Most of them were still significant even after considering the ER status and node status using the multiple variable model (Table 3).
For discrimination in modeling datasets, ANN was significantly higher than CLR in AUROC and accuracy in 16- and 6-variable models (Table 2).
As for calibration in modeling datasets, ANN had significantly lower HL Chi-squares and was more calibrated than CLR in 16-variable models (Table 3).
In the uni-variable model, the meta-regression coefficient for weight change was significantly higher in trials with a sample size ≥100 (1.95 kg; CI: 0.07, 3.82; p = 0.042), but none of the explanatory variables were significant predictors of intervention group weight change in the multi-variable model (Table 2).
For the harpaline clade, a model of two abrupt rate shifts, Model 6: yule3rAIC (AIC = -2.60), was chosen over other constant and variable rate models (Table 5).
While the model with equal rates among clades was strongly rejected in favor of variable rate models (Table 5), the highest marginal likelihood was assigned to a model with three different rates assigned to the clades I, II+III, and IV, respectively.
After adjustment for age at diagnosis, year of diagnosis and either neighborhood SES (model 1) or the combined neighborhood SES/enclave variable (model 2) (Table 2), foreign-born Hispanics remained 14%-15 14%-15likely to be diagnosed at a regional/distant stage than their US-born counterparts.
There were no predisposing variables significantly associated with utilization after statistically adjusting for the other variables in the models (Table 3).
The following table shows the assignment of selected variables to defined models (Table 2).
Tumor stage, nodal stage, tumor differentiation, performance status, vascular infiltration, and histology were independent prognostic variables in both models (Table 3).> -wrap-foot> †Overall significance as prognostic marker.
Age was kept as a continuous variable in the FP models (Table 4).
