AUTHOR=Foster Emma J. , Barlas Raphae S. , Bettencourt-Silva Joao H. , Clark Allan B. , Metcalf Anthony K. , Bowles Kristian M. , Potter John F. , Myint Phyo K. 

TITLE=Long-Term Factors Associated With Falls and Fractures Poststroke

JOURNAL=Frontiers in Neurology

VOLUME=Volume 9 - 2018

YEAR=2018

URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2018.00210

DOI=10.3389/fneur.2018.00210

ISSN=1664-2295

ABSTRACT=Background: Risk factors for post-stroke falls and fractures remain poorly understood. This study aimed to evaluate which factors increased risk of these events after stroke.
Methods: Data from 7,267 hospitalized stroke patients were acquired from the Norfolk and Norwich University Hospital Stroke Register from 2003–2015. The impacts of multiple patient level and stroke characteristics and comorbidities on post-discharge falls and fractures were assessed. Univariate and multivariable models were constructed, adjusting for multiple confounders, using binary logistic regression for short-term analysis (up to 1-year post-discharge) and Cox-proportional hazard models for longer-term analysis (1-3, 3-5 years and 0-10 year follow-up). 
Results: The mean age (SD) was 76.3±12.1 years at baseline. 1,138 (15.7%) participants had an incident fall; and 666 (9.2%) an incident fracture during the 10-year follow-up (total person years = 64,447.99 for falls and 67,726.70 for fractures). Half the sample population were female (50.6%) and the majority had an ischemic stroke (89.8%). After adjusting for confounders: age, sex, previous history of falls and atrial fibrillation were associated with an increased risk of both falls and fractures during follow-up. Furthermore, chronic kidney disease and hyperlipidemia were associated with an increased risk of falls, while previous stroke/TIA increased fracture risk. Total anterior circulation stroke and a prestroke modified Rankin Scale score of 3-5 were associated with decreased risk of both events, with hypertension and cancer decreasing risk of falls only. 
Conclusion: We identified demographic, stroke-related and comorbid factors associated with post-stroke falls and fracture incidence. Futher studies are required to examine and establish the relationship between reversible factors and further explore the role of preventative measures to prevent post-stroke falls and fractures.