Data for: Benchmark dose profiles for bivariate exposures

Akkaya-Hocagil et al (2024) present a methodology to conduct a benchmark dose analysis to characterize the joint effect of a two-dimensional exposure on a continuous outcome using a generalized additive model while adjusting for potential confounders via propensity scores. To illustrate this approach, the paper analyzes data from six U.S. longitudinal cohort studies to quantify levels and patterns of prenatal alcohol exposure that are associated with the development of clinically important cognitive deficits in children born to mothers who consume alcohol during pregnancy. More details can be found in Jacobson et al. (2024). The dataset included in this repository includes a cohort indicator (1-6), a standardized outcome variable (CogScore derived from a structural equation model analysis); three maternal alcohol consumption variables that are quantified in terms of the log(x+1) transformed amount of alcohol consumed on each drinking day during pregnancy (Alch 1), the frequency of drinking days (Alch2), and the interaction between the Alch1 and Alch2 (Alch3); and pre-computed and centered propensity scores (PS1, PS2, PS3 respectively).