Background: The CONCEPTT randomized trial of CGM in pregnant women with type 1 diabetes showed significant improvement in pregnancy outcomes, but no difference in mean CGM glucose between those allocated to CGM vs. capillary glucose monitoring. Whilst summary statistics are recommended for the reporting of CGM data, they do not give dynamic information about the glucose profile across 24 hours.

Aim: To determine, using novel functional data analysis (FDA), if differences in the temporal glucose profile occurred in women randomized to CGM.

Methods: CGM data was available from 200 pregnant women (100 CGM intervention; 100 control) in the CONCEPTT study, at baseline, 24 and 34 weeks gestation. FDA was applied to the glucose data to generate functional glucose curves and multivariable statistical analysis applied.

Results: Despite having the same mean CGM glucose as those on capillary glucose monitoring alone 121 vs. 126 mg/dL; 6.7 vs. 7.0 mmol/l p=0.14, FDA revealed that pregnant CGM users were in fact running a significantly lower glucose (by 9-14mg/dL; 0.5-0.8 mmol/l) for a total of 8 hours a day (08.30-12.30 and 15.00-19.00).

Conclusion: Analysis of temporal glucose profiles gives key information about glucose control, missed by commonly reported CGM metrics. The women allocated to CGM ran a significantly lower glucose for 33% of the day.

E.M. Scott: Advisory Panel; Self; Abbott. Speaker's Bureau; Self; Abbott, Eli Lilly and Company. H.R. Murphy: Advisory Panel; Self; Medtronic MiniMed, Inc. D. Feig: Advisory Panel; Self; Medtronic. Speaker's Bureau; Self; Medtronic. G. Law: None.


JDRF; Canadian Clinical Trials Network; National Institute for Health Research

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at