Comparing approaches for deriving diabetes care cascades to inform policy: a cross-sectional analysis using national data from 88 countries

Résumé fourni par la source

Objective: Care cascade indicators are widely used to monitor national diabetes control efforts. However, diabetes definitions used to derive care cascades vary across studies, which may markedly affect results and subsequent policy decisions. Here, we examine the magnitude of resultant differences between approaches. Research Design and Methods: We analyzed nationally representative, cross-sectional data of 800,348 individuals aged ≥25 years from 88 countries in 2008-2021. We employed two different diabetes definitions: elevated biomarkers (HbA1c≥6.5%; fasting plasma glucose ≥7.0mmol/L; or random plasma glucose ≥11.1mmol/L) or self-reported diagnosis (“diagnosis-based definition”) versus elevated biomarkers or self-reported treatment (“treatment-based definition”). Care cascade estimates included (1) proportions of diabetes cases who were diagnosed, and proportions of diagnosed diabetes cases who (2) received treatment and (3) attained glycemic control. We benchmarked results against World Health Organization (WHO) diabetes targets. Results: Diabetes prevalence was 12.9% (95%-CI: 12.1-13.8) applying the diagnosis-based definition and 11.3% (95%-CI: 10.5-12.1) with the treatment-based definition. Using the diagnosis-based rather than treatment-based diabetes definition to derive care cascades consistently increased percentages who attain diagnosis and control stages but decreased percentages receiving treatment. Across countries, median differences between approaches were 11.3% (IQR: 5.1-24.7) for diabetes diagnosis, 21.6% (IQR: 14.5-37.8) for treatment, and 16.4% (IQR: 7.8-26.8) for control. The WHO 80%-glycemic control target was met by 22% versus 6% of countries when using the diagnosis-based versus treatment-based definition, respectively. Conclusions: Care cascade estimates diverged substantially and consistently across diabetes definitions, skewing policy implications in predictable ways. Harmonizing diabetes performance metrics may improve decision-making and facilitate cross-country comparisons.