In univariate analysis age, treatment, and tumor location were identified for inclusion in the multivariate model, and all remained significant (p < 0.01).
The multivariate model and training set was used to predict GC retention times on a non-polar column (DB-5).
A nomogram was developed based on this multivariate model and validated internally using bootstrap methods, with a concordance index of 0.69.
Second, 13% of participants were missing data for one or more baseline variables included in our final multivariate model, and were hence excluded.
Maximum-likelihood estimation was then performed to analyze all factors available on ICU admission and significantly associated with the neurological outcome in the multivariate model, and the results were used to develop a predictive score (the MyeloScore).
All variables were included in the multivariate model and, after adjusting for other factors, those that were not predictive were excluded, with the exception of age, sex and Indigenous status.
Factors with p values less than 0.25 in univariate analysis were tested in the multivariate model, and p values less than 0.05 were considered significant.
The clinical prediction rule was then calculated using the formula based on the coefficients from the multivariate model and examined using various cut-points for the diagnosis of TBM.
Univariate predictors with a two-sided p-value ≤0.2 were included in the multivariate model, and backward elimination was applied to obtain a final model; two-sided p-values ≤0.05 were considered significant.
Significant independent variables from the univariate analysis were included into a multivariate model and analysed.
(c) Not significantly associated with smoking cessation in multivariate model and eliminated from final model.
