Skip to main content
. 2023 Sep 29;15:749–765. doi: 10.2147/NSS.S423331

Figure 1.

Figure 1

Directed acyclic graph of the MR framework investigating the causal relationship between OSA and HBP/T2DM. Instrumental variable assumptions: (1) the GIVs must be strongly associated with OSA; (2) the GIVs must not be associated with any potential confounder of OSA versus HBP/T2DM relationship; (3) the GIVs should only affect the risk of HBP/T2DM through OSA.

Abbreviations: GIVs, genetic instrument variables; SNPs, single-nucleotide polymorphisms; OSA, obstructive sleep apnea; HBP, high blood pressure; T2DM, type 2 diabetes mellitus; MR, Mendelian randomization.