Skip to main content

Journal of Population Sciences

Table 3 Mediation effect of maternal smoking on birth outcomes by cohort

From: Inequalities at birth: stable socioeconomic differences in birth outcomes in three British cohorts

 

NCDS

(1958)

B70

(1970)

MCS

(2001)

 

Mediation effect

(CI)

Mediation effect

(CI)

Mediation effect

(CI)

Panel A

   

Y = Birth Weight (g)

   

Maternal behavior (M)

   

Smoking in pregnancy

− 19.657

(− 23.322, − 16.4)

− 30.605

(− 35.02, − 26.556)

− 39.886

(− 45.68, − 33.822)

% Mediation

35.9%

65.6%

60.5%

Panel B

   

Y = LBW

   

Maternal behavior (M)

   

Smoking in pregnancy

0.004

(0.002, 0.005)

0.007

(0.007, 0.005)

0.009

(0.007, 0.011)

% Mediation

16.4%

46%

46.2%

N

15,500

15,552

14,619

  1. Panel A shows the mediation effect of maternal smoking in pregnancy on birth weight measured in grams. Panel B shows the mediation effect of maternal smoking in pregnancy on low birth weight. All the models reported include the following control variables: child’s sex; maternal age; maternal marital status at birth; birth order; timing of prenatal care access; maternal height (standardized by cohort). The control variables are defined, as shown in Table 1
  2. All the reported mediation effects are statistically significant