Removing age adjustment from the main multivariate models resulted in substantially larger estimates, very similar to the crude associations.
No main multivariate effect of smoking behavior was found nor was a multivariate interaction effect of scenario condition with smoking behavior found (ps > .49).49
A propensity score weighted generalized linear model was used as the main multivariate analysis method; age, baseline A1c, copay, initial year, region, health plan, diabetes education, baseline comorbidities, diabetes medication, health care utilizations, and costs were controlled in the model.
Instead, the factors found to be significant in the main multivariate reduced model were added to each of these individual lipid models to adjust for confounding, among those with available lipid data.
Odds ratio (OR) with 99% confidence interval (CI) and model fit (c-statistic and R ) were calculated per diagnosis group and per age group using the main multivariate logistic regression model.
Validating the association of CHAC1 and its transcript variants with survival within breast-cancer entities, a P-spline regression model confirmed the results of the main multivariate Cox model in sensitivity analysis (Supplementary Figure 1).
