Follow-up analyses evaluated a series of statistical interactions in the DTSA model to determine whether the interaction effects exist. Following the procedure of Singer & Willett (2003), these analyses were restricted to examine interactions among only those risk factors shown to be significant from the previous models. These interactions were tested by creating a cross-product term and adding it to a model that included the main effects of the covariates comprising the interaction. This model was compared to a model excluding the cross-product variable using a LRT to determine whether the inclusion of each interaction significantly improved overall model fit. All models were run in Mplus v5.1 (Muthen & Muthen, 1998–2007). In order to increase confidence in the final maximum likelihood values, automatically generated starting values with random perturbations (100 random sets of starting values with 30 full optimizations) were used for all models described below.
