Medical Statistics from Scratch E-Book

David Bowers

• Herausgeber: John Wiley & Sons
• Kategorie: Fachliteratur
• Sprache: Englisch

Seitenzahl: 740

Veröffentlichungsjahr: 2019

Beschreibung

Correctly understanding and using medical statistics is a key skill for all medical students and health professionals. In an informal and friendly style, Medical Statistics from Scratch provides a practical foundation for everyone whose first interest is probably not medical statistics. Keeping the level of mathematics to a minimum, it clearly illustrates statistical concepts and practice with numerous real-world examples and cases drawn from current medical literature. Medical Statistics from Scratch is an ideal learning partner for all medical students and health professionals needing an accessible introduction, or a friendly refresher, to the fundamentals of medical statistics.

Leseprobe

Medical Statistics from Scratch

An Introduction for Health Professionals

Fourth Edition

David Bowers

Visiting Research FellowLeeds Institute of Health SciencesSchool of MedicineUniversity of LeedsLeeds, UK

This edition first published 2020© 2020 John Wiley and Sons Ltd

Edition HistoryJohn Wiley and Sons (3e, 2014)

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 law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.

The rights of David Bowers to be identified as the author of this work has been asserted in accordance with law.

Registered Office(s)John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USAJohn Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

Editorial Office9600 Garsington Road, Oxford, OX4 2DQ, UK

For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.

Wiley also publishes its books in a variety of electronic formats and by print‐on‐demand. Some content that appears in standard print versions of this book may not be available in other formats.

Limit of Liability/Disclaimer of WarrantyThe contents of this work are intended to further general scientific research, understanding, and discussion only and are not intended and should not be relied upon as recommending or promoting scientific method, diagnosis, or treatment by physicians for any particular patient. In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of medicines, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each medicine, equipment, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

Library of Congress Cataloging‐in‐Publication Data

Names: Bowers, David, 1938– author.Title: Medical statistics from scratch : an introduction for health professionals / David Bowers.Description: 4th edition. | Hoboken NJ : Wiley, 2020. | Includes bibliographical references and index. |Identifiers: LCCN 2019015126 (print) | LCCN 2019015564 (ebook) | ISBN 9781119523925 (Adobe PDF) | ISBN 9781119523949 (ePub) | ISBN 9781119523888 (pbk.)Subjects: | MESH: Biometry | Statistics as TopicClassification: LCC RA409 (ebook) | LCC RA409 (print) | NLM WA 950 | DDC 610.72/7–dc23LC record available at https://lccn.loc.gov/2019015126

Cover Design: WileyCover Image: Courtesy of David Bowers

Preface to the 4th Edition

I noticed when looking at the prefaces to each of the first three editions, that there seems to be a gap of five years between each one. And this fourth edition is no exception – it’s five years since I wrote the preface to the third edition, in 2013. Spooky or what! But anyway, this new edition has given me an opportunity to completely refresh the book. I have added three shiny new chapters covering important areas in medical statistics:

On incidence, prevalence and standardisation.

On Poisson regression.

And on missing data.

I have also made numerous additions in most of the other chapters. These include, for example:

In the chapter on measures of spread, I have added material on testing for Normality using the Shapiro–Wilk and Kolmogorov–Smirnov tests, as well as the graphical QQ plot.

In the chapter on describing data with charts, I have added more material on the scatterplot, and introduced the bubble plot.

In the second of the chapters on research design I have added brief notes on

per protocol

analysis.

In the confounding chapter I have introduced the idea of confounding by indication, residual confounding, and propensity score matching.

In the chapter on comparing population parameters using confidence intervals, I have revisited the interpretation of confidence intervals and the size of corresponding samples to try and make it easier to understand. I wanted to emphasise that we should take a flexible approach to the interpretation of confidence intervals and p‐values.

In the hypothesis chapter I have introduced the notion of correcting for multiple comparisons.

In the logistic regression chapter I have added new material on goodness‐of‐fit to include the Cox and Snell, and Nagelkerke’s

pseudo

goodness‐of‐fit measures.

In the chapter on systematic review I have added a brief discussion of two statistical tests of funnel plot asymmetry: Begg’s test and Egger’s test. I have also added a little something on the trim and fill method for dealing with funnel plot asymmetry.

Apart from the above additions, I have added many new examples and exercises, and have rewritten much of the existing text to (hopefully) make it easier and more accessible.

Finally, I want to thank all of those who helped me in the writing of this book, and Barbara Noble in particular for bravely (and very competently) copy‐editing and proofreading the material. Any remaining errors are of course mine.

