Research for Evidence-Based Practice in Healthcare E-Book

Robert Newell

• Herausgeber: John Wiley & Sons
• Kategorie: Fachliteratur
• Serie: Vital Notes for Nurses
• Sprache: Englisch

Seitenzahl: 510

Veröffentlichungsjahr: 2011

Beschreibung

This is an essential, accessible introduction to the practicalities of research and evidence-based practice aimed at all pre-registration nursing and healthcare students. It places research and evidence in the context of clinical practice, introduces the main methodological approaches in qualitative and quantitative research, and describes the processes of research appraisal, dissemination and implementation.

The new edition of Research for Evidence-Based Practice in Healthcare has been updated to include information for a broader health care audience. It engages students with the research and evidence agenda, demonstrates the relevance of research and evidence to nursing practice, and provides the skills needed to explore these areas in greater detail.

Special features:

• A practical guide to research methods and evidence-based practice
• New edition of a successful student textbook
• Includes a glossary of common research terms
• Provides case studies, key points, further reading, and activities throughout
• Accompanying website with links to further reading

Leseprobe

Contents

List of Common Research Terms

SECTION 1 CONTEXTUAL MATERIALS

1 Introduction to Healthcare Research for Evidence-Based Practice

Introduction

The scope of healthcare research

Whose business is research in healthcare?

Using this book to get involved in healthcare research

The chapters

Notes on person and gender

2 The Research Process – Organising Your Research

Introduction

The title

The abstract

Aims and objectives

A review of the literature

The sample

Measures and materials

Data collection

Data analysis

Ethical issues

Financial issues

Writing the report

Publishing the findings

Timetable

Other issues

Organising your own research

3 Choosing Methodological Approaches

Introduction

When is a methodological approach fit for purpose?

Two different approaches to knowledge

Different broad methodological approaches and their appropriateness

What sorts of questions for what general approaches?

Do we need to do research at all?

4 Searching the Literature

Introduction

Purpose of literature searches

Systematic reviews

Search process

Designing a search strategy

Components of a search strategy

Retrieval and inclusion criteria

Resources

Reporting your search

5 Ethics of Healthcare Research

Introduction

Ethics and patient and public involvement in research

Ethics committees and research governance

What can you do?

6 Basic Concepts: Sampling, Reliability and Validity

Sampling in research

Samples and populations

Sampling as an everyday pursuit

Samples and representativeness

Validity and reliability

Validity, reliability and sampling

Horses for courses

SECTION 2 QUALITATIVE APPROACHES

7 Issues in Qualitative Data Collection

Introduction

Sampling for qualitative research

Could qualitative researchers use randomised samples?

Interviews and focus groups

Semi-structured interviews

Unstructured interviews

Other issues concerning interviews

Focus groups

Transcription

Problems with the interview method

Questionnaires

Observation

Published work

8 Case Studies

Introduction

The case study as ‘illustration’

The case study as a research method

Drawbacks

Endnote

9 Ethnography

Introduction

The stages of an ethnographic study

General methods used in ethnographic research

What should be observed?

Analysis

10 Phenomenology

Introduction

Philosophical background

Applications

Methods

11 A Pragmatic Approach to Qualitative Data Analysis

Introduction

Grounded theory

Thematic content analysis

A pragmatic approach to schematic content analysis

Conclusion

12 Limitations of Qualitative Research

Introduction

Reliability and validity

Generalisability

The nature of human beings

Data analysis

SECTION 3 QUANTITATIVE APPROACHES

13 Sampling, Reliability and Validity Issues in Data Collection and Analysis

Introduction

Representativeness, fairness and random sampling

Stratification

Cluster sampling

Quota samples

Systematic samples

Convenience/accidental samples

Sampling, representativeness and external validity

Internal validity

Reliability

Reliability of measures

Validity of measures

Conclusion

14 Cause and Effect, Hypothesis Testing and Estimation

Introduction

Cause and effect relationships in daily life

Examining cause and effect in healthcare

Variables

Hypotheses

Experimental and null hypotheses

Directional and non-directional hypotheses

Grounds for rejecting and accepting hypotheses

Power to reject the experimental hypothesis

Estimation

Exercises

15 Experimental and Quasi-Experimental Approaches

Introduction

True experiments

Quasi-experimental designs

Repeated measures approaches

Interrupted time series analysis

Independent groups designs

Control group time series designs

Matched pairs designs

Factorial studies

Conclusion

16 The Single Case Experiment

Single case experiments have good internal validity but poor external validity.What is a single case experimental design?

Why SCEDs

The value of SCEDs

SCEDs in clinical practice

Types of SCED

Measurement and quantification

The AB design

The ABAB design

The ABAC design

Multiple baseline designs

Data analysis of single case experiments

Shortcomings of SCEDs

Conclusion

17 Randomised Controlled Trials

Introduction

What is an RCT?

Explanatory versus pragmatic trials

18 Non-Experimental Approaches

Introduction

Descriptive designs

Causal comparative designs (comparative designs/ex post facto designs)

Case–control studies

Cohort studies

Correlational designs

Correlation and prediction

19 Surveys

Introduction

Background issues in survey research

Instrument design and question construction

Administration

Conclusion

20 The Role of Statistics

Why do I need to know about statistical tests?

What are statistics for

Basic ideas: visual displays, levels of data and statistics

Descriptive statistics

Normal distribution

Parametric and non-parametric statistical tests

Inferential statistics

Designs and statistics

Estimation and confidence intervals

Is this really all I need to know?

SECTION 4 APPRAISAL, DISSEMINATION AND IMPLEMENTATION

21 Evidence-Based Practice and Clinical Effectiveness

Evidence and clinical practice

What are EBP and clinical effectiveness?

The process of EBP in practice

Question formulation

Finding the evidence

Appraising the evidence

Applying the evidence

Evaluating one’s own performance

The nature of evidence to support practice

Criticisms of EBP

22 Critical Evaluation of Research Reports

Introduction

Reading the paper

Conclusion

23 Writing a Research Report

Introduction

Help the reader

General structure of the report

Basic style tips

24 Getting Research into Practice

Introduction

The nature of the change process

Effectiveness of interventions to get research into practice

Barriers to implementation

Creating clinical guidelines: a change strategy in your organisation?

