Attributable Costs of Stroke in Ontario, Canada and Their Variation by Stroke Type and Social Determinants of Health

Abstract

BACKGROUND: Estimates of attributable costs of stroke are scarce, as most prior studies do not account for the baseline health care costs in people at risk of stroke. We estimated the attributable costs of stroke in a universal health care setting and their variation across stroke types and several social determinants of health. METHODS: We undertook a population-based administrative database-derived matched retrospective cohort study in Ontario, Canada. Community-dwelling adults aged ≥40 years with a stroke between 2003 and 2018 were matched (1:1) on demographics and comorbidities with controls without stroke. Using a difference-in-differences approach, we estimated the mean 1-year direct health care costs attributable to stroke from a public health care payer perspective, accounting for censoring with a weighted available sample estimator. We described health sector–specific costs and reported variation across stroke type and social determinants of health. RESULTS: The mean 1-year attributable costs of stroke were Canadian dollars 33 522 (95% CI, $33 231–$33 813), with higher costs for intracerebral hemorrhage ($40 244; $39 193–$41 294) than ischemic stroke ($32 547; $32 252–$32 843). Most of these costs were incurred in acute care hospitals ($15 693) and rehabilitation facilities ($7215). Compared with all patients with stroke, the mean attributable costs were higher among immigrants ($40 554; $39 316–$41 793), those aged <65 years ($35 175; $34 533–$35 818), and those residing in low-income neighborhoods ($34 687; $34 054–$35 320) and lower among rural residents ($29 047; $28 362–$29 731). CONCLUSIONS: Our findings of high attributable costs of stroke, especially in immigrants, younger patients, and residents of low-income neighborhoods, can be used to evaluate potential health care cost savings associated with different primary stroke prevention strategies.

Advances in the Understanding of Social Determinants of Health in Stroke

“…Studies utilizing administrative data in Canada found that stroke incidence followed a J-shaped curve between the proportion of life spent in Canada and stroke incidence and disability; it was the highest among immigrants who had either immigrated at an early age or immigrated recently 7 ; and that the attributable cost of stroke among immigrants was significantly higher than among nonimmigrants. 8 Registry data in Denmark suggested that immigrants had greater prehospital delays compared with their nonimmigrant counterparts. 9 Furthermore, immigrants were less likely to receive acute reperfusion therapies even when adjusting for time to presentation and had lower odds of stroke unit admission, dysphagia screening, and early rehabilitation services than nonimmigrants.…”