Libro selection bias and covariate imbalances in randomized clinical trials De vance berger

portada selection bias and covariate imbalances in randomized clinical trials

Formato

Libro Físico

Editorial

wiley

Autor

vance berger

N° páginas

218.00

ISBN

0470863625

ISBN13

9780470863626

Categorías

Probabilidad y estadística

selection bias and covariate imbalances in randomized clinical trials

Vance Berger (Autor) · wiley · Libro Físico

Sin Stock

Reseña del libro "selection bias and covariate imbalances in randomized clinical trials"

this is the first book-length treatment of the form of selection bias most apt to compromise the integrity of randomised trials. chapter 1 provides an overview of the hierarchy of study designs, and justifies the placement of randomised trials at the top of this hierarchy. chapter 2 delves deeper into what randomisation is and isnt and how this can guard against some but not all biases. the concepts of masking and allocation concealment are also discussed at length in chapter 2. chapter 3 provides evidence that the form of selection bias described in chapter 2 actually occurs, and so is not merely a hypothetical concern. chapter 4 details the harm that selection bias causes when it does occur, and so this completes part i of the book, consisting of the first four chapters, which may be described as a statement of the problem. part ii may be described as the actions that one can take to manage the problem that is selection bias, and begins with prevention, covered in chapter 5. while selection bias can be completely eliminated, doing so introduces other problems (primarily chronological bias), and so elimination of selection bias is not generally an ideal solution. the maximal procedure is developed as the optimal solution to one statement of the problem of the trade-off between chronological bias and selection bias. chapter 6 presents methods that can be used to detect selection bias in randomised trials, including the berger-exner test based on predicting patient outcomes or responses with both the actual treatment assigned and the likelihood with which this treatment was to be assigned (the reverse propensity score). chapter 7 discusses methods to correct for selection bias, including the ivanova-barrier-berger method, which is based on likelihoods. the book concludes with chapter 8, which provides a comprehensive plan for managing baseline imbalances and selection bias in randomised trials, and also describes open problems for future research.