Contents
Cover
Title Page
Copyright
Dedication
Foreword
Preface
Editors
Contributors
Part One: Statistics In The Development Of Pharmaceutical Products
Chapter 1: Statistical aspects in ICH, FDA and EMA guidelines
Synopsis
1.1 Introduction
1.2 ICH guidelines overview
1.3 ICH guidelines for determining efficacy
1.4 ICH quality guidelines
1.5 Other guidelines
1.6 Statistical challenges in drug products development and manufacturing
1.7 Summary
Chapter 2: Statistical methods in clinical trials
Synopsis
2.1 Introduction
2.2 Hypothesis testing, significance levels, p-values, power and sample size
2.3 Bias, randomization and blinding/masking
2.4 Covariate adjustment and Simpson's paradox
2.5 Meta-analysis, pooling and interaction
2.6 Missing data, intent-to-treat and other analyses cohorts
2.7 Multiplicity, subgroup and interim analyses
2.8 Survival analyses
2.9 Propensity score
2.10 Bayesian versus frequentist approaches to clinical trials
2.11 Adaptive designs
2.12 Drugs versus devices
Chapter 3: Pharmacometrics in drug development
Synopsis
3.1 Introduction
3.2 Pharmacometric components
3.3 Pharmacokinetic/pharmacodynamic analysis
3.4 Translating dynamic processes into a mathematical framework
3.5 Nonlinear mixed-effect modeling
3.6 Model formulation and derivation of the log-likelihood
3.7 Review of the most important pharmacometric software characteristics
3.8 Maximum likelihood method of population analysis
3.9 Case study: Population PK/PD analysis in multiple sclerosis patients
3.10 Mathematical description of the dynamic processes characterizing the PK/safety/efficacy system
3.11 Summary
Chapter 4: Interactive clinical trial design
Synopsis
4.1 Introduction
4.2 Development of the Virtual Patient concept
4.3 Use of the Virtual Patient concept to predict improved drug schedules
4.4 The Interactive Clinical Trial Design (ICTD) algorithm
4.5 Summary
Acknowledgements
Chapter 5: Stage-wise clinical trial experiments in Phases I, II and III
Synopsis
5.1 Introduction
5.2 Phase I clinical trials
5.3 Adaptive methods for Phase II trials
5.4 Adaptive methods for Phase III
5.5 Summary
Chapter 6: Risk management in drug manufacturing and healthcare
Synopsis
6.1 Introduction to risks in healthcare and trends in reporting systems
6.2 Reporting adverse events
6.3 Risk management and optimizing decisions with data
6.4 Decision support systems for managing patient healthcare risks
6.5 The hemodialysis case study
6.6 Risk-based quality audits of drug manufacturing facilities
6.7 Summary
Chapter 7: The twenty-first century challenges in drug development*
Synopsis
7.1 The FDA's Critical Path Initiative
7.2 Lessons from 60 years of pharmaceutical innovation
7.3 The challenges of drug development
7.4 A new era in clinical development
7.5 The QbD and clinical aspects
Part Two: Statistics In Outcomes Analysis
Chapter 8: The issue of bias in combined modelling and monitoring of health outcomes
Synopsis
8.1 Introduction
8.2 Example I: Re-estimating an infection rate following a signal
8.3 Example II: Correcting estimates of length-of-stay measures to protect against bias caused by data entry errors
8.4 Discussion
Chapter 9: Disease mapping
Synopsis
9.1 Introduction
9.2 Epidemiological design issues
9.3 Disease tracking
9.4 Spatial data
9.5 Maps
9.6 Statistical models
9.7 Hierarchical models for disease mapping
9.8 Multivariate disease mapping
9.9 Special issues
9.10 Summary
Chapter 10: Process indicators and outcome measures in the treatment of acute myocardial infarction patients
Synopsis
10.1 Introduction
10.2 A semiparametric Bayesian generalized linear mixed model
10.3 Hospitals’ clustering
10.4 Applications to AMI patients
10.5 Summary
Chapter 11: Meta-analysis
Synopsis
11.1 Introduction
11.2 Formulation of the research question and definition of inclusion/exclusion criteria
11.3 Identification of relevant studies
11.4 Statistical analysis
11.5 Extraction of study-specific information
11.6 Outcome measures
11.7 Estimation of the pooled effect
11.8 Exploring heterogeneity
11.9 Other statistical issues
11.10 Forest plots
11.11 Publication and other biases
11.12 Interpretation of results and report writing
11.13 Summary
Part Three: Statistical Process Control In Healthcare