And finally...

Appendix 1 Research Log

Appendix 2 List of Useful Websites (In Order of Citation)

Index

To Phil’s wife and children, Sally, Aaron and Becky; and to Rob’s wife, Caroline and late mother, Olive. Also to Rob’s late Latin teacher, Agnes.

This edition first published 2011

© 2011 and 2006 by Robert Newell and Philip Burnard

Blackwell Publishing was acquired by John Wiley & Sons in February 2007. Blackwell’s publishing program has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell.

Registered office:

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

Editorial offices:

9600 Garsington Road, Oxford, OX4 2DQ, UK2121 State Avenue, Ames, Iowa 50014-8300, USA

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/wiley-blackwell.

The right of the author to be identified as the author of this work has been asserted in accordance with the UK 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.

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

Newell, Robert, 1954-

Research for evidence-based practice in healthcare/Robert Newell, Philip Burnard. – 2nd ed.p.; cm. – (Vital notes)Rev. ed. of: Vital notes for nurses : research for evidence-based practice/Robert Newell, Philip Burnard. 2006. Includes bibliographical references and index. ISBN 978-1-4443-3112-7 (pbk. : alk. paper)

1. Nursing-Research-Methodology. 2. Evidence-based nursing. I. Burnard, Philip. II. Newell, Robert, 1954- Vital notes for nurses. III. Title. IV. Series: Vital notes.

[DNLM: 1. Nursing Research-methods.2. Evidence-Based Nursing- methods. WY 20.5 N546r 2011]RT81.5.N46 2011610.73072-dc22

2010011213

List of Common Research Terms

Case studyA descriptive approach to research involving multiple sources of information about a person, organisation or other entityContent analysisThe organisation and categorisation of text materialDescriptive statisticsQuantitative data analysis approach used to simplify presentation of data from a given sampleEthnographyQualitative research method involving study of a cultureEvidence-based practiceAn approach to clinical problem solving involving a systematised synthesis of available informationExperimentQuantitative research method involving randomisation, manipulation of an independent variable, use of a control groupFindingsThe outcomes of the analysis of qualitative research (occasionally used broadly to mean any kind of results in a research study)Four principles approach to ethicsApproach emphasising; respect for autonomy; beneficence; non-maleficence; justiceHypothesisExplicit statement of the predicted relationship between independent and dependent variablesInferential statisticsQuantitative data analysis approach used to draw inferences from sample data to its populationPhenomenologyQualitative research method involving examination of people’s perceptions of their experiencesQualitative researchBroad approach to research emphasising analysis of text materialQuantitative researchBroad approach to research emphasising analysis of numerical materialQuasi-experimentQuantitative research method similar to an experiment but with control and/or random assignment missingRandomised controlled trialA type of experiment used to investigate effectiveness of therapeutic interventionsReliabilityThe extent to which a quantitative study examines the entity it says it does in a consistent and repeatable wayRespondentsThose who are studied or take part in a qualitative studyResultsThe outcomes of the analysis of quantitative researchSamplingProcess by which subjects are chosen to participate in a research projectSingle case experimentApplication of experimental and quasi-experimental methods to a single individualSubjectsPeople whose responses are researched by researchers, in quantitative studies. Participants in research other than the researchers themselvesSystematic reviewSystematised, rigorous approach to literature search and reviewValidityThe extent to which a study examines the entity it says it does

SECTION 1Contextual Materials

1

Introduction to Healthcare Research for Evidence-Based Practice

Introduction

This book is about research in healthcare, seen principally from the viewpoint of students undertaking pre-registration and postregistration educational programmes. We are both active healthcare researchers and are passionate believers in research by healthcare professionals (HCPs)1 for the benefit of the patients and clients we serve. Our own background is principally in mental health nursing, although, as teachers and researchers, we have widened the scope of our work beyond a purely mental health focus. Likewise, as researchers we have typically worked in multidisciplinary research, with the broad range of other HCPs, with medical practitioner researchers and with researchers from a range of non-clinical disciplines. In preparing this book, we have sought to make it relevant to colleagues from across the disciplines, and we hope this comes over in the material that follows. We both started to get interested in research from early on in our clinical careers, and did a good deal of research while still mainly working as clinicians. The real reason we started in research was because we were interested in whether the things we did with patients and clients made a difference to their experiences of illness, recovery and health. If you have picked up this book, we guess you have a similar interest, and we hope we can work with you in developing that interest and finding ways you can translate your ideas into practical projects, whether these be through your own research or your examination of the research literature to inform your practice.

In the following pages, we will try and take you through the various elements of undertaking a piece of research. Even if you do not have to do research yourself as part of your educational course, this book is still for you for three reasons. First, much of the material we present is essential, not just for doing your own research, but for understanding the research of others. Because almost all healthcare courses these days require you to critically appraise the research that is already out there, you need an understanding of how to do that. Books and articles which just deal with critical appraisal are fine, as far as they go, but you will certainly have a much better understanding of how to evaluate published research if you have a clear idea of the various elements that go into a research project, the methods used by researchers and the reasoning behind methodological choices they have made. This book will give you that information.

Second, you will almost certainly encounter, during your course or later, the need to undertake some project work, for example developing a new guideline or a new way of organising care. All the information given here will help you to organise and evaluate that project.

Finally, a great deal of clinical practice is investigated by medical practitioners and non-clinical researchers, partly because comparatively few HCPs other than medical practitioners go on to become fulltime researchers. We want to increase the number that do, so that HCPs are increasingly responsible for evaluating their own practice and get the credit for doing so. Ultimately, we would like you to be in a position to decide you want to be one of those people, so part of the job of this book is to give you a taste of what is involved, including some of its complexities, so that you will want to go on and find out more. We believe that research is essentially a practical skill and is best learnt through an apprenticeship system, and so the best piece of advice we can give you is to get hold of someone who has experience of actually doing research, translating research into practice, doing a systematic review and so on, and learn from them.

