Research Modified Rankin scale as a determinant of direct medical costs after stroke Sarah Dewilde 1,2 , Lieven Annemans 2,3 , Andre Peeters 4 , Dimitri Hemelsoet 5 , Yves Vandermeeren 6 , Philippe Desfontaines 7 , Raf Brouns 8,9 , Geert Vanhooren 10 , Patrick Cras 11 , Boudewijn Michielsens 12 , Patricia Redondo 13 and Vincent Thijs 14 Abstract Background: Resource use in the acute and subacute phases after stroke depends on the degree of disability. Aims: To determine if direct costs after stroke also vary by level of disability as measured using the modified Rankin scale at the chronic stage after stroke. Methods: In a multicentre study, we collected acute and chronic in- and outpatient resource use in survivors of ischemic stroke stratified by levels of disability according to the modified Rankin Scale. Statistical inference on costs at each level of the modified Rankin Scale was estimated using a general linear model for the first three months, the first year, and any subsequent year after ischemic stroke. Results: A total of 569 survivors of ischemic stroke with a mean age of 71.7 years were enrolled (41% female) from 10 academic and nonacademic centers. Costs varied substantially over time and with each modified Rankin Scale level. The total average costs in the first year were estimated $33,147 per patient, ranging from $9,114 for modified Rankin Scale 0 to $83,236 for modified Rankin Scale 5. In the second year, medical costs were on average $14,039, varying from $2,921 to $39,723 for patients with modified Rankin Scale 0–5. The level of disability based on the modified Rankin Scale was a major determinant of resource use, irrespective of age, gender, atrial fibrillation, and vascular risk factors. Conclusion: Long-term resource use after stroke is high and is mainly driven by degree of disability as measured by the modified Rankin scale. Keywords Stroke, costs, cost analysis, modified Rankin Scale Received: 21 May 2016; accepted: 8 December 2016 Introduction Most survivors of stroke will face long-term conse- quences resulting in a large number of patients requiring help from a third party or to be institutiona- lized. The direct cost determinants of stroke vary over time: initial costs reflect expenditures due to 1 Department of Public Health, Faculty of Medicine, UGent, Belgium 2 Services in Health Economics, Brussels, Belgium 3 Interuniversity Centre for Health Economics Research UGent, VUB, Belgium 4Cliniques Universitaires SaintLuc, Service de Neurologie, Brussels, Belgium 5 Department of Neurology, Ghent University Hospital, Ghent, Belgium 6 Neurology Department, Universite ´ catholique de Louvain (UcL), Yvoir, Belgium 7 Centre Hospitalier Chre ´tien (CHC), Site Saint-Joseph, Lie `ge, Belgium 8 Department of Neurology, Universitair Ziekenhuis Brussel, Brussels, Belgium 9 Center for Neurosciences (C4N), Vrije Universiteit Brussel (VUB), Brussels, Belgium 10 Department of Neurology, AZ Sint-Jan Brugge-Oostende AV, Bruges, Belgium 11 Department of Neurology, Born Bunge Institute, University and University Hospital, Antwerp, Belgium 12 Heilig Hart Ziekenhuis, Lier, Belgium 13 CHU Ambroise Pare ´, Mons, Belgium 14 Austin Health, Department of Neurology and Florey Institute of Neuroscience and Mental Health, University of Melbourne, Heidelberg, Victoria, Australia Corresponding author: Sarah Dewilde, Department of Public Health, Faculty of Medicine, UGent, Belgium. Email: sd@she-consulting.be International Journal of Stroke, 0(0) International Journal of Stroke 0(0) 1–9 ! 2017 World Stroke Organization Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1747493017691984 journals.sagepub.com/home/wso