Table 5b

 Multiple linear regression for VIQ for all children

Variablen (mean; SD)Fp†β Coefficient‡95% CIp‡
*Significant factors in multiple regression model.
F test based on type III sums of squares and p values show significance of variable in model.
‡Only the adjusted coefficients for variables that reached significance are shown. p Value for adjusted β coefficient (compared to reference variable for categorical variables).
Overall model characteristics R2 = 0.288, adjusted R2 = 0.245, F = 6.709; p = 0.000, SD residuals = 0.971.
The continuous variables included were of normal distribution and the standardised residuals in all analysis were normally distributed. Folic acid and smoking were not included in the model as the univariate modelling did not suggest that that they were significant.
IGE, idiopathic generalised epilepsy; LD, learning difficulties; SES, socioeconomic status.
Maternal age230 (26.1; 4.8)0.7280.394
Maternal IQ (NART)*230 (98.7; 11.5)37.5110.0000.600.41 to 0.790.000
SESProfessional (reference) 470.9030.407
Intermediate 43
Manual or never worked 140
Family Hx LD in first degree relativeNone (reference) 1910.8330.363
1 or more 39
Seizures during pregnancyNone (reference) 1022.5990.053
None, convulsive 46−2.13−7.47 to 3.210.432
1–4 tonic-clonic 44−4.36−9.75 to 1.020.112
5 or more tonic-clonic* 38−8.75−15.07 to −2.430.007
Epilepsy typeFocal (reference) 1590.0130.909
IGE 71
AED exposure in uteroNone (reference) 736.3780.000
Valproate* 73−7.36−13.29 to −1.440.004
Carbamazepine 510.80−4.58 to 6.170.771
Phenytoin 215.68−1.51 to 12.860.121
Polytherapy 482.08−3.92 to 8.090.495