Then, use this book as your workshop guide. If you cannot find such an experienced researcher, then we hope this book will be able to tell you some of the things they would have. The book is not heavily referenced (usually only a few per chapter), but each of the references is important, and is easily available, either from your library or from the internet. We have made considerable use of web sources so as to make it easy for you to find the best supporting information. We know web sources do not necessarily remain current forever, but we believe the best ones do. When we came to write the second edition of this book, we found only a few sites that were no longer available.

The scope of healthcare research

Until quite recently, healthcare research had a reputation for being an introspective pursuit which was more concerned with investigating its own workforce than undertaking clinical research. Some commentators have suggested that becoming a teacher of healthcare professionals frequently involved ceasing to have any clinical responsibility for patient care or, indeed, much contact with clinical settings at all. In consequence, those teachers wishing to do research had little access to patients or were often out of touch with issues which were important to patients. They did, however, have contact with students, and so ended up developing research interests related to education and the views and experiences of student nurses. Sometimes, it was difficult to see how this research would benefit patients. Although this criticism was chiefly aimed at nurse researchers, our experience with the broad range of healthcare professionals suggests it holds true equally in other disciplines, a belief reinforced by the fact that initiatives to increase research by non-medical health disciplines have typically been applied across the range of disciplines.

We do believe that healthcare research is changing for the better, though, and is nowadays much more concerned with patient care, rather than being overly inward looking towards its own professions. Whilst we recognise that it is important for research to be done into such things as the opinions and experiences of students and members of the healthcare professions, or the ways in which these professionals are educated, we also think that the eventual point of all healthcare research should be the greater good of patients. Therefore, we suggest, the vast bulk of research into such issues as the views of the professions themselves should have immediate consequences for patient care. If it does not, then why do we want healthcare researchers doing it, rather than, say, sociologists or educationalists? Surely, examination of, for example, occupational therapists’ opinions of their educational preparation, can be done as well by researchers from other disciplines. Given that healthcare researchers are a rare breed, we hope that the growing focus on clinical research, where nurses can make a distinctive contribution, will continue.

That said, many people are largely unaware of the contributions to research that healthcare professionals have made already, or the effect which that research has had on care. Just two fairly random examples from our own areas of interest give an idea of the scope of research by healthcare professionals. Professor Mary Jo Dropkin, from the University of Long Island, has written definitive studies of the psychosocial impact of head and neck cancer. Her work is cited by researchers across the whole range of health disciplines and has changed the way we think about head and neck cancer. In the UK, Professor Trudie Chalder from Kings College, London, is a world-recognised expert in fatigue, and developed the leading mode of treatment in this area. Once again, her work is referred to by all the healthcare professions. When you consider that fatigue is cited as a major symptom in almost all long-term physical conditions, it is easy to see the extent of this HCP’s contribution to the potential well-being of patients via her research.

Whose business is research in healthcare?

As you can tell from the above, we think it is primarily the business of HCPs themselves to evaluate and develop our care through research. These days, almost all large-scale research is undertaken in teams, and almost all these teams are multidisciplinary. All HCPs need to be equipped to take a full part in these teams. In the past, we have been ill equipped to do so, and, given the packed nature of healthcare pre-registration education, research often takes a back seat. As we said above, very few clinicians go on to be full-time researchers, but all of us are research users, even when we are not aware of it. Being a knowledgeable, aware consumer of research findings is integral to competent practice.

Apart from using research in our own clinical practice, we have a further ethical obligation concerning use of research, and one which exceeds the responsibility of members of the general public. For example, HCPs need to be sufficiently knowledgeable about research to help patients who may become involved in research projects run by other members of the clinical team to make reasonable, informed choices about, for example, participation in such a study. Similarly, patients may ask our views about treatment which is currently practiced, and we are unable to offer such advice without an informed understanding of the evidence base. This, in turn, comes from an understanding of the research approaches which have been used to generate this evidence.

Surprisingly, even in everyday life, away from clinical practice, our role as an HCP gives us a greater ethical obligation to understand the evidence behind healthcare interventions, because our role may give us a certain amount of authority when we communicate (even very informally) with others about healthcare matters and healthcare research. Accordingly, we have a special responsibility to ensure that we know what we are talking about. This implies, once again, a knowledge of research methods.

Which brings us to our final point in this section. Very little healthcare has been subject to robust clinical research. This leaves us, we believe, with two important responsibilities. First, we should be basing our care as far as possible on the best available evidence. This implies an ability to search for, appraise and implement that evidence. Appraisal requires a basic understanding of the merits of the studies we read, and a knowledge of research methods is essential to that understanding. Second, the knowledge base needs building, so involvement, at whatever level, in research to build it is as part of our ethical responsibility as HCPs in just the same way as use of current best evidence.

Perhaps the key questions at the heart of evidence-based practice in the health professions are as follows:

What works?

What works best?

How does it work?

Research provides a starting point in answering these questions.

Using this book to get involved in healthcare research

We want this book to be a practical guide. Part of being practical is being as easy to access as possible. This leaves us with a problem. We have tried to make each chapter as stand-alone as possible, but at the same time we wanted to avoid repetition, so we do not go over every piece of background information necessary in each chapter. This means there will inevitably be some shifting around for you between chapters, and we hope you will dip in and out to follow up things we have not been able to cover over and over again as they occur in different contexts. To help you do this, we are going to avoid giving you the traditional detailed chapter-by-chapter description of what is going to be covered in this book. Instead, here we are simply giving you each chapter title, followed by the key points section for that chapter. As a result, at any point, you will be able to flick back to this chapter (dog ear it now) and read through key points to give you an idea of where relevant issues are covered. Please do not feel you have to read the whole book. It is a tool. Use what you need and leave the rest. Maybe it will be of use later.

The chapters

Chapter 2: The research process – organising your research

The research process is a way of organising a research project.

The study aims and objectives guide the study.

The literature review provides the context for the study and determines the need for it.

All stages of the process should be clearly described with appropriate rationale.

Issues of ethics are fundamental to the research process.

