Skip to Content
Interactive Textbook on Clinical Symptom Research Logo


Home Button

Statistical Models for Prognostication
Author Bio
Introduction
Predictions: Statistical Models
Insight: Statistical Models
Currently selected section: Ingredients: Statistical Models
Theoretical Aspects
Central Concepts
Regression Models
Problems: Regression
Practical Advice
Example 1
Example 2




Chapter 8: Statistical Models for Prognostication: Ingredients of Statistical Models
        

Types of Regression Models

We illustrate the application of logistic regression in Example 1: Gallstones (Lee et al., 1995). For categorical variables, polytomous logistic regression models can be used. When an ordering is present, special types of polytomous models may be chosen (Harrell et al., 1998).

For survival variables, analyses are often performed with Kaplan-Meier curves.

Figure 4.2: Kaplan-Meier Survival Curves
Example of a Kaplan Meier survival curve, described in text
Illustration of Kaplan-Meier survival curves. The observed survival is shown
for four disease groups until five years after hospitalization. The five-year
survival is around 10% for patients hospitalized with coma or cancer, and
around 35% for patients for the other disease groups. The data are from a
random sample of 1000 patients in the SUPPORT study (Knaus et al., 1995); http://hesweb1.med.virginia.edu/biostat/s/data/.

Kaplan-Meier curves have advantages and disadvantages.

  • An advantage of Kaplan-Meier curves is that they are non-parametric, i.e. that no assumptions are made on the distribution of survival times.

  • A disadvantage is that only a few factors can be considered (dichotomous or categorical), and those continuous variables cannot be handled without categorization.

Previous Page