[Comparative analysis on type 2 diabetes incidence based on multisource data from China Kadoorie Biobank]

Area of Science:

• Epidemiology
• Public Health
• Endocrinology

Background:

• Type 2 diabetes is a growing public health concern in China.
• Accurate incidence estimation is crucial for effective prevention and management strategies.
• Previous studies have relied on single data sources, potentially leading to underestimation.

Purpose of the Study:

• To estimate the incidence of type 2 diabetes in Chinese adults.
• To evaluate the impact of different data sources on incidence estimation.
• To analyze the consistency of these differences across various population subgroups.

Main Methods:

• Utilized data from the second and third resurveys of the China Kadoorie Biobank (n > 23,000 each).
• Identified new type 2 diabetes cases through field surveys (questionnaire + blood glucose testing), self-reports, and healthcare database linkage.
• Calculated age-standardized incidence rates and used generalized linear mixed-effects models to compare data sources.

Main Results:

• Crude incidence rates varied significantly by data source, with field surveys yielding higher estimates (e.g., 8.4/1000 person-years in the second resurvey).
• Age-standardized incidence rates were consistent between resurveys (7.6 and 7.4/1000 person-years) using field survey data.
• The ratio comparing field survey to healthcare database linkage (IRR) decreased from 3.27 to 1.46, indicating improved case identification over time but still showing underestimation by databases in certain subgroups.

Conclusions:

• Integrating multi-source healthcare data substantially improves the identification of new type 2 diabetes cases in large cohorts.
• Relying on a single data source, particularly healthcare databases alone, can lead to significant underestimation of type 2 diabetes incidence.
• The findings highlight the importance of multi-modal data collection for accurate epidemiological surveillance of type 2 diabetes in China.