Dissemination and implementation complete the research process and start the research cycle with new questions.

Chapter 3: Choosing methodological approaches

Researchers tend to associate inductive reasoning with qualitative research and theory building, and deductive reasoning with quantitative research and theory testing.

Quantitative approaches emphasise cause–effect relationships and prediction.

Qualitative approaches emphasise exploration.

Researchers should examine the goals of their research when choosing methodological approaches.

Consider qualitative approaches first for studies of individuals’ experiences; research with excluded and hard to reach groups; pilot studies.

Consider quantitative approaches first for epidemiological studies of large groups; treatment comparison studies.

Chapter 4: Searching the literature

Literature searches are done primarily to ensure awareness of a field of research.

Systematic reviews examine the literature using systematised, transparent criteria.

A search strategy consists of the research question, its components, sources of information, search terms, retrieval and inclusion criteria, available resources.

Sensitivity (recall) refers to comprehensivity of a search strategy.

Specificity (precision) refers to relevance of a search strategy.

The scope of a search is determined by its search strategy.

Chapter 5: Ethics of healthcare research

Codifications of research ethics date from the Nuremberg code and the declaration of Helsinki.

Autonomy, beneficence, non-maleficence and justice are key principles in research ethics.

Autonomy refers to an individual’s freedom to choose and act.

Beneficence and non-maleficence require that we maximise good and minimise harm.

Justice is the maximising of fairness to all.

Research ethics committees exist to interpret these concepts for the protection of research participants and researchers.

All research should receive ethical scrutiny.

All NHS-related research must receive approval from a Research Ethics Committee (REC).

NHS REC approval is centrally organised and standardised.

All NHS research must receive Research Governance approval from the NHS institution in which it takes place.

Research Governance approval is locally organised by each institution and there is limited standardisation.

Chapter 6: Basic concepts – sampling, reliability and validity

Sampling is an everyday activity, not peculiar to research.

A population is a total group from which a sample is drawn.

Some populations are themselves samples from larger populations

Representativeness is key to sampling, but is defined differently by quantitative and qualitative researchers.

Samples may be random or non-random.

Sampling technique is guided by the aim of the research.

Validity is the extent to which a study examines the entity it says it does.

Reliability is that it does so in a systematised, repeatable way.

Quantitative and qualitative researchers place different emphasis on different aspects of validity and reliability.

Chapter 7: Issues in qualitative data collection

Data collection choices are made in response to research aims.

Sampling in qualitative research aims at illumination rather than representativeness.

Interviews may be structured, semi-structured or unstructured.

Interviews are normally transcribed verbatim.

Sometimes, qualitative data can be gleaned from questionnaires.

Observational studies benefit from painstaking field notes.

Published work can be subjected to similar data analysis to other research methods.

Chapter 8: Case studies

Case studies are descriptive pieces of qualitative research.

Case studies may be stand-alone investigations or illustrations from larger studies.

Case studies examine a particular person, group, situation or set of circumstances in detail.

Case studies are not necessarily typical of general experiences.

Case studies rely on a high level of detailed description.

Sample selection for case studies can lead to challenges in terms of typicality or inevitable comparisons with other settings.

Chapter 9: Ethnography

Ethnography involves the in-depth study of a culture.

Ethnographic approaches use elements of ethnography.

Ethnographic approaches can be combined with other methods.

Ethnographic approaches usually involve extended amounts of field work.

Ethnography combines observation with other methods such as interviews.

Formal recording of interviews may often be impossible, so field notes are particularly important.

Disconfirming evidence is actively sought.

The range of phenomena to be observed is potentially overwhelming.

Ethnography reminds us of the importance of cultural context.

Chapter 10: Phenomenology

Phenomenology is concerned with individuals’ perceptions of their experiences.

Phenomenology as a philosophy is concerned with seeing things without making value judgements.

Phenomenology frequently uses in-depth interviews and series of interviews.

Bracketing is the attempt to put aside one’s own thoughts, feelings and beliefs.

The researcher avoids explanations of people’s accounts so that the person’s own voice can emerge.

Phenomenology aims to create vivid personal insights.

Chapter 11: A pragmatic approach to qualitative data analysis

Content analysis refers to the organising and ordering of textual material.

Transcription involves writing out recordings of an interview.

Some degree of quantification is possible.

Categories can be pre-defined or can emerge from the data.

The pragmatic approach involves six stages:

taking memos after each interviewreading transcripts and making notes of general themesrepeated reading and generating open coding headings to describe all aspects of the datareducing the codes under higher order headingsreturning to the data with the higher order codescollating the organised data for reporting

Chapter 12: Limitations of qualitative research

Qualitative research does not claim to be scientific in the same way as quantitative research.

Samples in qualitative research are rarely representative.

The researcher’s own influence on the emerging data may be checked by bracketing and by discussion with respondents.

There is too much variability to allow replication of qualitative studies.

Qualitative research does not aim to generalise.

Neither researcher nor respondent is necessarily aware of their own biases.

Analysis of interviews can be affected by hindsight bias.

Validation by respondents is itself potentially problematic.

Respondents can offer explanations about things they have no way of knowing about.

The illuminative value of quantitative research is slowly gaining ground in healthcare.

Chapter 13: Sampling, reliability and validity issues in data collection and analysis

Random sampling decreases the likelihood that members of a sample are different from its population.

Stratification and cluster sampling both ensure adequate representation of population subgroups in a sample.

Quota sampling and systematic sampling approximate to random sampling.

Convenience sampling is the simplest form of sampling but is least likely to conform to its population.

External validity refers to the applicability of a study to the real world.

Population validity refers to the similarity between a study sample and its population.

Ecological validity refers to the similarity between study conditions and procedures and the real world.

Internal validity determines the confidence we can have in cause–effect relationships in a study.

Reliability consists of two concepts: consistency and repeatability.

Consistency implies that, if a phenomenon is unchanged, it will be measured as the same by several observers or by several measuring methods.

Repeatability means that, if a phenomenon is unchanged, it will be measured as the same on several occasions.

