Erroneous inflation of diabetes prevalence: Are there global implications?

Abstract

Comparison of the prevalence of type 2 diabetes mellitus (T2DM) in adults aged 25–64 years in selected Pacific Island countries using whole blood and plasma glucose cut‐off points. Unit records of STEPwise approach to Surveillance (STEPS) surveys obtained from Fiji, Samoa, and Tonga Ministries of Health; T2DM prevalence recalculated using whole blood and plasma cut‐off points. Shaded bars indicate T2DM prevalence based on correct glucose cut‐off points for the glucose meter used (fasting blood glucose [FBG] ≥6.1 mmol/L for early surveys1,3,5; fasting plasma glucose [FPG] ≥7.0 mmol/L for later surveys),2,4,6 whereas open bars show T2DM prevalence based on incorrect glucose cut‐off points (FPG ≥6.1 mmol/L for later surveys).2,4,6

Age and cohort rise in diabetes prevalence among older Australian women: Case ascertainment using survey and healthcare administrative data

“…records of medication and other health service use or hospital use) [14]. However, there are a number of limitations to these data which may contribute to unreliable case ascertainment and biased prevalence estimation [15][16][17].…”

Questioning glucose measurements used in the International Diabetes Federation (IDF) Atlas

“…Although fasting whole blood glucose concentrations are approximately 0.5 mmol/L lower than capillary glucose, with the difference increasing to nearly 1.5 mmol/L after a glucose load, 8 Taylor et al note that the same glucose cut-off points were applied for diagnosis of diabetes in the earlier and the more recent surveys, leading to the potential that type 2 diabetes prevalences are "erroneously inflated to approximately double the actual [type 2 diabetes] prevalence had the correct glucose cut-off point been applied". 7 Certainly, these questions are appropriate, and a review of the IDF figures would be welcome. At the same time, we must recognize the tremendous difficulty inherent in combining statistics from a wide variety of sources and, equally, must underscore the importance of the efforts of the IDF and other groups in gaining perspective on the diabetes epidemic, with its growing magnitude and ominous implications for global health.…”

Liberating non-communicable disease data

“…Low-income and middle-income countries (LMICs) lack good data, simply because of poor data repository systems, which make timely data unavailable for initiation of evidence-informed actions. 3,4 The health information system in Zambia, 5 for example, routinely collects information about HIV, but lacks systems for monitoring of the incidence and prevalence of NCDs. Thus, the focus should be on building the capacity of national health systems to develop surveillance systems that continuously monitor disease trends in the population.…”

“…Failure to recognise this problem, and application of the criterion for diabetes diagnosis in fasting capillary whole blood of 6•1 mmol/L or greater instead of 7•0 mmol/L or greater for plasma values has resulted in some reports of diabetes prevalence that are almost twice as high as they actually are. 4 In any event, diabetes prevalence, even when based on appropriate measurement of fasting plasma glucose, underestimates the true prevalence by around 30%. 5 Whether the NCD and Global Burden of Disease (GBD) risk-factor study collaborators, who have also used STEPS data in some of their reports, identifi ed and corrected for these shortcomings is unclear.…”