A multiple variable logistic regression model identified that lower household per capita income (odds ratio [OR] of quartile 1 = 6.36, OR of quartile 2 = 4.7, OR of quartile 3 = 3.87; P = 0.001 for linear trend), age (OR = 1.3; P < 0.001), and male gender (OR, 2.2; P = 0.001) were associated with grade retention; silent cerebral infarct (P = 0.31) and painful episodes (P = 0.60) were not.
With stepwise multivariate analysis, sequentially adding clinical, echocardiographic, and catheterization variables, the overall model identified independent predictors of survival as baseline functional status, baseline cardiac output, renal function, cachexia, female gender, left ventricular systolic function, and mitral regurgitation.
While the multiple linear regression model above identified variables likely to impact shedding concentrations, the logistic model identified variables important for detecting a positive sample (e.g., during surveillance efforts).
Multivariate analysis using the Cox's proportional hazard model identified the variable PROX1 expression as an independent predictor for poor survival in the full model (P=0.0275), as well as in the stepwise model (P=0.0269) (Table 5).
Multivariate analysis using the Cox's proportional hazard model identified the variable PROX1 expression as an independent predictor for poor overall survival, in the full model (P=0.0310) as well as in the stepwise model (P=0.0237) (Table 4).
A conservative penalized multinomial logistic regression model identified 14 variables that could accurately classify subjects, including frontal, temporal, parietal, occipital and cerebellum volume.
The final conditional logistic regression model identified 3 variables as independent risk factors for HPAI (H5N1) infection of backyard chickens in Bangladesh (Table 3).
Whereas the CART model identified three variables: type of medical; how many visits the employee had and site; the BRT model identified BMI and lung function as the most important factors predicting the proportion of missingness in the employees' health records.
The multivariate regression model identified six explanatory variables of environmental, socioeconomic and demographic nature.
Competing latent variable models were identified in previous studies.
