United States, seven million stroke patients live beyond the acute stroke phase with significant disability and impairment [1]. Most have varying levels of disability [2]. Information provided by clinicians to patients and their families is often focused on etiology or pathophysiological facts such as the size and location of brain lesions [3]. Due to the sudden onset of the disease, stroke patients are often uncertain about their long-term prognosis. Predictors related to long-term stroke outcomes, such as early rehabilitation, smoking, drinking, early stroke recognition, and social support, have been rarely communicated to clinicians [2,4,5,6]. Patient-level phenotypes are vital for designing personalized stroke management after an initial incident but are often underused
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
