Using non-parametric Bayes shrinkage to assess relationships between multiple environmental and social stressors and neonatal size and body composition in the Healthy Start cohort

Martenies, Sheena E.; Hoskovec, Lauren; Wilson, Ander; Moore, Brianna F.; Starling, Anne P.; Allshouse, William B.; Adgate, John L.; Dabelea, Dana; Magzamen, Sheryl

doi:10.1186/s12940-022-00934-z

Environmental Health

Table 3 Posterior means, 95% credible intervals, and posterior inclusion probabilities (PIPs) for each census tract level exposure regression coefficient in the NPB model, adjusted for all individual-level covariates. Results are presented for a 1 SD increase in each exposure

From: Using non-parametric Bayes shrinkage to assess relationships between multiple environmental and social stressors and neonatal size and body composition in the Healthy Start cohort

Variable	SD^c	Birth Weight (g)		Adiposity (% Fat Mass)
Variable	SD^c	Posterior Mean^a (95% CI)	PIP	Posterior Mean^a,b (95% CI)	PIP
Environmental Exposures
Mean PM_2.5 (μg/m³)	0.6	-2.4 (-29.3, 8.5)	0.31	-0.01 (-0.16, 0.11)	0.32
Mean O₃ (ppb)	3.1	-6.4 (-65.1, 6.9)	0.37	-0.02 (-0.24, 0.09)	0.36
Mean temperature (°F)	4.8	13.1 (-18.6, 189.3)	0.38	-0.01 (-0.20, 0.12)	0.32
Mean O₃ × Mean temperature		-162.0 (-242.3, -117.8)	1.00	-0.01 (-0.09, 0.00)	0.04
Tree cover (%)	3.1	-0.4 (-12.1, 11.6)	0.25	0.00 (-0.15, 0.11)	0.28
Impervious surfaces (%)	13.3	-0.7 (-14.4, 7.4)	0.25	-0.01 (-0.15, 0.09)	0.29
AADT (vehicles per day-km²)	8203	0.7 (-9, 18.2)	0.24	0.01 (-0.08, 0.21)	0.28
Distance to TRI sites (km)	2.6	-1.1 (-18.9, 8.9)	0.27	-0.01 (-0.17, 0.11)	0.32
Distance to NPL sites (km)	3.3	0.0 (-11.7, 14.6)	0.26	0.00 (-0.13, 0.14)	0.30
Distance to waste sites (km)	2.3	3.0 (-7.4, 39.2)	0.31	0.03 (-0.07, 0.34)	0.33
Distance to major emitters (km)	3.2	0.0 (-10.7, 13.1)	0.24	0.01 (-0.09, 0.19)	0.29
Distance to CAFOs (km)	6.8	-2.1 (-31, 14.2)	0.34	-0.02 (-0.24, 0.14)	0.35
Distance to mines or wells (km)	2.1	-1.7 (-22.6, 8.9)	0.31	-0.03 (-0.27, 0.07)	0.38
Social Exposures
CVD hospitalizations (n per 10,000)	45.2	-1.4 (-18.4, 5.9)	0.28	-0.02 (-0.22, 0.08)	0.35
Resp. hospitalizations (n per 10,000)	33.0	-2.5 (-26.8, 5.0)	0.30	-0.02 (-0.21, 0.05)	0.35
Violent crimes (n per 1,000)	6.3	-0.3 (-12.2, 9.2)	0.25	-0.01 (-0.16, 0.09)	0.32
Property crimes (n per 1,000)	36.0	-1.5 (-18.3, 5.2)	0.28	-0.06 (-0.35, 0.03)	0.49
Less than HS diploma (%)	12.7	-1.1 (-17.9, 10.6)	0.29	-0.01 (-0.20, 0.09)	0.32
Unemployment (%)	5.0	-6.6 (-51.3, 2.5)	0.40	-0.03 (-0.30, 0.05)	0.37
Households speaking limited English (%)	10.9	-1 (-16.2, 6.9)	0.25	0.00 (-0.12, 0.19)	0.31
Households in poverty (%)	8.3	-1.1 (-15, 6.6)	0.26	-0.02 (-0.22, 0.06)	0.33
Persons of color (%)	22.9	-0.5 (-13.1, 12.6)	0.26	0.00 (-0.13, 0.15)	0.29

^a Models are adjusted for: maternal race/ethnicity, maternal educational attainment, maternal pre-pregnancy BMI, maternal age at delivery, maternal smoking during pregnancy, second-hand smoke exposure during pregnancy, mean perceived stress scale score across pregnancy, mean postnatal depression score across pregnancy, season of conception, year of conception, longitude, latitude, and the interaction between longitude and latitude
^b Models of adiposity are also adjusted for the number of days between delivery and PEA POD measurements
^c Effect estimates are reported for a 1 standard deviation increase in each exposure variable

Back to article page

ISSN: 1476-069X

Contact us

General enquiries: journalsubmissions@springernature.com