We created storyboards, page outlines, computational infrastructure for entry of variables necessary for the ProVent model, and a better graphical display of prognosis.
To assess predictors of 100% compliance with the resuscitation bundle of the SSC, multivariable binary logistic regression was done with forward entry of variables that had P < 0.2 in univariate analysis.
Entry of variables to the model was determined using the Log-Likelihood ratio test, with p < 0.05.
Initial model development included entry of variables into a forward stepwise model, with the probability criterion for entry set at 0.05 and exit at 0.10.
Stepwise models were built by forward selection and entry of variables showing univariate associations with psychiatric disorder to a significance of p < or = 0.2.
The criterion for inclusion was forward selection based on the likelihood ratio test using a cut-off p-value of 0.05 for entry of variables into a preliminary model.
The strengths of the current study include use of validated theoretical and conceptual models, large sample size, inclusion of a broad range of neighborhood and community factors, entry of variables in blocks based on theoretical relationships to assess incremental effect, and focus on multiple self-care behaviors and outcomes.
The results are shown in Table 3. Premorbid IQ as measured by the NART was the only significant predictor of outright refusal at Wave 2, while both NART and MMSE scores were significant predictors of the Informant Only group at Wave 2. Stepwise entry of variables was also undertaken to further examine the role of predictors in the odds of refusal.
Here again, hierarchical regression analyses were computed, but now using the forced entry method of variable inclusion, making the results in the five imputed data sets directly comparable.
By contrast, for functional disability depression severity remained the strongest predictor, and after forced entry of this variable (R2 = 40.4%, F = 10.2, df = 1,16 p = 0.006), memory retention and SRCDs were no longer significant predictors.
