Grant: CIHR Fellowship

Stroke is one of the leading causes of disability and death in Canada. However, stroke affects men and women differently. Women experience greater disability compared to men after stroke. The reasons for these differences are unknown and are difficult to study in large populations. A new stroke outcome measure, 90-day home-time (the number of days a patient is living at home - and not in a hospital - in the first 90 days after their stroke), may make measuring stroke outcomes in large populations easier. However, 90-day home-time has properties which make it difficult to analyze and the optimal method for analyzing 90-day home-time is unknown. This research will use several methods from the machine learning literature to determine which is most appropriate for use with 90-day home-time. Once the most appropriate method is determined it will be used to help identify risk factors for poor outcome after stroke and if these are different between men and women. For example, perhaps smoking men compared to non-smoking men spend an average of 3 fewer days at home in the first 90 after their stroke while smoking women compared to non-smoking women spend an average of 6 fewer days at home in the first 90 after their stroke. This type of information will help inform the public on specific risk factors for poor outcome after stroke and which of these risk factors are different in men and women. For physicians and health care administrators, predicting the amount days a patient is at home in the first 90 days after their stroke helps plan for the amount of days a patient is in hospital in the first 90 days after their stroke. Knowing when 90-day home-time may be different among men and women will help plan for future hospital lengths of stay and how these lengths of stay may change if certain risk factors could be eliminated from the population.