Qualitative methods for health economics

Qualitative Methods
for Health Economics



Qualitative Methods
for Health Economics
Edited by Joanna Coast

London • New York


Published by Rowman & Littlefield International Ltd
Unit A, Whitacre Mews, 26-34 Stannary Street, London SE11 4AB
www.rowmaninternational.com
Rowman & Littlefield International Ltd. is an affiliate of Rowman & Littlefield
4501 Forbes Boulevard, Suite 200, Lanham, Maryland 20706, USA
With additional offices in Boulder, New York, Toronto (Canada), and Plymouth (UK)
www.rowman.com
Selection and editorial matter © Joanna Coast 2017
Copyright in individual chapters is held by the respective chapter authors.
All rights reserved. No part of this book may be reproduced in any form or by any
electronic or mechanical means, including information storage and retrieval systems,
without written permission from the publisher, except by a reviewer who may quote
passages in a review.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
ISBN:

HB 978-1-7834-8561-1
PB 978-1-7834-8562-8

Library of Congress Cataloging-in-Publication Data
Names: Coast, Joanna, editor.
Title: Qualitative methods for health economics / edited by Joanna Coast.
Description: London ; New York : Rowman & Littlefield International Ltd., [2017] |
Includes bibliographical references and index.
Identifiers: LCCN 2017010823 (print) | LCCN 2017011960 (ebook) |
ISBN 9781783485635 (ebook) | ISBN 9781783485611 (cloth) |
ISBN 9781783485628 (paper)
Subjects: | MESH: Qualitative Research | Economics, Medical |
Research Design | Evaluation Studies as Topic
Classification: LCC RA410.55.E83 (ebook) | LCC RA410.55.E83 (print) |
NLM W 20.5 | DDC 362.1068/1072—dc23
LC record available at />The paper used in this publication meets the minimum requirements of
American National Standard for Information Sciences—Permanence of Paper
for Printed Library Materials, ANSI/NISO Z39.48-1992.
Printed in the United States of America


This book is for Steve, Iain, Ned and Alfie, with all
my love, and thanks for much patience and many
cups of tea while I was putting it together!



Contents

List of Figures

xi

List of Tables

xiii

List of Boxes

xv

Acknowledgementsxvii
PART I:  CHOICES ABOUT METHODS

1

 1  Theoretical and methodological positions and the choice
to use qualitative methods
Joanna Coast and Louise Jackson

3

 2  Designing qualitative research studies in health economics
Joanna Coast, Philip Kinghorn and Amanda Owen-Smith

19

 3  Understanding sampling and recruitment
Amanda Owen-Smith and Joanna Coast

42

 4  Understanding data collection: Interviews, focus groups
and observation
Amanda Owen-Smith and Joanna Coast
 5  Understanding primary data analysis
Joanna Coast and Louise Jackson
 6  Understanding secondary data analysis
Hareth Al-Janabi and Amanda Owen-Smith

vii

59
92
119


viii

Contents

 7  Presenting and publishing qualitative research
in health economics
Joanna Coast

139

PART II:  CASE STUDIES

153

 8  Introduction to the case studies
Joanna Coast

155

PART II A: USING QUALITATIVE METHODS
TO DEVELOP AND ENHANCE
QUANTITATIVE RESEARCH
 9  Constructing statement sets for use in Q methodology studies
Rachel Baker, Neil McHugh and Helen Mason
10  Qualitative methods in discrete choice experiments:
Two case studies
Jane Vosper, Joanna Coast and Terry Flynn
11  Using qualitative methods to develop a preference-based
health-related quality of life measure for use in economic
evaluation: The development of the Child Health Utility 9D
Katherine Stevens
12  Incorporating novel qualitative methods within health
economics: The use of pictorial tools
Alastair Canaway, Hareth Al-Janabi, Philip Kinghorn,
Cara Bailey and Joanna Coast
13  Using non-participant observation and think-aloud to
understand and improve modelling processes
Samantha Husbands, Sue Jowett, Pelham Barton
and Joanna Coast
PART II B: USING QUALITATIVE METHODS
TO INTERPRET QUANTITATIVE FINDINGS

161

163

175

193

205

217

229

14  Using ‘think-aloud’ and interview data to explore patient
and proxy completion of health and capability measures
at the end of life
Cara Bailey, Philip Kinghorn, Rosanna Orlando
and Joanna Coast

231

15  Getting beyond the numbers: Using qualitative methods to
explore the validity of health state valuation techniques
Suzanne Robinson

