Search

Background: We undertook this study to determine whether the widely replicated link between maternal smoking and conduct disturbance (Cd) is better explained by a model of direct causation or of mother-offspring transmission of a latent Cd variable., Methods: Family data collected on 538 adolescent twin boys from the Virginia Twin Study of Adolescent Behavioral Development (VTSABD) was used to compare two alternative models: 1) a model composed of a latent transmissible factor that influences mother's juvenile conduct symptoms, smoking during pregnancy, and subsequent Cd and smoking in her adolescent boys; and 2) a model specifying a direct causal path from mother's smoking to child Cd., Results: The maternal-offspring transmission model fit the data as well as a model specifying a direct causal path from maternal smoking to child Cd., Conclusions: Our results suggest that the association between maternal smoking during pregnancy and boys' Cd symptoms may be attributed to the transmission of a latent Cd factor and not to a direct effect of the smoking. Our results challenge previous findings of a direct effect of prenatal smoke exposure on risk to Cd once other etiologic factors are considered.

The expression of many important behavioral outcomes is contingent on passing a developmental milestone, such as puberty, or exposure to an age-dependent risk factor, such as alcohol or drugs. Current models for genetic effects on development have not provided explicit theoretical insight about possible patterns of epistasis and genotype x environment interaction generated by genetic and environmental heterogeneity in the timing of developmental milestones that influence gene expression. We explore theoretically the effects of individual differences in the timing of a developmental milestone on the expression of genetic differences in outcome. The consequences of additive genetic variation and shared environmental variation on timing are examined. Additive genetic effects on timing generate transient epistatic effects on outcome because they modulate the expression of otherwise additive genetic differences in outcome. Epistatic effects will be greatest at ages when there is the most heterogeneity in developmental maturity. Epistasis will gradually disappear as proportionately more individuals mature. Similarly, shared environmental effects on the timing of a milestone will generate transient genotype x shared environment interaction appearing as shared environmental effects on outcome that are greatest when there is the most heterogeneity in developmental maturity. The impact of these effects on age patterns in twin correlations is examined and their relevance to current data on developmental change discussed briefly.

Background: The aim was to compare the psychometric properties of absolute versus differential ratings of the within-family environments and to examine their associations with conduct problems in 8-16-year-old twins., Methods: The sample comprised 1117 pairs of like-sex male and female twins and their parents, recruited from the school population of Virginia. The within-family environment was assessed from the Twin Inventory of Relationships and Experiences (TIRE), which provided measures of differential parental criticism, parental preference for one twin, and twins' peers' conduct problems, as rated by mothers, fathers and each of the twins. Twins' conduct problems were assessed by a modified version of the Olweus questionnaire., Results: Although differential ratings and absolute ratings had similar psychometric properties, the use of absolute ratings to differentiate the twins resulted in a construct with weak inter-rater agreement and low stability over time. Differential ratings of the twins' environment showed significant associations with differences in conduct problems between the twins, whereas the difference in absolute ratings of each twin did not. The differential parental criticism effect was uninfluenced by the overall level of criticism in the family., Conclusions: A direct contrast between siblings or twins, as part of the rating procedure (as provided by differential ratings), may be a more efficient way to assess aspects of the within-family environment that are reliable and which are associated with psychopathology. Both differential parental criticism and peers' conduct disturbance were associated with conduct problems in the twins.

Akiake's Information Criterion (AIC) is commonly used in univariate twin modeling of a discrete trait to prune a full model into a more parsimonious submodel. It is possible that this practice could introduce bias and inaccuracy, and we could identify no prior systematic study of these issues. Thus, we used simulation to investigate the performance of AIC-guided modeling across a broad range of parameters. Our simulations indicated that the use of the AIC to determine the "best" univariate model for a discrete trait tended to yield the incorrect model rather frequently. Moreover the parameter estimates of the "best" model by AIC were biased sharply upward as were the associated 95% confidence intervals. These results suggest that the use of AIC to guide twin modeling for univariate discrete traits should either be abandoned or used with great caution.