Reassessment of the new diagnostic thresholds for gestational diabetes mellitus: an opportunity for improvement

In reply: Callaway and McElduff are dismissive of the concerns I raised in my recent article,1 asserting incorrectly that they are old arguments. The potential reduction of the risk of macrosomia (birthweight > 90th centile; large for gestational age babies) when one or more blood glucose levels (BGLs) on an oral glucose tolerance test are normal was only suggested recently, in February 2014.2 In response, new data3 confirmed the statistical flaw in the new diagnostic criteria for gestational diabetes mellitus (GDM). These data showed that (i) nearly 50% of women having only one elevated BGL test result do not reach the diagnostic risk threshold, and (ii) women having two or more BGL results just below the new diagnostic levels may be at greater risk, yet will not be identified.1

The many international organisations mentioned by Callaway and McElduff were early adopters of the new criteria, and the impact of this interaction was not considered. Australia has an opportunity to develop a better, statistically valid, diagnostic approach. The rate of GDM and its management can then be benchmarked against other countries that have adopted the new statistically flawed diagnostic criteria. The benefit of this approach cannot be understated.

Zheng and colleagues question the statement that it is a concern that women with GDM may be charged a higher insurance premium. Their hypothetical applicant had been diagnosed with GDM 10 years previously. The Australian Diabetes in Pregnancy Society states that the risk of developing diabetes is up to 50%.4 Most cases occur within 10 years and the risk appears to plateau after that time.5 Thus, it would appear that the insurance actuaries have read carefully the medical literature about the conversion to type 2 diabetes with minimal additional risk after 10 years. The industry response may be different if the application for insurance was made 3 months after delivery. This may explain why many of my patients and a colleague who had GDM were recently quoted higher premiums when applying for life insurance soon after the delivery of their children. The hypothetical case does not invalidate the real concerns expressed in the article.

My article was factual, not alarmist. It suggests that the available data can be used to establish the diagnosis more accurately. Improving the identification of women who truly have an odds ratio for the threshold level for risk of 1.75 or greater will still result in more women being diagnosed with GDM than currently. However, these women will belong to a higher risk group and their management should result in even greater improvements in obstetric and neonatal outcomes.