Table 2. Associations of Poor Family Functioning With Brain Morphologya.
| Mother-reported poor family functioning | Global brain measures | Specific brain volumetric measures | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total brain volume, cm3 | Cerebral white matter volume, cm3 | Total gray volume, cm3 | Amygdala volume, cm3 | Hippocampus volume, cm3 | |||||||||||
| B (95% CI) | P value | P FDR b | B (95% CI) | P value | P FDR | B (95% CI) | P value | P FDR | B (95% CI) | P value | P FDR | B (95% CI) | P value | P FDR | |
| Poor prenatal family functioning (FAD), per score | |||||||||||||||
| Model 1 | −26.8 (−34.6 to −18.9) | <.001 | .002 | −9.76 (−13.3 to −6.20) | <.001 | .002 | −16.7 (−21.3 to −12.2) | <.001 | .002 | −0.01 (−0.03 to 0.02) | .59 | .74 | −0.04 (−0.09 to 0.04) | .08 | .11 |
| Model 2 | −9.54 (−17.9 to −1.19) | .03 | .09 | −4.15 (−8.01 to −0.29) | .04 | .09 | −5.24 (−10.1 to −0.41) | .03 | .09 | −0.02 (−0.05 to 0.06) | .13 | .26 | −0.08 (−0.13 to −0.02) | .004 | .04 |
| Model 3 | −9.13 (−17.4 to −0.74) | .03 | .11 | −4.05 (−7.92 to −0.18) | .04 | .11 | −5.05 (−9.93 to −0.19) | .04 | .11 | −0.02 (−0.04 to 0.09) | .19 | .38 | −0.07 (−0.13 to −0.03) | .003 | .03 |
| Age 5 y poor family functioning (FAD), per score | |||||||||||||||
| Model 1 | −15.6 (−24.9 to −6.36) | .001 | .002 | −5.05 (−9.27 to −0.83) | .02 | .03 | −10.4 (−15.8 to −4.97) | <.001 | .002 | 0.01 (−0.02 to 0.03) | .943 | .94 | −0.01 (−0.06 to 0.05) | .87 | .94 |
| Model 2 | −5.47 (−14.7 to 3.81) | .25 | .35 | −1.86 (−6.14 to 2.45) | .40 | .50 | −3.52 (−8.90 to 1.86) | .20 | .57 | −0.01 (−0.03 to 0.03) | .82 | .82 | −0.02 (−0.07 to 0.04) | .53 | .59 |
| Model 3 | −4.18 (−14.1 to 4.50) | .31 | .44 | −1.47 (−5.80 to 2.83) | .50 | .59 | −2.91 (−8.31 to 2.49) | .29 | .44 | −0.01 (−0.03 to 0.03) | .93 | .93 | −0.02 (−0.08 to 0.04) | .53 | .59 |
| P homogeneity for age of FAD assessmentc | NA | .23 | NA | NA | .01 | NA | NA | .18 | NA | NA | .13 | NA | NA | .001 | NA |
Abbreviations: FAD, Family Assessment Device; FDR, false discovery rate; ICV, intracranial volume; MRI, magnetic resonance imaging; NA, not available.
Linear regression analysis of FAD and brain morphology outcome. B statistics are averaged from 10 imputed data sets. Model 1 is adjusted for child age at brain MRI scan, child sex, and total ICV. Model 2 is additionally adjusted for maternal age, race/ethnicity, education, marital status, parity, maternal psychopathology, and smoking and alcohol consumption. Model 3 is additionally adjusted for harsh parenting assessed when child was aged 3 years. Global brain measures are not adjusted for total ICV.
PFDR indicates the false discovery rate correction for multiple testing. Number of tests = 5 outcomes and the 2 relevant exposure periods (prenatal and early childhood). Critical value for FDR = 0.05.
P homogeneity for age of FAD assessment is a multiple-partial test used to test whether exposure from different time points of poor family functioning relates in the same manner to brain morphology outcome.