Chapter 14: Cause and effect, hypothesis testing and estimation

Assertions about cause and effect are probabilistic, not definitive.

We make cause–effect predictions in daily life.

Cause–effect predictions in healthcare research are the same as those in daily life.

Quantitative researchers are concerned with independent variables, dependent variables and intervening variables.

Researchers manipulate independent variables, observe any changes to dependent variables and attempt to account for intervening variables.

Hypotheses are explicit statements of the predicted relationship between independent and dependent variables.

Directional hypotheses are made only when there is reason to think a relationship operates only in one direction.

Adequate statistical power is essential to the safe acceptance of the null hypothesis.

Hypothesis testing is an all-or-nothing statement of relatedness, but estimation emphasises the extent of a relationship.

Chapter 15: Experimental and quasi-experimental approaches

Experiments consist of three elements: manipulation of the independent variable, use of a control group, random allocation to experimental conditions.

These three elements allow experiments to claim a high degree of internal validity.

Quasi-experimental designs use a similar general approach to true experiments but control and/or random assignment may be missing.

Repeated measures approaches use the same participants for all experimental conditions.

Repeated measures approaches are vulnerable to order effects.

Order effects may be reduced by counterbalancing.

Independent groups designs use different participants for different experimental conditions.

Independent groups designs without randomisation are vulnerable to differences in participant characteristics (subject variability).

Matched pairs designs match participants on important variables to combat subject variability.

Factorial designs study interactions between several different independent variables.

Chapter 16: The single case experiment

Single case experiments apply experimental methods to treatment of single individuals.

Single case experiments allow you to do research in the course of clinical practice.

In AB designs and variants, different interventions are introduced and withdrawn sequentially.

In multiple baseline designs, responses to different interventions are compared cross-sectionally.

Systematised measurement is essential to single case experiments.

Data analysis is often confined to visual inspection of changes in scores.

Visual inspection can involve examining raw scores, means, levels, trends and latencies.

Statistical approaches have been developed because the interpretation of visual data can be subject to bias.

Single case experiments have good internal validity but poor external validity.

Chapter 17: Randomised controlled trials

A randomised controlled trial (RCT) is a type of experiment.

RCTs involve randomisation, a control group and manipulation of the independent variable.

RCTs attempt to control bias.

RCTs are the most effective way of examining cause and effect relationships, including the effect of treatment on patients.

Explanatory trials possess high internal validity and establish what principle, mechanism or theory accounts for a change in patient condition.

Pragmatic trials possess high external validity and establish how well a principle, a mechanism or a theory translates into the real clinical world.

RCTs are essential in providing patients with accurate information about healthcare interventions.

Chapter 18: Non-experimental approaches

Non-experimental approaches involve observation of naturally occurring relationships and differences between variables.

Non-experimental approaches are useful when experimental approaches are unethical or impractical.

Causal inference is weak in non-experimental approaches.

Descriptive designs only describe relationships between variables.

Causal–comparative designs (comparative/ex post facto designs) compare two or more naturally occurring groups.

Correlational designs explore relationships between variables in a single group.

Some correlational approaches allow prediction through the assignment of predictor variables.

Chapter 19: Surveys

Surveys are often wrongly assumed to be simple to carry out properly.

Sampling technique and sample size are important in determining the margin of error in a survey.

Survey designs are typically observational, but may also use experimental and quasi-experimental approaches.

Decisions about question construction and questionnaire administration are fundamental to a successful survey.

A recent systematic review has revealed that some ‘research folklore’ guidelines about survey methods and question construction are not supported by empirical research.

Chapter 20: The role of statistics

Statistics are important because they convey large amounts of data in an understandable way.

Statistics describe data and allow us to draw inferences based on probability from these data.

Data are divided into continuous (ratio, interval, ordinal) and categorical (nominal/categorical) data.

Statistical tests are divided into parametric and non-parametric tests.

Parametric tests usually assume data are at interval level, normally distributed and possess homogeneity of variance.

Different statistical tests are designed to deal with different types of data.

Different tests are designed to deal with different sources of variability from different research designs.

Chapter 21: Evidence-based practice and clinical effectiveness

Research aims to add to the evidence to inform clinical practice.

Evidence-based practice (EBP) and clinical effectiveness are related terms, but clinical effectiveness has the narrower focus.

EBP is a process involving asking answerable questions, finding evidence, appraising the evidence, applying evidence to practice, evaluating performance of EBP.

The hierarchy of evidence runs from most to least trustworthy on the basis of freedom from bias.

EBP is intended to enhance clinical decision-making, not replace it.

Chapter 22: Critical evaluation of research reports

The ability to weigh evidence is a required skill for competent practitioners.

A good literature review should critically examine the literature and set

the current study in that context. The method section should be detailed and demonstrate appropriate choices.

Different specific methodological issues are associated with qualitative and quantitative research.

The results section should be clear.

Quantitative results should distinguish between significant and nonsignificant results.

Qualitative results should avoid claims as to generalisability.

The discussion section should set the results in the context of the literature, implications for practice and further research.

Study weaknesses should be honestly and comprehensively reported.

Chapter 23: Writing a research report

The report is a product – help the reader.

All reports follow a similar general structure.

The executive summary is all most readers will read.

Do not spend too long on general methodological debate.

When reporting results:

be clearbe comprehensivedistinguish clearly between results and comment

A good discussion amplifies the study results, shows the importance of the study results, relates the results to earlier research and theory, admits shortcomings of the study, shows implications for future research and for practice.

Follow assignment or journal guidelines exactly.

Write with short words, sentences and paragraphs.

Use linking sentences to join paragraphs.

Avoid punctuation you do not understand.

Write in written English, not spoken English.

Read work aloud to help with punctuation.

Chapter 24: Getting research into practice

The theory of diffusion of innovation divides change adopters into innovators, early adopters, early and late majority, and laggards.

Change diffusion is mediated by personal characteristics and accessibility of the innovation.

Journal articles are ineffective in changing practice.

A clear dissemination and implementation strategy should be devised on the basis of a diagnostic analysis.

The analysis should inform broad-based interventions to introduce the change.