245




Contentsix

16  Use of deliberative methods to facilitate and enhance
understanding of the weighting of survey attributes
Philip Kinghorn, Alastair Canaway, Cara Bailey
and Joanna Coast
PART II C: USING QUALITATIVE METHODS
TO ENHANCE AND GENERATE
ECONOMIC UNDERSTANDING
17  Using qualitative methods in impact evaluation: The case
of the results-based financing for maternal and neonatal
care impact evaluation in Malawi
Manuela De Allegri
18  An assessment of an international declaration on aid
effectiveness through qualitative methods at the
country level
Melisa Martínez-Álvarez
19  Using cost-effectiveness evidence in making priority
gradings: The case of the Swedish National Guidelines
for heart diseases
Nathalie Eckard and Ann-Charlotte Nedlund

257

275

277

293

307

20  Contributions of qualitative methods to the study of priority
setting and resource allocation in health care
Neale Smith

319

PART III: AFTERWORD 

333

21  Afterword: Walking the disciplinary tightrope
Joanna Coast, Hareth Al-Janabi, Louise Jackson,
Philip Kinghorn and Amanda Owen-Smith

335

References339
Index377
Contributors385



List of Figures

  9.1. Q grid and sorting instruction

165

  9.2. An illustration of Q factor analysis

166

12.1. Example of completed hierarchical map

212

16.1. Recruitment process for public

263

16.2. Task one example question

265

16.3. Patients versus close persons (task three)

266

16.4. Number of participants recording a change on one
or more tasks

269

18.1. Example of analysis of the theme ‘mutual accountability’

302

19.1. Schematic description of the work process for the National
Guidelines (Socialstyrelsen 2008)

310

xi



List of Tables

  3.1. Extract from AOS doctoral thesis: qualitative sampling
in practice

44

  5.1. Example definitions associated with first- and second-level
analytic categories for citizen–agent example (Coast 2000b)

111

  6.1. Reciprocal translation of selected themes across two
qualitative studies

123

  6.2. Qualitative data repositories currently available

126

  6.3. Translation of the ‘concept’ of control into individual
studies in a meta-ethnography of six studies of the

experience of informal care (Al-Janabi 2009)

132

  9.1. Sources of statements for study of values and life extension
and terminal illness

169

  9.2. Example of Q set design matrix

172

10.1. Attributes and levels included in the TREAD discrete
choice experiment

183

10.2. Attributes and levels included in the drug treatment discrete
choice experiment

189

16.1. Coding and the emergence of themes

268

18.1. Key informant and in-depth interview participants

298

19.1. Identified themes for the use of cost-effectiveness evidence
in priority gradings

316

xiii



List of Boxes

  4.1. Example topic guide for patient interview

64

  4.2. Example topic guide for citizen interview

65

  4.3. Example of a distress protocol

89

  5.1. Excerpt from first analytic account for citizen informant,
based on first four interviews transcribed

113

  5.2. Excerpt from final analytic account for citizen informant,

based on all interviews

115

14.1. Stages of the think-aloud interview

236

15.1. Details of the main valuation study

246

15.2. Outline of the interview schedule

251

xv



Acknowledgements

The inspiration for this book on the use of qualitative methods in health economics came about through discussions with Alison Howson, and I would
like to thank her for that and for her backing for the book. I would also like
to thank two other publishing staff for their assistance later in the project:
Martina O’Sullivan and Michael Watson.
There are a number of people, institutions and funding bodies that I would
like to thank. I would particularly like to thank Jenny Donovan, who introduced me to qualitative research and provided invaluable mentoring during
my early work in this area. Over the years I have enjoyed working with, and
learning from, a number of researchers using qualitative methods in health

economics. I would like to thank them all for their enthusiasm and persistence
in working in an area that falls between disciplines, but particularly Hareth
Al-Janabi, Cara Bailey, Alastair Canaway, Oya Eddama, Sue Horrocks, Sam
Husbands, Tom Keeley, Phil Kinghorn, Amanda Owen-Smith, Rosanna
Orlando, Eileen Sutton and Jane Vosper.
This book was written while I was based in two universities, University of
Bristol and University of Birmingham, and I would like to thank both institutions for their support and encouragement.
Acknowledgements for funding bodies for the individual case studies are
given at the end of each chapter, where applicable. A number of organizations
also require thanks for funding the different qualitative research projects that
are discussed in the first part of the book, including the Economic and Social
Research Council, European Research Council, Medical Research Council,
National Institute for Health Research and the South and West Research and
Development Directorate.

xvii


xviii

Acknowledgements

I would like to thank the International Society for the Scientific Study of
Subjectivity for permission to reproduce Figure 9.2.
Finally, I would like to offer my greatest thanks to all of the chapter contributors for their hard work and the informants who participated in the various research studies without whom there would be no qualitative research
to report!


