Clinical Prediction Models - Ewout W. Steyerberg - Bog

PrefaceviiAcknowledgementsxi
Chapter 1Introduction1
1.1Diagnosis, prognosis and therapy choice in medicine1
1.1.1Predictions for personalized evidence-based medicine1
1.2Statistical modeling for prediction5
1.2.1Model assumptions5
1.2.2Reliability of predictions: aleatory and epistemic uncertainty6
1.2.3Sample size6
1.3Structure of the book81.3.1Part I: Prediction models in medicine8
1.3.2Part II: Developing internally valid prediction models8
1.3.3Part III: Generalizability of prediction models9
1.3.4Part IV: Applications9
Part I: Prediction models in medicine11
Chapter 2Applications of prediction models13
2.1Applications: medical practice and research13
2.2Prediction models for Public Health14
2.2.1Targeting of preventive interventions14*2.2.2Example: prediction for breast cancer14
2.3Prediction models for clinical practice17
2.3.1Decision support on test ordering17
*2.3.2Example: predicting renal artery stenosis17
2.3.3Starting treatment: the treatment threshold20
*2.3.4Example: probability of deep venous thrombosis20
2.3.5Intensity of treatment21
*2.3.6Example: defining a poor prognosis subgroup in cancer22
2.3.7Cost-effectiveness of treatment23
2.3.8Delaying treatment23
*2.3.9Example: spontaneous pregnancy chances24
2.3.10Surgical decision-making26
*2.3.11Example: replacement of risky heart valves27
2.4Prediction models for medical research28
2.4.1Inclusion and stratification in a RCT28
*2.4.2Example: selection for TBI trials29
2.4.3Covariate adjustment in a RCT30
2.4.4Gain in power by covariate adjustment31
*2.4.5Example: analysis of the GUSTO-III trial32
2.4.6Prediction models and observational studies32
2.4.7Propensity scores33
*2.4.8Example: statin treatment effects342.4.9Provider comparisons35
*2.4.10Example: ranking cardiac outcome35
2.5Concluding remarks35
Chapter 3Study design for prediction modeling37
3.1Studies for prognosis37
3.1.1Retrospective designs37
*3.1.2Example: predicting early mortality in esophageal cancer37
3.1.3Prospective designs38
*3.1.4Example: predicting long-term mortality in esophageal cancer393.1.5Registry data39
*3.1.6Example: surgical mortality in esophageal cancer39
3.1.7Nested case-control studies40
*3.1.8Example: perioperative mortality in major vascular surgery40
3.2Studies for diagnosis41
3.2.1Cross-sectional study design and multivariable modeling41
*3.2.2Example: diagnosing renal artery stenosis41
3.2.3Case-control studies41
*3.2.4Example: diagnosing acute appendicitis42
3.3Predictors and outcome42
3.3.1Strength of predictors42
3.3.2Categories of predictors42
3.3.3Costs of predictors43
3.3.4Determinants of prognosis44
3.3.5Prognosis in oncology44
3.4Reliability of predictors45
3.4.1Observer variability45
*3.4.2Example: histology in Barrett''s esophagus45
3.4.3Biological variability46
3.4.4Regression dilution bias46
*3.4.5Example: simulation study on reliability of a binary predictor46
3.4.6Choice of predictors47
3.5Outcome47
3.5.1Types of outcome47
3.5.2Survival endpoints48
*3.5.3Examples: 5-year relative survival in cancer registries48
3.5.4Composite endpoints49
*3.5.5Example: composite endpoints in cardiology49