Chapter 12: The use of control charts in healthcare
Synopsis
12.1 Introduction
12.2 Selection of a control chart
12.3 Implementation Issues
12.4 Certification and governmental oversight applications
12.5 Comparing the performance of healthcare providers
12.6 Summary
Acknowledgements
Chapter 13: Common challenges and pitfalls using SPC in healthcare
Synopsis
13.1 Introduction
13.2 Assuring control chart performance
13.3 Cultural challenges
13.4 Implementation challenges
13.5 Technical challenges
13.6 Summary
Chapter 14: Six Sigma in healthcare
Synopsis
14.1 Introduction
14.2 Six Sigma background
14.3 Development of Six Sigma in healthcare
14.4 The phases and tools of Six Sigma
14.5 DMAIC overview
14.6 Operational issues of Six Sigma
14.7 The way forward for Six Sigma in healthcare
14.8 Summary
Chapter 15: Statistical process control in clinical medicine
Synopsis
15.1 Introduction
15.2 Methods
15.3 Clinical applications
15.4 A cautionary note on the risk-adjustment of observational data
15.5 Summary
Appendix A
Acknowledgements
Part Four: Applications To Healthcare Policy And Implementation
Chapter 16: Modeling kidney allocation: A data-driven optimization approach
Synopsis
16.1 Introduction
16.2 Problem description
16.3 Proposed real-time dynamic allocation policy
16.4 Analytical framework
16.5 Model deployment
16.6 Summary
Acknowledgement
Chapter 17: Statistical issues in vaccine safety evaluation
Synopsis
17.1 Background
17.2 Motivation
17.3 The self-controlled case series model
17.4 Advantages and limitations
17.5 Why use the self-controlled case series method
17.6 Other case-only methods
17.7 Where the self-controlled case series method has been used
17.8 Other issues that were explored in improving the SCCM
17.9 Summary of the chapter
Chapter 18: Statistical methods for healthcare economic evaluation
Synopsis
18.1 Introduction
18.2 Statistical analysis of cost-effectiveness
18.3 Inference for cost-effectiveness data from clinical trials
18.4 Complex decision analysis models
18.5 Further extensions
18.6 Summary
Chapter 19: Costing and performance in healthcare management
Synopsis
19.1 Introduction
19.2 Theoretical approaches to costing healthcare services: Opportunity cost and shadow price
19.3 Costing healthcare services
19.4 Costing for decision making: Tariff setting in healthcare
19.5 Costing, tariffs and performance evaluation
19.6 Discussion
19.7 Summary
Part Five: Applications To Healthcare Management
Chapter 20: Statistical issues in healthcare facilities management
Synopsis
20.1 Introduction
20.2 Healthcare facilities management
20.3 Operating expenses and the cost savings opportunities dilemma
20.4 The case for baselining
20.5 Facilities capital … is it really necessary?
20.6 Defining clean, orderly and in good repair
20.7 A potential objective solution
20.8 Summary
Chapter 21: Simulation for improving healthcare service management
Synopsis
21.1 Introduction
21.2 Talk-through and walk-through simulations
21.3 Spreadsheet modelling
21.4 System dynamics
21.5 Discrete event simulation
21.6 Creating a discrete event simulation
21.7 Data difficulties
21.8 Complex or simple?
21.9 Design of experiments for validation, and for testing robustness
21.10 Other issues
21.11 Case study no. 1: Simulation for capacity planning
21.12 Case study no. 2: Screening for vascular disease
21.13 Case study no. 3: Meeting waiting time targets in orthopaedic care
21.14 Case Study no. 4: Bed Capacity Implications Model (BECIM)
21.15 Summary
Chapter 22: Statistical issues in insurance/payor processes
Synopsis
22.1 Introduction
22.2 Prescription drug claim processing and payment
22.3 Case study: Maximizing Part D prescription drug claim reimbursement
22.4 Looking ahead
22.5 Summary
Chapter 23: Quality of electronic medical records
Synopsis
23.1 Introduction
23.2 Quality of electronic data collections
23.3 Data quality issues in electronic medical records
23.4 Procedure to enhance data quality
23.5 Form design and on-entry procedures
23.6 Quality of data evaluation
23.7 Summary
Index
This edition first published 2012
© 2012 John Wiley & Sons, Ltd
Registered office
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, United Kingdom
For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com.