Part I

CHOICES ABOUT METHODS



Chapter 1

Theoretical and methodological
positions and the choice to
use qualitative methods
Joanna Coast and Louise Jackson

It is assumed that the main audience for this book comprises (health) economists with an interest in using qualitative methods in their own research; this
chapter therefore begins the book with a relatively accessible discussion of
issues around ontology, epistemology and methodology. The nature of training in economics (Lawson 1997; Lee, Pham, et al. 2013) means that many
economists reading this book may be unaware of these terms and will not
have thought about these issues previously (Coast 1999). Indeed, when publishing her first article exploring these issues in a health economics journal in
the late 1990s, Coast was asked to include a glossary detailing these terms,
as they were deemed to be terms that health economists would generally
be unfamiliar with (Coast 1999). It has been noted that economists’ training does not tend to provide an understanding of data collection processes
or the role of qualitative methods in complementing quantitative research,
with Berik stating that ‘such training would be an eye-opening and humbling
experience – an essential antidote for arrogance – for economists who do not
question their data’ (Berik 1997, p. 124). Those with an existing sophisticated
understanding (e.g., qualitative researchers reading the book because they are
working with economists) may want to read this chapter not so much for its
content but to become aware of the different understandings of ontology and
epistemology that the economists they are working with may have.
This chapter begins by considering theories of knowledge and their application in economics, drawing on Coast’s earlier paper (Coast 1999). There is
much work of an explanatory nature examining issues relating to the nature
of health care systems as well as health more generally. The vast majority of

this work is conducted within the neoclassical framework (indeed Murphy
and colleagues have noted the ‘narrowness of the range of economic thought
that is currently being deployed on the problems of health technology and
3


4

Chapter 1

health services’ [Murphy, Dingwall, et al. 1998, p. viii]) although explanatory work can also be conducted within alternative frameworks. Normative
(welfare or extra-welfare), institutional, behavioural and feminist economics
are all briefly considered, as is the potential for approaches based on critical
realism. The chapter then details the nature of qualitative research methods
as well as the constructivist philosophy with which they are most frequently
associated. From this discussion it is apparent that such a philosophy is very
different from that usually associated with the discipline of economics. The
nature of enquiry associated with qualitative research is briefly examined,
covering details of specific data collection methods, sampling, analysis of
data and ensuring rigour. Despite differences in philosophy from mainstream
economics, it is apparent that qualitative methods are useful to economists,
and the use of qualitative methods in health economics is then examined. It
is argued that robust qualitative research can provide scientifically valuable
contributions to health economic knowledge both through improving quantitative data collection, such as when generating attributes for discrete choice
attributes, and through contributing directly to economic theory, particularly
in complex areas such as priority setting or developing financial incentives.
UNDERSTANDING RELEVANT TERMINOLOGY
Health economists embarking on qualitative research first need some understanding about the ways in which different types of researchers understand
the world. Many are likely to have a view similar to that of Coast when she
started work as a junior health economist in 1990. She had completed a first

degree in economics and a master’s degree in health economics, and throughout these degrees, theories had been largely presented either as self-evident,
and thus without the need for any supporting evidence, or in terms of economists developing hypotheses based on deductive thinking and then testing
these hypotheses using quantitative data.
An example of the former was the basic underlying principles of demand
and supply, and the elaborations on these such as Giffen goods. These were
presented as clear, logical and internally consistent interpretations of how the
world works. But there was little in the way of evidence for many of these
theories, perhaps underlining the views of some that economics is largely a
branch of applied mathematics, with little relationship to the world outside
(Rosenberg 1992).
An example of the latter might be the theories of supplier-induced demand,
where economists have developed theories about how medical physicians
might act in ways that shift the demand curve for health interventions from its
‘true’ level (the level that fully informed consumers might choose), perhaps