The BARRIERS scale identifies four elements of barriers to change the adopter, the organisation, the innovation and the communication.

Notes on person and gender

We wrote this book together, with the idea of a series of conversations with you, the reader, in mind. For that reason, it is written largely in the first person plural, and we often address you directly. Occasionally, we vary this slightly and use the first person singular. This is usually to recount something which has happened to one of us and which has influenced our development as individuals and as researchers. Really, this is just to avoid cumbersome expressions like ‘one of the authors (RN)’ and so on. We hope this direct style is one you can engage with and will help you get into the spirit of a conversation with us about research.

We have made no definite choice about the use of the personal pronoun, and we refer to he or she pretty much indiscriminately throughout the book.

Good luck with this book and your research.

Endnote

1. Throughout this book, we will use the expression ‘Healthcare Professionals’ to refer to the broad range of clinical professions in healthcare (e.g. nurses, midwives, physiotherapists, occupational therapists, radiographers, podiatrists, clinical prosthetists and many, many others). We are excluding medical practitioners from this category for no other reason than that their educational and clinical tradition has developed very differently, and led to a different (and, perhaps greater) preparation for and engagement with the research that provides evidence for practice.

2

The Research Process – Organising Your Research

Key points

The research process is a way of organising a research project.The study aims and objectives guide the study.The literature review provides the context for the study and determines the need for it.All stages of the process should be clearly described with appropriate rationale.Issues of ethics are fundamental to the research process.Dissemination and implementation complete the research process and start the research cycle with new questions.

Introduction

This book is about all aspects of the research process as it relates to healthcare research. This chapter offers an overview of what is involved in that process and can be read in conjunction with other, more specific, chapters. Those chapters will, for example, discuss the differences between qualitative and quantitative approaches to research. Whatever approach is chosen, the same broad process is undertaken and it is this broad process that is under discussion here. Many writers have described the research process, often providing subtly different components. We do not subscribe to any one description of the process rather than another. Instead, we offer below a series of suggestions as to how a research project might be organised, based on the following view of the research cycle: research questions, literature review, selection of sample, selection of measures, data collection, data analysis, report writing and dissemination. Indeed, much of this book is constructed with this cycle in mind.

In a way, research is always a cyclical process. Once findings from a study have been established, they nearly always bring us back to new questions that need answering. It is not a good idea to view research as establishing ‘the truth’. It is merely offering the best view we have on a given topic at a particular time in history.

The title

A research project needs a descriptive title. When other researchers are looking to see what research has been done before, it is the title of the project that will alert them, either to want to read the report of that project or to pass over it. The title, then, should be concise and, as clearly as possible, describe the nature and content of the research project. Thus, a title such as:

‘A qualitative study of student midwives’ views of reflective practice’

is preferable to:

‘The mirror of the mind: students and reflective practice’

The first title states clearly both the nature and the content of the study, whilst the second one is less clear. It is probably best to avoid what might be called ‘romantic’ titles, or titles that use an analogy (such as ‘The mirror of the mind...’), as these may be confusing or may miss the point. The title, wherever possible, should indicate both the research approach used and the topic of the study.

The title should, until the research is finished, be considered a ‘work in progress’. It may be found that the title changes as the study is completed. Considerable time should be given to ensuring that the title is as accurate and as concise as possible. It is usually intimately related to the research questions.

The abstract

One of the final tasks in designing a research project is to write an abstract for the study. It is discussed here because it is usually one of the first pages in a research report or proposal for funding. An abstract is normally a one-page description of all aspects of the study. Thus, the researcher will report, briefly, the aims of the study, the sample, the data collection and analysis methods used and the findings. In writing proposals for funding, it is usually a requirement that the abstract is written in lay terms. This in itself can be helpful in guiding the researcher towards highly focused research aims and objectives.

Although the abstract is one of the first pages of a research proposal, it is often best written when the proposal is complete. In this way, the abstract more easily becomes a concentrated summary of what the project is about.

Aims and objectives

While the heading of this section is ‘aims and objectives’, in practice, various terms can be used here. A research project has to be clearly focused, because it is this focus which provides the direction for the research process as the project takes shape. The researcher must know, in advance, what he or she is trying to find out. Most beginning researchers attempt to do too much and the process of narrowing down the focus of interest can be a lengthy and painful one. A useful means of knowing that you have finally identified the exact focus for a study is when you can state it in one sentence. For example, the beginning researcher might say, when asked what his or her project is about: ‘I want to look at the things students think about reflection and reflective practice. I want to know what their problems with it are and how it affects their clinical practice. I also want to hear about their teachers’ experiences of teaching reflective practice and see if they feel that it makes a difference to clinical practice’. After discussion with a research supervisor and a reading of the literature, the answer to the question: ‘What is your project about?’ might be: ‘It is a descriptive study of student midwives’ attitudes towards reflective practice’.

Having identified exactly what the project is about, the researcher will write either a set of aims and objectives for the project or a number of research questions. Aims are broad, general statements about what the research will be about, while objectives are smaller statements about the actions to be taken. Thus, an aim might be to ‘Identify the attitudes of student midwives towards reflective practice’, while two objectives might be to ‘(1) identify an appropriate, convenience sample of student midwives; (2) use a validated instrument to identify student midwives’ attitudes towards reflective practice’.

A research question is not dissimilar to a research aim but the focus of the research is demonstrated in a question. For example, ‘What are some student midwives’ attitudes towards reflective practice?’

In the case of quantitative studies, a hypothesis will be stated – a very specific statement that is to be tested by the research and found to be either ‘supported’ or ‘not supported’ at the end of the study. Occasionally, the statistical term, null hypothesis, is used to state that something is not likely to be the case at the end of the study (e.g., from the pharmacy world, drug ‘A’ will not be demonstrated as being more effective in the treatment of hypotension than drug ‘B’). Hypothesis testing is discussed in more detail in Chapter 14.

