Table 3

Differences in fat mass, android to gynoid ratio and lean mass (95% CI) at age 60–64 years between the hypothetical lowest and highest paternal educational attainment (slope index of inequality), with sequential adjustment for potential mediators

Men (n=479)Fat mass index (kg/m1.2)p ValueAndroid to gynoid fat mass ratiop ValueAppendicular lean mass index (kg/m2), adjusted for fat mass indexp Value
 1. Paternal education (6 years)0.93 (−0.25 to 2.11)0.128.50 (3.49 to 13.52)<0.001
 2. Model 1+birth weight0.93 (−0.25 to 2.12)0.128.31 (3.34 to 13.28)<0.001
 3. Model 1+weight gain from 0 to 7 years1.13 (−0.04 to 2.29)0.068.30 (3.31 to 13.29)<0.001
 4. Model 1+weight gain from 7 to 20 years0.72 (−0.39 to 1.82)0.217.87 (2.83 to 12.92)<0.001
 5. Model 1+own education and adult SEP0.24 (−1.04 to 1.51)0.725.55 (0.12 to 10.98)0.05
Women (n=515)
 1. Paternal education (6 years)4.22 (2.62 to 5.82)<0.00110.62 (6.69 to 14.55)<0.001−0.27 (−0.49 to −0.06)0.01
 2. Model 1+birth weight4.21 (2.61 to 5.82)<0.00110.52 (6.59 to 14.45)<0.001−0.26 (−0.48 to −0.05)0.02
 3. Model 1+weight gain from 0 to 7 years4.24 (2.64 to 5.83)<0.00110.50 (6.57 to 14.42)<0.001−0.24 (−0.45 to −0.03)0.03
 4. Model 1+weight gain from 7 to 20 years3.58 (2.06 to 5.10)<0.0019.88 (5.95 to 13.80)<0.001−0.26 (−0.47 to −0.05)0.01
 5. Model 1+own education and adult SEP3.38 (1.54 to 5.23)<0.0019.77 (5.29 to 14.26)<0.001−0.12 (−0.36 to 0.13)0.35
  • Occupational class was that of the highest in the household and derived using the Registrar General's classification; analyses were restricted to those with valid measures for each indicator of SEP, relevant potential mediators and all body composition outcomes. Cells are blank where analyses were not included due to lack of evidence for association in univariable analyses (p>0.05).

  • SEP, socioeconomic position.