As always I am happy to receive comments and criticism of my book, and I welcome any suggestions for improvement.

“Reader, I married him.” Whoops! Wrong book.

David Bowers, 2019

Preface to the 3rd Edition

The 1st edition of this book was published in 2002 and the 2nd edition in 2008. I was surprised when I discovered it was quite such a long time ago. Where did the time go! Anyway, over the course of the last five years, I have received many favourable comments from readers of my book, which of course is immensely gratifying. I must be doing something right then.

This edition contains a completely new chapter (on diagnostic tests), there is a quite a lot of new material and most of the chapters have received an extensive re‐write. I have also updated virtually all of the examples drawn from the journals and added many new exercises. I hope that this gives the book a fresh feel – as well as a new lease of life.

The book should appeal, as before, to everybody in health care (students and professionals alike) including nurses, doctors, health visitors, physiotherapists, midwives, radiographers, dieticians, speech therapists, health educators and promoters, chiropodists and all those other allied and auxiliary professionals. It might possibly also be of interest to veterinary surgeons, one of whom reviewed my proposal fairly enthusiastically.

My thanks to Jon Peacock and all the others at Wiley who have shepherded me along in the past and no doubt will do so in the future. I must also thank Barbara Noble, who patiently acted as my first‐line copyeditor. She read through my manuscript, discovered quite a few errors of various sorts and made many valuable suggestions to improve readability. Any remaining mistakes are of course mine.

I also want to acknowledge my great debt to Susanne, who always encourages me, enthusiastically, in everything I attempt.

Finally, I would like to mention another book which might be of interest to any readers who are thinking of embarking on research for the first time – Getting Started in Health Research, Bowers et al., Wiley, 2012. This book covers both quantitative and qualitative research. It will guide you through the research process, from the very first idea to the interpretation of your results and your conclusions.

David Bowers, 2013

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 his or her particular discipline is – who wants either a simple introduction to the subject or a gentle reminder of stuff that they might have forgotten. I have aimed the book at:

students doing either a first degree or a diploma in clinical and health care courses

students doing post‐graduate clinical and health care studies

health care professionals doing professional and membership examinations

health care professionals who want to brush up on some medical statistics generally or who want a simple reminder of a particular topic

anybody else who wants to know a bit of what medical statistics is about.

The most significant change in this edition is the addition of two new chapters, one on measuring survival and the other 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 & Sons, for her enthusiastic support, to Liz Renwick and Robert Hambrook and all the other people in Wiley for their invaluable help and my special thanks to my copyeditor Barbara Noble for her truly excellent work and enthusiasm (of course, any remaining errors are mine).

I am happy to get any comments from you. You can e‐mail me at: [email protected].

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, it has the absolute minimum of maths. The exceptions however are Chapters 17 and 18, which have maths on linear and logistic regressions. It is 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 production of odds ratios, which I felt they must go in. Of course, you do not have to read them. It should appeal to anyone training or working in the health care arena – whatever his or her particular discipline is – who wants a simple, not‐too‐technical introduction to the subject. I have aimed the book at:

students doing either a first degree or a diploma in health care‐related courses

students doing post‐graduate health care studies

health care professionals doing professional and membership examinations

health care professionals who want to brush up on some medical statistics generally or who want a simple reminder of a particular topic

anybody else who wants to know a bit of what medical statistics is about.

I intended originally to make this book as 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 regressions. I am happy to get any comments and criticisms from you. You can e‐mail me at: [email protected].

Introduction

My purpose in writing this book was to provide a guide to all those health care students and professionals out there who either want to get started in medical statistics or who would like (or need) to refresh their understanding of one or more of the topics which are common in medical statistics. I have tried to keep the mathematics to a minimum, although this is a bit more difficult with the somewhat challenging material on modelling in later chapters.

I have used lots of appropriate examples drawn from clinical journals to illustrate the ideas, as well as many relevant exercises which the readers may wish to work through to consolidate their understanding of the material covered (the solutions are at the end of this book).

I have included some outputs from SPSS and Minitab which I hope will help the readers interpret the results from these statistical programs.

Finally, for any tutors who are using this book to introduce their students to medical statistics, I am always very pleased to receive any comments or criticisms they may have which will help me improve the book for any possible future editions. My e‐mail address is: [email protected].

Enjoy!

David Bowers

ISome Fundamental Stuff

1First things first – the nature of data

Learning objectives

When you have finished this chapter, you should be able to:

Explain the difference between nominal, ordinal, and metric data.