A review of the literature

As we shall see, all researchers need to do a thorough search of the literature that has preceded the study. This search of the literature is to find out what previous research has been done, what others have written about on the topic, and to place the present study in context. Research projects do not simply arise out of the blue. They are always linked, closely, to a particular time in history, a particular set of beliefs that are current about a subject and must always be clearly linked to what is already known about the topic in question.

The development of computer software that allows huge databases of research reports and publications to be searched means that most researchers can now undertake systematic reviews of the literature. The aim of a systematic review is to attempt to identify almost every paper that relates to the current topic being researched. This is no easy task, even with computerised search engines. The researcher must identify key words in order for the engines to identify appropriate papers. Once a range of papers has been identified, the researcher then needs to filter these down to exactly the papers that are relevant to his or her study. Finally, the papers themselves have to be obtained, read, summarised and their reference formally recorded. Searching the literature is covered in detail in Chapter 4.

In reviewing the literature, in this way, the researcher is also required to read critically. What are the limitations of the study report that is being read? What, if any, mistakes were made in the research process? What else needs to be done? Asking these questions can help further focus on the study being undertaken by the researcher who is doing the review. The critical evaluation of research papers is described near the end of this book in Chapter 22.

It is not always necessary for the researcher to design an entirely new project. There is considerable value in undertaking a replication study. In such a study, the methods used in a previous study are used again. If the replication is very close to the original piece of research, much can be learned about whether or not the original work was sound and whether or not the findings from the original are still valid or current.

There is a certain hierarchy in the value of papers that are reviewed. The soundest evidence comes from original report papers. After that, in degrees of value, come theoretical papers, which summarise the research and develop theoretical perspectives. A long way behind these two come ‘opinion papers’. These are usually short, one-page papers that appear in many healthcare journals and magazines. They may help to illustrate some of the current debates about certain issues but, in the end, usually only reflect the writer’s own views. As a result, they are not particularly useful in generating information for a review.

The most thorough researcher also considers the ‘grey literature’. This is that literature which is not formally published in journals or books. Examples of the grey literature include documents such as college reports, curriculum documents and handouts. Again, their status may not be very high in the evidence hierarchy but they can help to identify current trends in a particular discipline.

Some reviewers of the literature attempt a type of content analysis of that literature. This is to say that they count the occurrence of certain issues that arise across a spread of papers and then tabulate their findings. Such counting needs to be done carefully and is probably only really valid when a researcher has managed to identify all the literature in a given topic.

A systematic review of the literature means being systematic in other ways too. The researcher must set up appropriate and careful systems for managing the literature that they find. He or she must also be careful to record the exact reference to a given piece of work. Such references (e.g. author, date, title of book or paper, publisher of book or paper, page numbers [in the case of a journal paper]) can be stored either in a dedicated, computerised, reference database (such as End Note or Reference Manager) or can be stored on cards or in a notebook. What is essential is that the researcher is very clear about how to properly cite references, using either the Harvard or Vancouver systems of citation, and is also very careful to record all the details of a reference for further use in writing the research proposal or report. This point cannot be repeated too much: you MUST know how to reference properly. In our experience, many students at various levels in their academic careers have difficulty with the particular aspect of scholarship. It is worth learning to reference very early on in that career.

The sample

A sample is a slice of any given population. For example, we might consider ‘the population’ to be those who live in London. A ‘sample’ would be a selected group of those people.

In quantitative research, there are careful means for selecting a representative sample: that is to say, a sample that can, statistically, represent the larger population. The findings from such a sample can sometimes be generalised out to that larger population. There are computer programs for generating a power calculation which allows the researcher, in quantitative research, to identify the minimum number required in a sample for representativeness, given the size of the total population. Careful selection of a sample, in quantitative research, is essential for identifying the degree to which findings can or cannot be generalised out to the larger population. These issues are discussed further in Chapters 6 and 13

Generally, in qualitative research, three types of sample are routinely used. Researchers may use a convenience sample, a purposive sample or a snowball sample. A convenience sample is one that is made up of people who happen to be available to take part in the research. For example, in a midwifery department, a particular class or group of classes may constitute a convenience sample. A purposive sample is one made up of people who are reasonably likely to be able to offer information or views on a given topic. For example, it would be pointless asking student midwives who had no experience or knowledge of reflective practice for their views of reflective practice, but much more helpful to ask students who did have such experience or knowledge. We could sample specifically for students who had this experience or knowledge, and such a group would be a purposive sample. Surprisingly, if the whole population of students had such experience, no purposive selection of them according to that criterion would be possible. It would not be appropriate to refer to students from this group who eventually participated in the study as a purposive sample simply because they possessed characteristics we would have wished to select for, because the whole population have this characteristic.

A snowball sample is rather different and is usually used, in qualitative research, when interviews are being conducted. Here, the researcher asks the interviewee, at the end of the interview, to recommend someone else that he or she might interview. Thus, in a study of healthcare education, the interviewer may interview a course leader about the economics of such education. After the interview, the researcher may ask the course leader to recommend another person for interview and the interviewee may suggest the local finance officer. Thus, the pattern of sampling ‘snowballs’. Clearly, the snowball sample is also a purposive one. Issues around sampling in qualitative and quantitative research are covered in detail in Chapters 7 and 13, respectively.

Measures and materials

Measurement is usually a feature of quantitative research and involves the selection of measures and scales appropriate to accurate examination of the issues to be explored during the study. In qualitative research, although formal measurement is less frequent, the researcher will want to create detailed interview schedules or guidelines for the completion of field notes. There may also be stimulus materials to be shown to participants. For example, focus group interviews sometimes ask participants to comment on such things as the appropriateness of patient information booklets. In all cases, a clear rationale for the use of particular measures and materials should be described.

Data collection

The most common forms of data, from social science and healthcare research, arise in one of two forms: numbers or words. Occasionally, photographs and other forms of data are also collected and the collection of these seems likely to increase in the future as more sophisticated and computerised methods of data storage and analysis become available.

In quantitative research, the data are mostly likely to be numerical. These figures can come from various sources: they may be scores on a questionnaire, arise out of direct counting of things or may come out of other data sets. In qualitative research, data (in the form of words) usually come from transcripts of interviews or from the researchers’ field notes.

