Use and Utility of Administrative Health Data for Stroke Research and Surveillance.

Despite declining age-standardized stroke incidence in high-income countries, stroke incidence is rising in lowand middle-income countries. Globally, the absolute burden of stroke was high in 2010 with 16.9 million first-ever strokes, 5.9 million stroke-related deaths, and 102 million disabilityadjusted life-years lost. These numbers are projected to increase. Surveillance provides an understanding of stroke frequency, burden, distribution, and determinants. These data are essential for monitoring trends over time, guiding judicious resource allocation, and for the design, implementation, and evaluation of interventions aimed at stroke prevention, treatment, and rehabilitation. Collecting data specifically for research purposes can be costly and time-consuming, limiting the sample size, period of follow-up, and geographical distribution of subjects. Surveillance requires continuous data collection in large geographic areas over years; therefore, attention has been paid to secondary use data. Health services utilization data, or administrative health data, provide a wealth of information for health services researchers and for stroke surveillance. However, the information collected and ascertainment methods are heterogeneous between countries and even between jurisdictions within a country, making the data vulnerable to selection and measurement bias. Comparing international data is also challenging. In this review, we discuss the strengths and weaknesses of administrative health data for stroke surveillance.