DEVELOPING AND ASSESSING A TRANSPORTABILITY FRAMEWORK FOR CROSS-BORDER REAL-WORLD EVIDENCE GENERATION IN CANADA: A DEMONSTRATION USING US CLAIMS DATA FOR PATIENTS WITH MULTIPLE SCLEROSIS
Author(s)
Tuhin Maity, PhD, Saina Sehatkarlangrodi, PhD, Rayna Crawford, MS.
McKesson Canada, Mississauga, ON, Canada.
McKesson Canada, Mississauga, ON, Canada.
OBJECTIVES: Real‑world healthcare resource utilization and cost data for specific diseases remain limited in Canada due to fragmented healthcare systems, lack of data connectivity, and access barriers. Leveraging comprehensive, analysis-ready data from other jurisdictions offers a potential solution. However, use of other jurisdictional data in the absence of local data is constrained by the lack of a standardized and validated adaptation framework. This study aimed to develop and demonstrate a robust transportability and cost‑conversion framework to adapt US real‑world claims data for Canadian patients with multiple sclerosis (MS) and generate Canadian‑equivalent cost estimates.
METHODS: Adults with MS were identified in a large US administrative claims database (McKesson Compile; N = 151,920), comprising 12.9 million pharmacy and 55.5 million medical claims (2016-2025). The cohort was aligned to published Canadian reference targets for demographics, socioeconomic status, and comorbidity prevalence using two co-primary approaches: logit‑based calibration weighting and greedy distance‑minimization matching. A transparent cost‑conversion pipeline mapped US drug and procedure codes to Canadian formularies and fee schedules to generate Canadian‑equivalent unit prices, which were applied to patient‑level utilization.
RESULTS: Both approaches achieved strong alignment with Canadian targets across demographic, socioeconomic, and clinical dimensions. Estimated annual direct healthcare costs were CAD 18,970 per person‑year (weighting) and CAD 26,397 (matching). These estimates are aligned with published Canadian estimates (CAD 20,763), supporting the validity and robustness of the methods.
CONCLUSIONS: Although local data remains the preferred source for real‑world evidence generation, access to comprehensive, analysis‑ready datasets in Canada is limited. This study demonstrates that cross‑border claims data can be systematically adapted to generate Canadian‑relevant cost estimates using an integrated, transparent framework. While the overall results were aligned with Canadian benchmarks, some differences were observed. Further investigation to help refine the Canadianization approach will strengthen the robustness of future real‑world evidence generation using cross‑border data allowing for broader use.
METHODS: Adults with MS were identified in a large US administrative claims database (McKesson Compile; N = 151,920), comprising 12.9 million pharmacy and 55.5 million medical claims (2016-2025). The cohort was aligned to published Canadian reference targets for demographics, socioeconomic status, and comorbidity prevalence using two co-primary approaches: logit‑based calibration weighting and greedy distance‑minimization matching. A transparent cost‑conversion pipeline mapped US drug and procedure codes to Canadian formularies and fee schedules to generate Canadian‑equivalent unit prices, which were applied to patient‑level utilization.
RESULTS: Both approaches achieved strong alignment with Canadian targets across demographic, socioeconomic, and clinical dimensions. Estimated annual direct healthcare costs were CAD 18,970 per person‑year (weighting) and CAD 26,397 (matching). These estimates are aligned with published Canadian estimates (CAD 20,763), supporting the validity and robustness of the methods.
CONCLUSIONS: Although local data remains the preferred source for real‑world evidence generation, access to comprehensive, analysis‑ready datasets in Canada is limited. This study demonstrates that cross‑border claims data can be systematically adapted to generate Canadian‑relevant cost estimates using an integrated, transparent framework. While the overall results were aligned with Canadian benchmarks, some differences were observed. Further investigation to help refine the Canadianization approach will strengthen the robustness of future real‑world evidence generation using cross‑border data allowing for broader use.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
EE276
Topic
Economic Evaluation
Topic Subcategory
Cost/Cost of Illness/Resource Use Studies
Disease
Systemic Disorders/Conditions (Anesthesia, Auto-Immune Disorders (n.e.c.), Hematological Disorders (non-oncologic), Pain)