A short list of data collection methods to be found in social science and healthcare research might include, at least, the following:

Data collection approaches

Direct, non-participative, observation by the researcherParticipative observation (observation made while the researcher is working alongside other people)Questionnaires and scalesInterviewsField notesFocus groups (or group interviews)Historical review of documentsMeta-analysis of findings from other studies

You will see that no research designs are given here. We have not mentioned, for example, ethnography, randomised controlled trials or systematic reviews. This is because designs are basically approaches to organising the way in which data are collected. So, for example, a randomised controlled trial could involve direct observation and the use of scales.

Procedure

Strictly speaking, the procedure is not necessarily data collection, but refers more generally to everything you, as a researcher, do with a research participant. This includes, for example, initial meetings with potential participants to recruit them into the study, gaining consent from them, communicating with them to inform them further about the study, and so on. However, in healthcare research, the idea of a procedure commonly carries the sense of some intervention with participants. Most often these are clinical interventions with patients (e.g., the initiation by physiotherapists of a group class in balance for older patients admitted to hospital following a fall), but quite possibly interventions with staff (comparison of two different initiatives to encourage evidence-based practice) or healthy volunteers (comparison of two methods of presenting information on which to base informed consent to enter a research study).

Data analysis

Although much more is discussed about the analysis of data in other chapters of this book, two broad approaches can be noted here. Numerical data are normally summarised and presented in the form of percentages of responses within the sample. If the correct form of sampling has been used,inferential statistics may also be used with these summarised data to identify the degree to which it is safe or not safe to generalise the findings out to the larger population. Perhaps the most widely used computer program for handling numerical data, in social science and healthcare research is SPSS (Statistical Package for the Social Sciences).

Textual data are normally content analysed in some way. In this case, transcripts of interviews, group interviews or other written material are summarised under a range of headings and subheadings. As we shall see, there are various approaches to tackling the task of identifying these headings and subheadings. Various computer programs exist for managing qualitative, textual data (e.g., Atlas ti and The Ethnograph). It should be noted, however, that whilst a numerical package such as SPSS analyses numerical data, a textual data package such as Atlas ti, does not. Qualitative data management programs do not analyse data they merely assist the researcher in that process. By the same token, statistical packages such as SPSS cannot decide for the researcher which test is to be used on the data. The package merely assists by running that test.

Ethical issues

Various ethical issues arise in almost all social science and healthcare research. For example, is the study in question worth doing? Is it likely to do harm to anyone who takes part in the study? Are the methods used likely to help the researcher to achieve his or her aims or answer his or her question? These questions have to be answered: poor or dangerous research is worse than no research at all. Most hospitals and other healthcare-related institutions have their own research ethics committees, to which full research proposals must be submitted and approved – before the research is started. In the UK, there are Local Research Ethics Committees, comprising of medical and healthcare staff and lay members, who consider all larger research proposals. NHS Trusts and universities are now insisting that all research that involves people (staff or patients) must be subject to formal ethical approval before a study is begun.

Financial issues

All research that is done in a place of work, involves the spending of money. It is good practice (a) to formally identify the cost of a particular project and (b) to seek approval for the funding of that project. Financial considerations that need to be made include the cost of staff time, the cost of the use of equipment (and, sometimes, the cost of the equipment itself), stationary and photocopying costs, postage and secretarial costs, the cost of any software required for data analysis, travel and related costs. Many colleges and universities have departments which will help provide an accurate costing for a proposed project. If this is available, it is probably best to have such a department work out costs: they will have a better understanding of, in particular, staff costs than the average researcher.

If a study is being submitted for a grant or other sort of funding, the grant-awarding or funding body will insist on a fully worked out costing of the project. Again, this is where a university department that exists to cost projects can be very useful.

Writing the report

Generally, there are two approaches to the writing of a research report. One is to begin the write up at the start of the project and continue to write it throughout the life of the project. The second is to write up the project after the findings have been identified. There is much to commend the former approach. Most sections of a research report reflect the order in which research activities are carried out, so it makes sense to write down what happens in each of the research stages as they happen. The one exception to this is the review of the literature. This is likely to continue throughout the life of the research. One of the last tasks, in any research project, should be to make a final check of the available literature. Most researchers, despite this, have been saddened to find that, immediately after finishing their study, two or three important papers suddenly appear in print!

Most research reports, in their ordering, reflect the stages of the research process as reflected in this chapter. Thus, after an initial introduction, the report will identify the aims and objectives of the study (or the hypothesis or research question). It will then identify the sample and explain how it was derived. The following sections will describe the data collection methods and data analysis methods used. Another section will report the findings of the study. The final section will usually be a discussion of the findings, with the implications of the study being drawn out. There is some debate as to whether such a section should or should not identify the strengths and weaknesses of the study.

Those in favour of reporting strengths and weaknesses might argue that this is a means of evaluating what has been done. No research project is perfect and we can all identify things that went wrong or issues that had not been anticipated. Those who propose not reporting strengths and weaknesses might argue that a given study should stand on its own merits. Presumably, the researcher did the best job that he or she could do and, also, should have been able to anticipate or put right any errors that became apparent. A cynical view might also be that no researcher is going to evaluate the negative points of his or her study to the point where such critical commentary undermines the project.

Publishing the findings

Research is not finished until the study is published. This is the final stage of the research process, or, if we regard that process as cyclical, the starting point for the generation of further research questions, both by the researchers and by those who read their reports of their work. Most researchers aim at publishing their findings in a doubleblind refereed journal (e.g., in Occupational Therapy International, Journal of Clinical Nursing, British Medical Journal, British Journal of Clinical Psychology). That is to say that their work is considered, anonymously, by at least two referees, who normally are the researcher’s peers in the field. These reviewers’ judgements are involved in deciding whether a research report is published or not. Some researchers publish shortened versions of their work in more frequently published, professional and popular journals (e.g., in nursing, this might be in the Nursing Times, whilst in radiography, Radiography is a journal which has a mass readership amongst radiographers, but still maintains a considerable academic profile, which is reflected in the depth of many of its papers).