The right of the author to be identified as the author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.
Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought.
Library of Congress Cataloging-in-Publication Data
Statistical methods in healthcare / [edited by] Frederick W. Faltin, Ron S. Kenett, Fabrizio Ruggeri.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-0-470-67015-6 (cloth)
I. Faltin, Frederick W. II. Kenett, Ron S. III. Ruggeri, Fabrizio.
[DNLM: 1. Statistics as Topic–methods. 2. Data Collection–methods. 3. Delivery of Health Care–statistics & numerical data. WA 950]
362.102′1–dc23
2012009921
A catalogue record for this book is available from the British Library.
ISBN: 978-0-470-67015-6
To Donna, Erin, Travis and Madeline
– Frederick W. Faltin
To Jonathan, Alma, Tomer, Yadin, Aviv and Gili
– Ron S. Kenett
To Anna, Giacomo and Lorenzo
– Fabrizio Ruggeri
Foreword
Twenty-five years ago we launched an interesting experiment, ‘The National Demonstration Project for Quality Improvement in Healthcare’. It was a modest experiment bringing together twenty-one healthcare providers with twenty-one top industrial companies to explore whether industrial quality methods would work in healthcare settings. The results of this experiment were published as Curing Health Care: New Strategies for Quality Improvement. The statistical methods used by most of these healthcare providers were fairly basic tools of quality improvement; yet, many of the improvements were significant.
Looking back this many years later, there was no reason to be surprised by these results. Statistical methods had been used in many areas of healthcare for almost as many years as statistical methods had been used by any organization. Florence Nightingale was one of the first honorary members of the American Statistical Association, an organization that celebrated its 150th anniversary twenty-two years ago. Her pioneering work using clear, simple graphical methods to discover causes of death in hospitals during the Crimean War and alter British barracks was well known and celebrated. Basic, simple statistical methods to explore, understand and present data are as effective in healthcare applications as in any other endeavour.
But somehow, the science of quality control and quality improvement had passed healthcare by. Starting with Shewhart's control chart in 1924, statistical quality control had progressed quickly during the Second World War, and had been widely adopted and used by post-war Japan to become a leading producer of high-quality products. It had been rediscovered in the United States in the 1980s, and widely applied throughout the world in the 1980s and 1990s by companies in almost every competitive industry. Healthcare had evolved many methods of quality assurance, risk management and quality measurement, for the most part independent of what was happening in industry.
In some areas of healthcare, particularly in drug and medical device development and production, sophisticated methods had been created and widely used. Researchers in biostatistics, biometrics and clinical trials had developed and employed some of the most advanced statistical methods, and in turn contributed much to the statistical literature. These methods, however, did not seem to translate easily to the practice of continuous quality improvement in hospital-based care or general clinical practice. There was a considerable gap between what we knew how to do and what we were doing.
The National Demonstration Project evolved into the Institute for Healthcare Improvement, and the growing network of healthcare providers became increasingly adept in learning from sources outside of healthcare, adapting these methods to healthcare applications, and sharing encouraging results with each other. It was not only the statistical tools. The healthcare organizations picked up the methods of putting these tools to use in a scientific approach to improvement using PDSA (Plan-Do-Check/Study-Act), Juran's Quality Improvement Steps, Motorola/General Electric's Define-Measure-Analyse-Improve-Control (Six Sigma Quality), and full-scale implementations of the Toyota Production System (Lean).
Healthcare organizations around the world have formed collaboratives, networks and not-for-profit organizations to share these methods and statistical tools. Thousands of doctors, nurses and other practitioners now routinely attend healthcare quality conferences and daily participate in online courses, web-based sharing and local working groups. Organizations such as the Institute for Healthcare Improvement have tried to structure some of this learning through devices such as IHI's Improvement Roadmap and the Open School, but there has not been a simple place to find the statistical tools used in healthcare improvement until now.
Faltin, Kenett and Ruggeri have brought together leading researchers and practitioners in statistical methods to provide a wealth of methods in one place. Starting with some of the most sophisticated methods used in the development of pharmaceutical products and medical devices, and ending with applications to healthcare management, they have managed to cover amazing ground. The chapters on control charts bring together some of the best methods of statistical process control (SPC) in healthcare, and even cover some of the abuses in the use of control charts. The chapter ‘Six Sigma in Healthcare’ gives a remarkably thorough discussion of both Six Sigma and how it is being applied by many healthcare organizations in Europe and the USA.