Identify the type of any given variable.

Explain the non‐numeric nature of ordinal data.

Variables and data

Let’s start with some numbers. Have a look at Figure 1.1.

Figure 1.1 Some numbers. Actually, the birthweight (g) of a sample of 100 babies.

Source: Data from the Born in Bradford cohort study (Born in Bradford, 2012).

These numbers are actually the birthweights of a sample of 100 babies (measured in grams). We call these numbers sample data. These data arise from the variable birthweight. To state the blindingly obvious, a variable is something whose value can vary. Other variables could be blood type, age, parity, and so on; the values of these variables can change from one individual to another. When we measure a variable, we get data – in this case, the variable birthweight produces birthweight data.

Figure 1.2 contains more sample data, in this case, for the gender of the same 100 babies.

Figure 1.2 The gender of the sample of babies in Figure 1.1.

Moreover, Figure 1.3 contains sample data for the variable smoked while pregnant.

Figure 1.3 The variable “smoked while pregnant?” for the mothers of the babies in Figure 1.1.

The data in Figures 1.1–1.3 are known as raw data because they have not been organised or arranged in any way. This makes it difficult to see what interesting characteristics or features the data might contain. The data cannot tell its story, if you like. For example, it is not easy to observe how many babies had a low birthweight (less than 2500 g) from Figure 1.1, or what proportion of the babies were female from Figure 1.2. Moreover, this is for only 100 values. Imagine how much more difficult it would be for 500 or 5000 values. In the next four chapters, we will discuss a number of different ways that we can organise data so that it can tell its story. Then, we can see more easily what’s going on.

Exercise 1.1. Why do you think that the data in Figures 1.1–1.3 are referred to as “sample data”?

Exercise 1.2. What percentage of mothers smoked during their pregnancy? How does your value contrast with the evidence which suggests that about 20% of mothers in the United Kingdom smoked when pregnant?

Of course, we gather data not because it is nice to look at or we’ve got nothing better to do but because we want to answer a question. Such as “Do the babies of mothers who smoked while pregnant have a different (we’re probably guessing lower) birthweight than the babies of mothers who did not smoke?” or “On average, do male babies have the same birthweight as female babies?” Later in the book, we will deal with methods which you can use to answer such questions (and ones more complex); however, for now, we need to stick with variables and data.

Where are we going …?

This book is an introduction to medical statistics.

Medical statistics is about doing things with data.

We get data when we determine the value of a variable.

We need data in order to answer a question.

What we can do with data depends on what type of data it is.

The good, the bad, and the ugly – types of variables

There are two major types of variable – categorical variables and metric variables; each of them can be further divided into two subtypes, as shown in Figure 1.4.

Figure 1.4 Types of variables.

Each of these variable types produces a different type of data. The differences in these data types are of great importance – some statistical methods are appropriate for some types of data but not for others, and applying an inappropriate procedure can result in a misleading outcome. It is therefore critical that you identify the sort of variable (and data) you are dealing with before you begin any analysis, and we need therefore to examine the differences in data types in a bit more detail. From now on, I will be using the word “data” rather than “variable” because it is the data we will be working with – but remember that data come from variables. We’ll start with categorical data.

Categorical data

Nominal categorical data

Consider the gender data shown in Figure 1.2. These data are nominal categorical data (or just nominal data for short).

The data are “nominal” because it usually relates to named things, such as occupation, blood type, or ethnicity. It is particularly not numeric. It is “categorical” because we allocate each value to a specific category. Therefore, for example, we allocate each M value in Figure 1.2 to the category Male and each F value to the category Female. If we do this for all 100 values, we get:

Male

56

Female

44

Notice two things about this data, which is typical of all nominal data:

The data do not have any units of measurement.

1

The ordering of the categories is

arbitrary

. In other words, the categories cannot be ordered in any meaningful way.

2

We could just as easily have written the number of males and females in the order:

Female

44

Male

56

By the way, allocating values to categories by hand is pretty tedious as well as error‐prone, more so if there are a lot of values. In practice, you would use a computer to do this.

Exercise 1.3. Suggest a few nominal variables.

Ordinal categorical data

Let’s now consider data from the Glasgow Coma Scale (GCS) (which some of you may be familiar with). As the name suggests, this scale is used to assess the level of consciousness after head injury. A patient’s GCS score is judged by the sum of responses in three areas: eye opening response, verbal response, and motor response. Notice particularly that these responses are assessed