, 2007). Demographic and obstetric information such as maternal age, ethnicity, and education, was assessed using self-report. Parental educational level and national origin were defined according to the classification of Statistics Netherlands (Statistics Netherlands, 2004a and Statistics Netherlands, 2004b).
Parental psychopathology selleck chemicals llc was measured using the Brief Symptom Inventory (BSI), a validated 53-item (5-point scale) self-report symptom inventory that ascertains the psychological state (Derogatis and Melisaratos, 1983); the Global Severity Index subscale was used to determine general psychopathology symptoms. Multiple linear regression models were conducted to examine the associations of maternal cannabis use, tobacco use and non-use of cannabis and tobacco with the continuous scores of child behavioral
problems. AZD2281 All models controlled for age and gender of the child, parental education, national origin and psychopathology. Correction for alcohol use and other drug use did not significantly change the results; therefore we report results obtained by correcting only for covariates that significantly influenced the model. Identical analyses were carried out to examine the association between paternal cannabis use and child behavioral problems. The paternal models were additionally corrected for maternal cannabis and/or tobacco use. We performed stratified analyses for the effects of parental cannabis use and behavioral problems for boys and girls, because the multiplicative interaction
terms between gender and maternal cannabis and tobacco use were significant. Using a categorical distinction to classify children as having behavioral problems in the clinical range, we performed logistic regression analyses. The main reason for the use of this Phosphatidylinositol diacylglycerol-lyase categorical variable was to determine the clinical relevance. The data were somewhat skewed (0.78 for attention and 0.76 aggression) and kurtotic (0.28 for attention and 0.63 for aggression); these numbers lie within the range for normal distribution. Measures of association (Beta’s (B) and Odds Ratios, OR’s) are presented with the 95% Confidence Intervals (CI’s). Statistical analyses were performed using the Statistical Package of Social Sciences version 15.0 for Windows (SPSS Inc., Chicago, IL). Table 1 depicts the sample characteristics used in the study. Women using cannabis predominantly used cannabis in early pregnancy, 19.3% of the mothers who used cannabis continued their use after the first trimester. Compared to non-using mothers, mothers who used cannabis during pregnancy often drank alcohol during pregnancy, were less educated, and had lower psychopathology scores. Additionally, women using cannabis during pregnancy were more often of Surinamese (12.5% in the cannabis-users vs. 5.4% in the non-users) and Antillean national origin (14.5% in the cannabis-users vs. 1.