But this book goes much further than the typical statistics text and addresses serious policy issues such as kidney allocation and offers advanced statistical methods as an approach to this critical problem. Another critical issue in healthcare, vaccine safety evaluation, is also addressed. In this time of crises in healthcare costs, the economics of healthcare is becoming a major issue. Here too, statistical methods have a large part to play.
The core of healthcare is, of course, clinical outcomes. Statistical methods play a critical role in outcomes analysis. Bias in modelling and monitoring health outcomes are addressed in a chapter by Grigg. Biggeri and Catelan discuss disease tracking. Guglielmi, Ieva, Paganoni and Ruggeri address process indicators and outcome measures in an important area, and Negri gives an excellent discussion of the special tool of meta-analysis.
We no longer need to discuss the value of statistical tools and quality improvement methods in healthcare. The value has been demonstrated thousands of times. What is needed is a comprehensive compilation of these tools in one place written by careful, knowledgeable authors. We should all be grateful to Faltin, Kenett and Ruggeri for providing it.
A. Blanton Godfrey
Dean, College of Textiles and Joseph D. Moore Distinguished University Professor
North Carolina State University
and Chair of the Board of Directors (2009–2012)
Institute for Healthcare Improvement
Preface
This book has its origins in the confluence of two realizations. First, that the availability and quality of healthcare is the defining issue of our time. And second, that statistics as a discipline pervades every aspect of the healthcare field.
Statistical Methods in Healthcare illustrates the spectrum of statistical applications to healthcare. From pharmaceuticals to health economics, drug product development to facilities management, clinical outcomes to electronic medical records, risk assessment to organ allocation, statistics has permeated every corner of healthcare. Accordingly, we have assembled here an array of chapters, prepared by a broadly international group of leading authors, which address all of these topics, and many more. Our objective was not to touch upon every area of statistical application in healthcare – that would be impossible. Rather, our purpose has been to span, as best we can, the diverse domains to which statistics has been applied and, thereby, to contribute to the evolution of statistical methods in healthcare applications.
The book consists of 23 chapters organized in five parts:
Not surprisingly, such an effort has been the work of contributors from many fields. Statistical Methods in Healthcare integrates contributions from statisticians, economists, physicians, epidemiologists, operations researchers, actuaries and managers, among others. The outcome captures perspectives from all of these disciplines, providing an integrated interdisciplinary view reflecting the richness and complexity of healthcare applications.
Our hope and belief is that this collective effort will prove valuable to those in a wide array of professions which in some way touch upon healthcare. Not only statisticians, but researchers, physicians and administrators will find here statistical applications with detailed examples representing a variety of problems, models and methodologies. Students and practitioners alike will discover opportunities to innovate via the use of statistical methods.
We'd like to acknowledge and thank the many people whose contributions have made this work possible. These include, first and foremost, our esteemed colleagues who have contributed chapters to the work, and the outstanding editorial, production and copy-editing teams at Wiley, who followed up our work together on The Encyclopedia of Statistics in Quality and Reliability with another successful outing. And of course, our thanks go especially to our families, for their patience with us while we were preoccupied or otherwise disengaged throughout the duration of this project.
This book includes an accompanying website www.wiley.com/go/statistical_methods_healthcare
Editors
Frederick W. Faltin
Founder and Managing Director
The Faltin Group
25 Casper Drive
Cody, WY 82414, USA
Ron S. Kenett
Chairman and CEO, The KPA Group
KPA Ltd, PO Box 2525
Hattaassia Street, 25
Raanana 43100, Israel
and
Research Professor
Università degli Studi di Torino
10134 Turin, Italy
Fabrizio Ruggeri
Research Director
CNR IMATI
Via Bassini 15
I-20133 Milano, Italy
Contributors
Benjamin M. Adams
Department of Information Systems Statistics and Operations Management
University of Alabama
Tuscaloosa, AL
USA
Zvia Agur
Optimata Ltd.
Ramat Gan, Israel
and
Institute for Medical Biomathematics (IMBM)
Bene Ataroth, Israel
Robert Bauer
ICON Development Solutions
University Blvd.
