Medical Statistics from Scratch An Introduction for Health Professionals Second Edition David Bowers Honorary Lecturer, School of Medicine, University of Leeds, UK Medical Statistics from Scratch Second Edition Medical Statistics from Scratch An Introduction for Health Professionals Second Edition David Bowers Honorary Lecturer, School of Medicine, University of Leeds, UK Copyright C 2008 John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England Telephone (+44) 1243 779777 Email (for orders and customer service enquiries): cs-books@wiley.uk Visit our Home Page on www.com or www.com 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, scanning or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP, UK, without the permission in writing of the Publisher. Requests to the Publisher should be addressed to the Permissions Department, John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England, or emailed to permreq@wiley.uk, or faxed to (+44) 1243 770620. 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. Other Wiley Editorial Offices John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany John Wiley & Sons Australia Ltd, 33 Park Road, Milton, Queensland 4064, Australia John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809 John Wiley & Sons Canada Ltd, 6045 Freemont Blvd, Mississauga, Ontario, L5R 4J3 Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.
Library of Congress Cataloging-in-Publication Data Bowers, David, 1938– Medical statistics from scratch : an introduction for health professionals / David Bowers. Includes bibliographical references and index. Medicine—Research—Statistical methods.72’7—dc22 2007041619 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN 978-0-470-51301-9 Typeset in 10/12pt Minion by Aptara Inc., New Delhi, India Printed and bound in Great Britain by Antony Rowe Ltd., Chippenham, Wilts This book is printed on acid-free paper responsibly manufactured from sustainable forestry in which at least two trees are planted for each one used for paper production. This book is for Susanne Contents Preface to the 2nd Edition xi Preface to the 1st Edition xiii Introduction xv I Some Fundamental Stuff 1 1 First things first – the nature of data 3 Learning Objectives 3 Variables and data 3 The good, the bad, and the ugly – types of variable 4 Categorical variables 4 Metric variables 7 How can I tell what type of variable I am dealing with? 9 II Descriptive Statistics 15 2 Describing data with tables 17 Learning Objectives 17 What is descriptive statistics? 17 The frequency table 18 3 Describing data with charts 29 Learning Objectives 29 Picture it! 29 Charting nominal and ordinal data 30 Charting discrete metric data 34 Charting continuous metric data 35 Charting cumulative data 37 4 Describing data from its shape 43 Learning Objectives 43 The shape of things to come 43 viii CONTENTS 5 Describing data with numeric summary values 51 Learning Objectives 51 Numbers R us 52 Summary measures of location 54 Summary measures of spread 57 Standard deviation and the Normal distribution 65 III Getting the Data 69 6 Doing it right first time – designing a study 71 Learning Objectives 71 Hey ho! Hey ho! It’s off to work we go 72 Collecting the data – types of sample 74 Types of study 75 Confounding 81 Matching 81 Comparing cohort and case-control designs 83 Getting stuck in – experimental studies 83 IV From Little to Large – Statistical Inference 91 7 From samples to populations – making inferences 93 Learning Objectives 93 Statistical inference 93 8 Probability, risk and odds 97 Learning Objectives 97 Chance would be a fine thing – the idea of probability 98 Calculating probability 99 Probability and the Normal distribution 100 Risk 100 Odds 101 Why you can’t calculate risk in a case-control study 102 The link between probability and odds 103 The risk ratio 104 The odds ratio 105 Number needed to treat (NNT) 106 V The Informed Guess – Confidence Interval Estimation 109 9 Estimating the value of a single population parameter – the idea of confidence intervals 111 Learning Objectives 111 Confidence interval estimation for a population mean 112 Confidence interval for a population proportion 116 Estimating a confidence interval for the median of a single population 117 CONTENTS ix 10 Estimating the difference between two population parameters 119 Learning Objectives 119 What’s the difference? 120 Estimating the difference between the means of two independent populations – using a method based on the two-sample t test 120 Estimating the difference between two matched population means – using a method based on the matched-pairs t test 125 Estimating the difference between two independent population proportions 126 Estimating the difference between two independent population medians – the Mann–Whitney rank-sums method 127 Estimating the difference between two matched population medians – Wilcoxon signed-ranks method 131 11 Estimating the ratio of two population parameters 133 Learning Objectives 133 Estimating ratios of means, risks and odds 133 VI Putting it to the Test 139 12 Testing hypotheses about the difference between two population parameters 141 Learning Objectives 141 The research question and the hypothesis test 142 A brief summary of a few of the commonest tests 144 Some examples of hypothesis tests from practice 146 Confidence intervals versus hypothesis testing 149 Nobody’s perfect – types of error 149 The power of a test 151 Maximising power – calculating sample size 152 Rules of thumb 152 13 Testing hypotheses about the ratio of two population parameters 155 Learning Objectives 155 Testing the risk ratio 155 Testing the odds ratio 158 14 Testing hypotheses about the equality of population proportions: the chi-squared test 161 Learning Objectives 161 Of all the tests in all the world.
