Skip to main content
. 2020 May 15;26:9. doi: 10.1186/s40885-020-00142-2

Fig. 1.

Fig. 1

Flowchart. For the cross-sectional analyses, data from the 2014 cycle (n = 1727) were used. For the longitudinal analyses, only participants in whom hypertension was not prevalent at baseline in 2014 were included (n = 419)