Table 3.

Recommendations to improve estimation and interpretation of diabetes burdens

Category	Benefits
Data infrastructure
Enhance vital statistics coverage	Decreases biases related to where deaths occur
Enable linkage between community, healthcare and vital registration data systems	Increases validity of reported events
Denominators and numerators
Validate consensus definitions for diabetes and disseminate	Improves understanding of screening and diagnostic practices and their influence on rates
Expand and standardise routine collection and use of data from healthcare settings	Adds more credible biochemical and clinician-coded indicators to self-reported data
Develop classification structure for diabetes complications (traditional, emerging and other comorbidities)	Elevates importance of non-fatal and non-traditional diabetes complications that affect quality of life
Establish denominators based on standardised definitions	Permits credible comparison of trends within and across countries
Analysis
Expand capacity to manage data and analyses (especially in LMICs)	Improves surveillance of diabetes burdens and impacts of policies and programmes
Reporting
Standardise chronic disease surveillance reporting through checklists that recommend providing critical contextual information regarding case definitions and how these are operationalised in the data	Helps analysts and users of data to contextualise and compare the findings