Establishing causal relationships between insomnia and gestational diabetes mellitus using Mendelian randomization

Abstract

Background: Gestational diabetes mellitus (GDM) is a common condition observed globally, and previous studies have suggested a link between GDM and insomnia. The objective of this study was to elucidate the causative relationship between insomnia and GDM, and to investigate the influence of factors related to insomnia on GDM.

Methods: We performed bidirectional Mendelian randomization (MR) analyses using single nucleotide polymorphisms (SNPs) as genetic instruments for exposure and mediators, thereby minimizing bias due to confounding and reverse causation. The Cochran Q test was utilized for heterogeneity analysis, MR-Egger regression for pleiotropy assessment, and the leave-one-out method for evaluating the robustness of the results. Additionally, we determined the causal relationships between GDM and other factors such as coffee consumption, alcohol intake, and household income.

Results: Insomnia was positively associated with GDM, as indicated by 39 SNPs (OR = 1.27, 95 % CI 1.12-1.439, P-value = 0.008). Conversely, the MR analysis did not reveal any causal relationship between GDM and insomnia (OR = 1.032, 95 % CI 0.994-1.071, P-value = 0.99). Additionally, no causal relationship was observed between coffee consumption, alcohol intake, household income, and GDM (all P-values >0.05).

Conclusion: Our study indicates that insomnia elevates the risk of GDM, thereby establishing a causal link with GDM, independent of coffee consumption, alcohol intake, and household income.

Keywords: Causality; Genetic association; Gestational diabetes mellitus; Insomnia; Mendelian randomization.

Fig. 1

Mendelian randomization model of insomnia and risk of gestational diabetes. The design is under the assumption that the genetic variants are associated with insomnia, but not with confounders, and the genetic variants influence gestational diabetes only through insomnia. SNP indicates single nucleotide polymorphism.

Fig. 2

Odds ratios for the associations between genetically predicted GDM and risk of insomnia, alcohol intake, coffee intake and family income. SNPs: the number of SNPs used as instrumental variables; P: P-value of the causal estimate; OR: odds ratio; CI: confidence interval.