Deriving a method to estimate incidence of stroke in Ireland

This paper assesses the validity of estimating stroke incidence employing discharge data from the Hospital Inpatient Enquiry (HIPE) database, adjusted to reflect evidence from the North Dublin Population Stroke Study (NDPSS). This analysis compares contemporaneous HIPE stroke discharges among North Dublin residents to hospitalised stroke patients in the NDPSS and finds no significant difference between the datasets' measures of hospitalised stroke cases in five North Dublin hospitals. We find that in estimating incidence, HIPE discharges with principal and secondary stroke diagnoses should be included but patients admitted for rehabilitation should be excluded to avoid duplication. The HIPE-based estimate should be further adjusted to reflect the NDPSS evidence of non-hospitalised stroke incidents.