Table 2.
Relation of prenatal alcohol exposure to cortical thickness in 3 regions of interest, controlling for potential confounding variables
| ROI | Occipitotemporal N = 78 (LOSO N = 78) |
Parietal N = 78 (LOSO N = 73) |
Occipital N = 78 (LOSO N = 70) |
|||
|---|---|---|---|---|---|---|
| r | β | r | β | r | β | |
| AA/day | −0.35** (−0.30**) | −0.25*,a (−0.22†,a) | −0.44*** (−0.24**) | −0.44*** (−0.22†,b) | −0.27* (0.02) | −0.22†,c (0.02) |
| AA/occasion | −0.48*** (−0.34**) | −0.41***,a (−0.27*,a) | −0.48*** (−0.22†) | −0.48*** (−0.19b) | −0.35** (0.06) | −0.32**,c (0.06) |
| Drinking Days/week | −0.29** (−0.24*) | −0.14a,b (−0.13a) | −0.36** (−0.16) | −0.35**,a,b (−0.12b) | −0.24* (0.09) | −0.20†,c (0.09) |
r represents the simple correlation between alcohol exposure and cortical thickness in the ROI.
β is the standardized regression coefficient, after adjustment for the potential confounding variables listed below. Control variables indicated with superscripts were retained in the regression only if their entry in the model altered the standardized regression coefficient associated with prenatal alcohol exposure by at least 10%.
Regression coefficients in parentheses are those obtained from the LOSO analysis before and after controlling for confounding variables. Ns differ where no cluster survived threshold in the appropriate anatomical region for one of the LOSO analyses and thickness for the left-out subject could not be obtained.
aMaternal education.
bSocioeconomic status.
cCigarettes/day.
†P < 0.10, *P < 0.05, **P < 0.01, ***P < 0.001.