the chi-squared (χ 2 ) test 162 VII Getting up Close 169 15 Measuring the association between two variables 171 Learning Objectives 171 Association 171 The correlation coefficient 175 x CONTENTS 16 Measuring agreement 181 Learning Objectives 181 To agree or not agree: that is the question 181 Cohen’s kappa 182 Measuring agreement with ordinal data – weighted kappa 184 Measuring the agreement between two metric continuous variables 184 VIII Getting into a Relationship 187 17 Straight line models: linear regression 189 Learning Objectives 189 Health warning! 190 Relationship and association 190 The linear regression model 192 Model building and variable selection 200 18 Curvy models: logistic regression 213 Learning Objectives 213 A second health warning! 213 Binary dependent variables 214 The logistic regression model 215 IX Two More Chapters 225 19 Measuring survival 227 Learning Objectives 227 Introduction 227 Calculating survival probabilities and the proportion surviving: the Kaplan-Meier table 228 The Kaplan-Meier chart 230 Determining median survival time 231 Comparing survival with two groups 232 20 Systematic review and meta-analysis 239 Learning Objectives 239 Introduction 240 Systematic review 240 Publication and other biases 244 The funnel plot 244 Combining the studies 246 Appendix: Table of random numbers 251 Solutions to Exercises 253 References 273 Index 277 Preface to the 2nd Edition This book is a ‘not-too-mathematical’ introduction to medical statistics. It should appeal to anyone training or working in the health care arena – whatever their particular discipline – who wants either a simple introduction to the subject, or a gentle reminder of stuff they might have forgotten. I have aimed the book at: r Students doing a first degree or diploma in clinical and health care courses. r Students doing post-graduate clinical and health care studies.
r Health care professionals doing professional and membership examinations. r Health care professionals who want to brush up on some medical statistics generally, or who want a simple reminder of a particular topic. r Anybody else who wants to know a bit of what medical statistics is about. The most significant change in this second edition is the addition of two new chapters, one on measuring survival, and one on systematic review and meta-analysis.
The ability to understand the principles of survival analysis is important, not least because of its popularity in clinical research, and consequently in the clinical literature. Similarly, the increasing importance of evidence-based clinical practice means that systematic review and meta-analysis also demand a place. In addition, I have taken the opportunity to correct and freshen the text in a few places, as well as adding a small number of new examples. My thanks to Lucy Sayer, my editor at John Wiley, for her enthusiastic support, to Liz Renwick and Robert Hambrook, and all the other wiley people, for their invaluable help and special thanks to my copy-editor Barbara Noble, for her truly excellent work and enthusiasm (of course, any remaining errors are mine).
I am happy to get any comments and criticisms from you. You can e-mail me at: slothist@hotmail. Preface to the 1st Edition This book is intended to be an introduction to medical statistics but one which is not too mathematical—in fact has the absolute minimum of maths. The exceptions however are Chap- ters 17 and 18, on linear and logistic regression.
It’s really impossible to provide material on these procedures without some maths, and I hesitated about including them at all. However they are such useful and widely used techniques, particularly logistic regression and its pro- duction of odds ratios, that I felt they must go in. Of course you don’t have to read them. It should appeal to anyone training or working in the health care arena—whatever their particular discipline—who wants a simple, not-too-technical introduction to the subject.
I have aimed the book at: r students doing either a first degree or diploma in health care-related courses r students doing postgraduate health care studies r health care professionals doing professional and membership examinations r health care professionals who want to brush up on some medical statistics generally, or who want a simple reminder of one particular topic r anybody else who wants to know a bit of what medical statistics is about. I intended originally to make this book an amalgam of two previous books of mine, Statistics from Scratch for Health Care Professionals and Statistics Further from Scratch. However, although it covers a lot of the same material as in those two books, this is in reality a completely new book, with a lot of extra stuff, particularly on linear and logistic regression. I am happy to get any comments and criticisms from you.
You can e-mail me at: slothist@hotmail. Introduction Before the spread of personal computers, researchers had to do most things by hand (by which I mean with a calculator), and so most statistics books were full of equations and their derivations, with many pages of the necessary statistical tables. Analysing anything other than small samples could be time-consuming and error prone. You also needed to be reasonably good at maths.
Of course, for the statistics specialist there is still a need for books that deal with statistical theory, and the often complex mathematics which underlies the subject.