Contents
A user’s guide
Alejandro R Jadad
Introduction
Alejandro (Alex) R. Jadad
Dundas, 11 June 1998
This is a book for busy readers who need, wish or have to understand the basic principles of randomised controlled trials (RCTs) and their role in health care decisions. It is my best effort to fill a gap in the literature that I was not able to fill with any single source during the past 15 years.
In the 1980s, as a medical student, intern, resident, and novice researcher, I could not find a single source that could help me really understand what RCTs were about, their strengths and limitations, and how to use them while making health care decisions. During that decade, I had to rely on many sources, most of which had been written in a language that was difficult for me to understand, that presented information in formats that I could not adapt to my busy life or to my rapidly changing needs. I attributed my failed attempts to find a single, easy-to-read source of basic information to the fact that I was in Colombia, a developing country where it was often difficult to access the biomedical literature.
In 1990 I moved to Oxford, England, where I spent 5 years as a research fellow, as a practising clinician and as a graduate student. At the beginning of this period, the need for a good source of information to fill my most basic knowledge gaps intensified. My colleagues at Oxford were seasoned trialists. They not only designed and analysed their own trials, but also criticised, quite strongly, RCTs done by others. As I did not want to bother my colleagues with basic questions (or to sound dumb), I started looking, again, for a source of information that could meet my most immediate needs. To my surprise, I could not find one. I started to list my questions and decided to get answers as soon as possible, from as many sources as necessary. I spent vast amounts of time (and not insignificant amounts of money) trying to get small pieces of information from different books, most of which targeted doers of research, not users. I also started asking for key references from colleagues, and tracked down articles mentioned by speakers at conferences or included in the reference lists of the articles I was collecting from other sources. As my list of questions grew, so did my list of references and contacts. After a while, I felt more comfortable to talk to my colleagues about trials, started to design and lead some trials myself, and was invited to coordinate journal club sessions, where the work of others was rigorously appraised. Two years after my arrival in Oxford, one of my RCTs was published in the Lancet! This experience, however, created a new set of challenges. Colleagues started to invite me to give lectures on pain research and trial design. I found myself under a different type of pressure. I had the opportunity, for the first time, to transmit to others what I had learnt. I began to feel confident about acknowledging my own uncertainties and knowledge gaps. People seemed to like my presentations and I started to receive invitations to give lectures. It did not take me long to realise that the attendees were asking similar questions from lecture to lecture, and that those questions were similar to those I had listed before. I started recording the questions and made every effort to address them during each successive lecture. Soon, some of the new questions from the audience started to coincide with my own questions at the time. I kept adding any new questions to my list for the five years I was in England. The questions I collected, approximately 100, form the backbone of this book.
In 1992, I was accepted as a graduate student at the University of Oxford and started to work on a thesis on meta-analysis of randomised trials in pain relief. As part of my thesis work, I coded over 15,000 citations and created a database with over 8,000 RCTs. I also led the development of what is regarded by many as the first validated tool to appraise the quality of RCTs, and created new statistical methods to combine data from different RCTs addressing the same topic. When I tried to use the newly developed methods to answer clinical important questions, I realized that despite having thousands of RCTs at my fingertips, relevant data were often unavailable or poorly reported. I also started to formulate questions around RCTs that had not yet been answered at all and became interested in the design of methodological studies to answer them. During this period, I started to meet people who shared my interest in addressing unanswered methodological questions. Most of them had published a number of studies that looked at the RCT as the subject of research, rather than as the research tool. Through my interaction with this different breed of researchers and their work, and my own methodological studies, I became aware of how vulnerable the RCT can be to bias, imprecision, irrelevance and politics at all stages of its development and in all areas in health care.
Soon after I completed my thesis work, I moved to McMaster University, where I expected to have a quieter time, as a postdoctoral fellow, applying all my new knowledge on RCTs and systematic reviews, and exploring ways to involve consumers in health decisions. I was pleasantly surprised. I found lots of exciting work on RCTs and systematic reviews at McMaster; but through my interaction with my new colleagues, I became aware of the profound influence that factors other than evidence from RCTs can have on clinical decisions and health outcomes. Under their influence, I began to grasp the complexity of human-human, human- information, human-technology and information-technology interactions. My world became richer, and once again I faced new challenges. I started to look on RCTs as powerful but vulnerable tools that can make important but small contributions to a large puzzle with many pieces that are continuously changing their shape, size and patterns.
At that point, writing a book on RCTs moved from being a relatively easy task to a daunting one. I almost turned down the offer to write this book. It took me more than a year to draft the first chapter, and more than another year to complete the book.
At first, I was tempted to target the book to both doers (i.e. fellow researchers and academics) and users (i.e., clinicians, policy makers and managers, and if brave enough, consumers) of research. After drafting many tables of contents, I realized that trying to meet the needs of such a diverse group at the same time would be a mistake. After some deliberation, I decided to produce an introductory guide for busy readers. Primarily, this book is aimed at clinicians, young researchers, research staff and trainees in any health profession. Nevertheless, I hope that health planners, managers, journal editors, peer-reviewers, journalists, members of consumer advocacy groups and even seasoned researchers could also find it relevant and useful.
The book is divided into eight chapters, and addresses 100 questions with short answers and key references. The first five chapters focus on individual trials and include definitions of key elements of trials, descriptions of different types of RCTs and sources of bias, discussions of different approaches to quality assessment, and tips on trial reporting and interpretation. The sixth chapter focuses on groups of RCTs and includes discussions on different types of reviews, meta-analyses and clinical practice guidelines. The seventh chapter addresses the role of RCTs in health care decisions, their relationship with the other types of information, values, preferences and circumstances. This chapter also introduces the basic principles of evidence based decision-making and highlights its strengths and limitations. The last chapter describes ‘my wish list’. In it, I highlight what I think are the most important barriers to the optimal use of RCTs in health care, and propose some strategies that could be used to overcome them.
Writing each of the chapters of this book was an extraordinary experience, full of challenges and lessons. The main challenge was to ensure that I could complete the book without affecting my family life or my responsibilities as a new faculty member. The only way in which I could do this was by modifying my sleep pattern. This book was written, essentially, from 11 pm to 2 am or from 4 am to 6 am. Working at these hours gave me more time than I had had in years to think about trials and to put together my thoughts without interruptions. Putting each chapter together forced me to be concise and to use easy-to-read language to describe even the most complex aspects of RCTs. I made every effort to describe the key elements and implications of RCTs without statistics, and in a way that would appeal to busy readers. Each chapter created new questions and opened new avenues for me to explore. During each session I struggled, continuously, to keep the balance between the need to describe the basic elements of RCTs and the urge to express my concerns and expectations about health care research as a whole.
In sum, I did my best to provide you with an enjoyable, basic, balanced, useful and broad view of RCTs and their role in health care. I hope I succeeded.
|
Home | Contents | Foreword | Introduction | Acknowledgments | How to order
© BMJ Books 1998. BMJ Books is an imprint of the BMJ Publishing Group. First published in 1998 by BMJ Books, BMA House, Tavistock Square, London WC1H 9JR. A catalogue record for this book is available from the British Library. ISBN 0-7279-1208-9