Ellicot City, MD
USA
James C. Benneyan
Healthcare Systems Engineering Institute
Northeastern University
Boston, MA
USA
Paola Berchialla
Department of Public Health and Microbiology
University of Torino
Turin
Italy
Annibale Biggeri
Department of Statistics ‘G. Parenti’
University of Florence
Florence, Italy
and
Biostatistics Unit
ISPO Cancer Prevention and Research Institute
Florence, Italy
Dolores Catelan
Department of Statistics ‘G. Parenti’
University of Florence
Florence, Italy
and
Biostatistics Unit
ISPO Cancer Prevention and Research Institute
Florence, Italy
Shirley Y. Coleman
Industrial Statistics Research Unit
Newcastle University
Newcastle upon Tyne
UK
Caterina Conigliani
Department of Economics
University of Roma Tre
Rome
Italy
Anja Drescher
Operations Analytics
Integrated Facilities Management
Jones Lang LaSalle Americas, Inc.
Minnetonka, MN
USA
Dario Gregori
Unit of Biostatistics, Epidemiology and Public Health
Department of Cardiac, Thoracic and Vascular Sciences
University of Padova
Padua
Italy
Olivia A. J. Grigg
CHICAS
School of Health and Medicine
Lancaster University
Lancaster
UK
Alessandra Guglielmi
Department of Mathematics
Politecnico di Milano
Milan
Italy
Serge Guzy
POPPHARM
Albany, CA
USA
Francesca Ieva
Department of Mathematics
Politecnico di Milano
Milan
Italy
Telba Irony
General and Surgical Devices Branch
Center for Devices and Radiological Health
US Food and Drug Administration
Silver Spring, MD
USA
Victoria Jordan
Office of Performance Improvement
University of Texas MD Anderson Cancer Center
Houston, TX
USA
Caiyan Li
Baxter Healthcare Corporation
Round Lake, IL
USA
Andrea Manca
Centre for Health Economics
The University of York, York
UK
Patrick Musonda
School of Medicine, Norwich Medical School
University of East Anglia
Norwich, UK
and
Centre for Infectious Disease Research in Zambia (CIDRZ)
Lusaka, Zambia
Eva Negri
Department of Epidemiology
Istituto di Ricerche Farmacologiche ‘Mario Negri’
Milan
Italy
Daniel P. O’Neill
Healthcare Solutions
Jones Lang LaSalle Americas, Inc.
Chicago, IL
USA
Anna Maria Paganoni
Department of Mathematics
Politecnico di Milano
Milan
Italy
Melissa Popkoski
Pharmacy Administrative Services
Horizon Blue Cross Blue Shield of New Jersey
Newark, NJ
USA
Allan Sampson
Department of Statistics
University of Pittsburgh
Pittsburgh, PA
USA
Anne Shade
Good Decision Partnership
Ingleneuk, Strathdrynie
Dingwall, Scotland
UK
Phyllis Silverman
General and Surgical Devices Branch
Division of Biostatistics
Center for Devices and Radiological Health
US Food and Drug Administration
Silver Spring, MD
USA
Yafit Stark
Innovative R&D Division
TEVA Pharmaceutical Industries, Ltd.
Netanya
Israel
Andrea Tancredi
Department of Methods and Models for Economics, Territory and Finance
University of Roma ‘La Sapienza’
Rome
Italy
Rosanna Tarricone
Department of Policy Analysis and Public Management
Centre for Research on Health and Social Care Management – CERGAS
Università Bocconi
Milan
Italy
Aleksandra Torbica
Department of Policy Analysis and Public Management
Centre for Research on Health and Social Care Management – CERGAS
Università Bocconi
Milan
Italy
Per Winkel
The Copenhagen Trial Unit
Centre for Clinical Intervention Research
Rigshospitalet, Copenhagen University Hospital
Copenhagen
Denmark
William H. Woodall
Department of Statistics
Virginia Tech
Blacksburg, VA
USA
Inbal Yahav
The Graduate School of Business
Department of Information Systems
Bar Ilan University
Israel
Shelemyahu Zacks
Department of Mathematical Sciences
Binghamton University
Binghamton, NY
USA
Nien Fan Zhang
Statistical Engineering Division
National Institute of Standards and Technology
Gaithersburg, MD
USA
Part One
STATISTICS IN THE DEVELOPMENT OF PHARMACEUTICAL PRODUCTS