A New Validated Tool Helps Predict Lifetime Health Outcomes for Prediabetes and Type 2 Diabetes in Chinese Populations

Closeup on hands holding a glucometer
Source : Stanford University Medical Center

Diabetes is one of the most rapidly expanding health problems of the twenty-first century. The Asia-Pacific area is on the frontlines of the pandemic increase in the number of individuals who have diabetes, which has occurred in recent years. China, in particular, has by far the highest absolute burden of diabetes, with an estimated 116 million people living with the illness, accounting for one-quarter of all diabetic patients worldwide, according to the International Diabetes Federation. By 2045, it is projected that the number of people living with diabetes in the nation would have increased to 147 million, excluding the huge diaspora population that China has created throughout the globe.

It is critical for the development of health policy that we evaluate the health and economic consequences of diabetes and its complications. Because type 2 diabetes predictors have been established and validated in historical European populations, they may not be relevant to East Asian people, who have a different epidemiology and set of complications from European ones. Furthermore, the current models are usually restricted to diabetes alone and do not take into account the transition from prediabetes to type 2 diabetes. The absence of a suitable simulation model for East Asian people with prediabetes is a significant limitation in the economic assessment of health treatments in this region.

These constraints are being addressed through new joint research. Karen Eggleston, Director of the Asia Health Policy Program at the Asia Pacific Study Center, is a member of the research team. Using a patient-level simulation model, the researchers were able to predict the lifelong health consequences of prediabetes and type 2 diabetes in East Asian populations. The model was built and verified. They published their results in the peer-reviewed journal PLOS Medicine.

Because of the long-term nature of health outcomes and the time lag between treatments and changes in patient outcomes, the chronic progression to diabetes-related problems is an excellent candidate for computer simulation modeling. Although it is possible to assess the effects of health interventions on East Asian populations with diabetes using current models, doing so is difficult since the models were created and validated in European and North American populations with quite different epidemiology and outcomes.

Eggleston and her colleagues set out to close this gap by developing and validating an outcomes model for the development of diabetes and associated comorbidities in Chinese communities. A new model developed in China and Hong Kong, called the Chinese Hong Kong Integrated Modeling and Evaluation (CHIME), was compared to two widely used existing models developed and validated in two countries: the United Kingdom (the United Kingdom Prospective Diabetes Study Outcomes Model 2, or UKPDS-OM2) and the United States/Canada (known as the United States/Canada Prospective Diabetes Study Outcomes Model, or US/Canada) (called Risk Equations for Complications of type 2 Diabetes, or RECODe). Contrary to the fact that risks are continuous over a range of risk factor values, the two current models fail to account for the transition from prediabetes to diabetes.

The CHIME combines prediabetes and diabetes into a complete model with 13 outcomes that is used to diagnose and treat patients. Mortality, micro- and macrovascular problems, and the development of diabetes are all risks associated with smoking. A population-based cohort of 97,628 individuals in Hong Kong with type 2 diabetes (43.5 percent) or prediabetes (56.5 percent) from 2006 to 2017 was used to create the CHIME simulation model, according to the researchers. This group of technologies, collectively known as the Hong Kong Clinical Management System (CMS), is one of the biggest Chinese electronic health information systems, capable of storing comprehensive clinical data.

In the next phase, the CHIME model was tested against individual-level data from the China Health and Retirement Longitudinal Study (CHARLS) cohort (2011-2018), which was a nationally representative longitudinal cohort of middle-aged and elderly Chinese citizens aged 45 and older. With simulated cohorts of 100,000 people, the researchers tested their CHIME model against six outcomes measures reported in the CHARLS data as well as an additional 80 endpoints from nine previously published studies of diabetes patients.


Categories: Clinical