"Several limitations should be considered. Our participants were from a single center and of white ethnicity only, which may limit how applicable our findings are to other healthcare settings or populations. The small sample size of the study may limit the power to detect weaker associations with insulin requirements. We did not have weight measurements available at the time of insulin start. Further work could explore whether this weight may be a better predictor of insulin dose requirement; however, given the very limited correlations we have seen, we believe this is unlikely to provide a substantial improvement over pre-pregnancy weight."