For example, Burke and colleagues (2015) developed a multivariate model to predict treatment gains from a rehabilitation program in patients with stroke.
We assessed outcomes in social/health maintenance organizations (S/HMOs) and Medicare fee-for-service (FFS) care using a multivariate model to estimate active life expectancy (ALE).
Finally, we perform a multivariate model to identify variables associated with the occurrence of lymphopaenia, but we did not specifically investigate the lymphocyte count as a continuous variable.
We did not insert baseline assessments or the abovementioned covariates in the same multivariate model to avoid overfitting (Hawkins, 2004; Zhang, 2014).
We used these three criteria in a multivariate model to develop a scoring system (0 to 10), which allowed a highly specific diagnosis of acute myocardial infarction to be made.
To quantify the importance of the health-related factors as a whole, we also compared the final multivariate model to a model containing only demographic and criminal justice variables.
As described in Methods, we fit a multivariate model to identify patient and ED characteristics that might predict adoption of at least a "basic" EMR.
Established prognostic variables were then added to the TIMI UA/NSTEMI multivariate model to evaluate their incremental benefit to model discrimination [11].
A full multivariate model is therefore more accurate than its corresponding risk score, and refitting of an externally-derived multivariate model to the study population further improves model discrimination [21], [22].
Age and race were also included in the initial multivariate model to assess potential confounding.
A stepwise procedure then allowed the final multivariate model to be obtained.