Theoretical and methodological positions5

to achieve higher income for themselves (Evans 1984). Hypotheses are then
generated based on these theories, and various quantitative econometric
techniques might then be utilized to test these theories. For example, the
hypothesis might be around physicians not being able to achieve their target
income if more doctors arrive in an area and thus choose then to increase their
provision of services to achieve this income. This might then be tested using
quantitative information about the levels of consumption of medical services
alongside data about numbers of physicians working in a location (Evans
1980). The hypothesis can then be accepted or rejected.
In applying economics to health, health economists constantly come up

against the problem that theories developed to explain ‘the economy’ (with
varying degrees of success and with varying degrees of acknowledgement
of that varying success) will very often not apply readily and simply to the
‘health economy’. For Coast, her first job, as a lone health economist in a university department of epidemiology and public health in the United Kingdom,
made this abundantly clear, as her relatively ‘neat’ views of the world did not
reflect those of her new non-economist colleagues, nor as she could quickly
see for herself the complexities of health care provision in the UK health
economy. There was clearly a need for something more, and qualitative methods seemed to have potential in achieving a better understanding of the health
economy and how it works. Even a small foray into this different research
paradigm, however, suggested a very different perception of the world.
UNDERSTANDING ONTOLOGY, EPISTEMOLOGY
AND METHODOLOGY IN (HEALTH) ECONOMICS
Ontology means the view of the nature and form of reality (Lincoln 1992).
Epistemology is concerned with the theory and nature of knowledge, in
particular the relationship between what is to be known and the researcher
(Lincoln 1992). There are very different views of the nature and form of reality, although many health economists are unlikely to have given these issues
much thought. Indeed, they may well assume that their own view of reality
is universally shared because of the way in which the nature of the discipline
is conveyed.
As suggested in an earlier paper (Coast 1999), most health economics
explanatory research is achieved within an orthodox neoclassical framework.
The standard view of reality within this ‘positive’ or ‘explanatory’ economics
has had a number of influences including positivism, the notion that there is
a single reality that can be researched (Hollis 1994), and the falsificationism
of Popper (Blaug 1992; Hausmann 1994). Essentially, within this world view,
there is a belief that there is a single, knowable reality and that research can


6

Chapter 1

help to find out about this reality. This reality comprises a set of rules that
it is possible to determine (Lawson 1997; Graça Moura and Martins 2008).
Methodology is concerned with the means by which the person inquiring
tries to find out what is to be known, the practice of research (Lincoln 1992;
Creswell 2013). Methods of investigation within explanatory neoclassical
economics are long standing (Mill 1994) and start from the individual; that
is, assumptions about ‘individual’ preferences, both from consumers and
from firms, are used to generate the behavioural assumptions from which
the economic theory is built. This is known as methodological individualism
(Hodgson 1986; Dugger 1994; Toboso 2001). In an explanatory research conducted within a neoclassical paradigm, the economist starts with a set of these
assumptions, referred to as axioms, which are high-level principles that are
regarded as being established and accepted (although the accuracy of these
starting assumptions is often seen as less important than the resultant predictions [Friedman 1994]). From these axioms, often concerned with rationality
or optimization (Lawson 2006), the economist can infer logically, through
a deductive process, what might be predicted, and these predictions can be
tested; these predictions then form the basis for explanation.
The testing of these assumptions draws on a particular view of the relationship between the researcher and what he or she is researching, in which it is
assumed that the economist tries to be objective. The economist develops and
tests hypotheses without allowing himself or herself to influence the findings
or be influenced by these findings, so that he or she can confidently accept or
reject the hypothesis/hypotheses that have been tested and thus confirm (or
not) the accuracy of the theory generated. In general, at least theoretically,
it is seen as better to try to test whether hypotheses can be falsified (Blaug
1992). Although the logical sequelae of falsifying a theory is to reject it and
move on to generate new theories and test these, it has been noted previously
that in economics this may not be the ultimate outcome (Coast 1999), with
views expressed that economists may instead try to refine their mathematical
expressions (Ormerod 1994), alter assumptions and try to bring more variables into the explanation (Lawson 1997). Indeed, in relation to the example

of supplier-induced demand used earlier, it is notable that there has been difficulty in reaching clear conclusions (Labelle, Stoddart, et al. 1994) despite
the problem being formulated in a number of different ways.
In testing hypotheses, the aim is to eliminate bias and obtain true knowledge
about how reality works. Options for achieving this include the randomized
controlled trial with which most health economists will be familiar, given the
place of this method within health research as the ‘gold standard’ basis for
determining the effectiveness of health interventions (Bennett 2007). Indeed,
many health economists will have particular familiarity with this method
through their conduct of economic evaluation alongside such randomized