. 2019 Dec 3;11(12):2934. doi: 10.3390/nu11122934

Table 4.

zCES-D_(mean), zDASH_(mean) total score and zBMI_(v3) at second follow-up: mediation model relaxing the assumption of no interaction between exposure and mediator and stratifying by sex, race and poverty status, HANDLS 2004–2018 ¹.

	Controlled Direct Effect			Natural Direct Effect			Natural Indirect Effect			Marginal Total Effect
	β	(SEE)	P_wald	β	(SEE)	P_wald	β	(SEE)	P_wald	β	(SEE)	P_wald
CES-D→DASH→BMI ²
Overall sample, n = 1201	−0.007	0.030	0.82	−0.007	0.030	0.82	+0.003	0.005	0.59	−0.004	0.030	0.89
Men, n = 469	−0.060	0.050	0.22	−0.060	0.050	0.18	+0.001	0.003	0.76	−0.060	0.050	0.19
Women, n = 732	+0.030	0.040	0.46	+0.024	0.042	0.57	+0.003	0.008	0.72	+0.027	0.042	0.53
Whites, n = 587	−0.010	0.047	0.82	−0.030	0.046	0.52	+0.009	0.012	0.44	−0.020	0.047	0.67
African Americans, n = 714	+0.017	0.042	0.68	+0.017	0.042	0.69	+0.001	0.002	0.74	+0.017	0.042	0.68
Below poverty, n = 483	+0.013	0.047	0.79	+0.018	0.047	0.70	−0.003	0.008	0.66	+0.014	0.047	0.76
Above poverty, n = 718	−0.034	0.043	0.42	−0.035	0.043	0.41	+0.004	0.005	0.35	−0.031	0.043	0.47
DASH→CES-D→BMI ³
Overall sample, n = 1201	−0.14	0.03	<0.001	−0.14	0.03	<0.001	+0.001	0.001	0.63	−0.14	0.03	<0.001
Men, n = 469	−0.05	0.05	0.26	−0.06	0.05	0.22	+0.001	0.002	0.73	−0.06	0.05	0.23
Women, n = 732	−0.14	0.04	<0.001	−0.15	0.04	<0.001	+0.001	0.002	0.75	−0.15	0.04	<0.001
Whites, n = 587	−0.18	0.04	<0.001	−0.18	0.04	<0.001	+0.003	0.004	0.48	−0.18	0.04	<0.001
African Americans, n = 714	−0.03	0.04	0.43	−0.03	0.04	0.43	−0.001	0.002	0.76	−0.04	0.04	0.42
Below poverty, n = 483	−0.19	0.05	0.001	−0.19	0.05	<0.0001	−0.001	0.002	0.76	−0.20	0.05	<0.0001
Above poverty, n = 718	−0.07	0.04	0.071	−0.07	0.04	0.066	+0.001	0.002	0.48	−0.07	0.04	0.072

Abbreviations: DASH = Dietary Approaches to Stop Hypertension; HANDLS = Healthy Aging in Neighborhood of Diversity across the Lifespan; CES-D = The Center for Epidemiologic Studies Depression scale; OLS = Ordinary Least Square; SEE = Standard Error of the Estimate. ¹ Multivariate OLS models adjusted for baseline age, sex, race, poverty status, educational attainment, literacy, employment status, current smoking status, current drug use, self-rated health, mean of total energy intake and mean energy intake from grocery store. For CDE, zDASH_(mean) was set at a value of zero. Only results with significant total effects at type I error of 0.05 were presented. ² DASH diet as a mediator. ³ CES-D as a mediator.