Skip to main content
. 2018 Oct 22;22(2):295–306. doi: 10.1017/S1368980018002112

Fig. 3.

Fig. 3

(colour online) The non-linear association meta-analysis between adjusted hazard ratios of all-cause mortality and sodium intake among studies using (a) dietary intake assessment and (b) urinary sodium excretion. Sodium intake was modelled with restricted cubic splines in a multivariate random-effects dose–response model (Inline graphic, linear model; Inline graphic, spline model; Inline graphic, 95% CI); the vertical axis is on a log scale