Table 4.
Mediation tests evaluating indirect effects of education on 10-year estimated CHD risk, stratified by gender. Point estimates (indirect effects= [(exp(β)-1)×100]) represent % change in relative 10-year risk for specific education category vs. referent education category, due to the addition of the mediator into the model. Numerical values in parentheses represent 95% confidence intervals.
| Women (n=207) | Men (n=139) | |||
|---|---|---|---|---|
|  | ||||
| <High School | ≥High School, <College | <High School | ≥High School, <College | |
| Poverty income ratio | −4.8 (−14.6, 1.0) | −2.8 (−9.4, 0.4) | −1.1 (−11.7, 5.4) | −0.8 (−7.8, 3.9) | 
| Body mass index | 15.4 (−0.4, 33.5) | 4.2 (−6.9, 13.1) | 0.3 (−1.2, 5.6) | 0.7 (−0.7, 5.4) | 
| Depressive symptoms | 11.8 (2.5, 26.6)* | 3.1 (0.4, 8.8)* | −0.8 (−8.7, 1.5) | 0.2 (−2.8, 0.9) | 
| Sense of Control | ||||
| Perceived mastery | 1.8 (−1.0, 10.4) | 2.1 (−0.3, 8.5) | −0.7 (−8.4, 1.9) | 0.9 (−6.7, 2.4) | 
| Perceived constraint | 6.7 (0.7,19.1)* | 4.7 (0.8,11.3)* | 0.8 (−1.3, 7.7) | 0.2 (−1.1, 3.5) | 
| Time preference | 1.8 (−1.6, 10.0) | 1.3 (−0.7, 6.3) | −0.6 (−7.7, 2.0) | 0.0 (−2.0, 2.7) | 
| Literacy | ||||
| Aural | −0.8 (−6.0, 2.6) | −0.3 (−4.1, 1.0) | 0.0 (−3.0, 3.8) | −0.0 (−2.9, 2.0) | 
| Oral | −0.6 (−6.4, 2.1) | −0.1 (−3.7, 2.4) | −0.8 (−9.3, 7.5) | −0.1 (−7.8, 6.7) | 
| Numeracy | 5.4 (−2.1, 15.3) | 2.8 (−1.1, 8.3) | 5.4 (−1.4,18.6) | 6.0 (−2.8,17.1) | 
| Reading comprehension | 7.2 (0.7, 19.4)* | 0.3 (−4.5, 1.0) | 7.2 (0.8,19.1)* | 5.1 (0.5,13.9) | 
All models are adjusted for age, race/ethnicity, and childhood confounders (parental socioeconomic index, age 7 intelligence, childhood chronic illness, childhood introversion, age 7 body mass index).
p<0.05 for group vs. referent (≥College) point estimate (β=0).