Supplementary Material for: Risk of first-ever stroke in the 90 days after hospital encounters: A case- crossover study

Background and Aim
Hospital encounters preceding stroke appear to be common, but the reasons for these encounters are largely unexplored. We investigated the association between hospital encounters and the risk of first-ever stroke within 90 days.
Methods
A case-crossover study was undertaken in Tasmania, Australia (2007-2020). First-ever strokes were identified in linked data using International Classification of Diseases, Australian Modification (ICD-10-AM) codes. A 90-day lookback period was used to identify prior hospital encounters comprising emergency department [ED] presentations and hospital admissions. Conditional logistic regression was used to derive odds ratios (OR) and 95% confidence intervals (95% CI) for associations between prior hospital encounters and first-ever stroke in the 90 days before stroke (case period) vs. 1-year corresponding control period before stroke. Most common groups of diagnoses preceding in the 90 days before stroke were described.
Results
Among 4,907 first-ever stroke patients (78% aged ≥65 years; male 52%), 1,240 (25%) had a hospital encounter in the 90 days before stroke (21% ED presentation and/or 17% hospital admission). Hospital encounters were associated with more than two-fold greater stroke risk within 90 days (OR 2.6 95% CI 2.3-2.9), than the control period. The association was strongest in the 7 days immediately before stroke; ED presentations (OR 8.5 95% CI 6.1-11.8) exhibited a stronger association with odds of stroke than hospital admissions (OR 4.2 95% CI 3.1-5.8). In the 90 days before stroke, the most common diagnoses for hospital encounters included non-specific symptoms and signs, circulatory system disorders and injuries.
Conclusion
Hospital encounters are associated with an increased risk of stroke within 90 days. These may present opportunities to reduce misdiagnosis or target interventions for the primary prevention of stroke.