Delphi Technique in Nursing and Health Research

Dr Sinead Keeney is a Senior Lecturer in the Institute of Nursing Research at the University of Ulster. Felicity Hasson is a Senior Lecturer in the Institute of Nursing Research at the University of Ulster. Professor Hugh McKenna is Dean of the Faculty of Life and Health Sciences at the University of Ulster.

• Preface ixAcknowledgements x1 The Delphi Technique 1Introduction 1History of the technique 2What is the Delphi technique? 3Defining the Delphi technique 3The Delphi process 4Original Delphi 4Idea generation 5Priority setting versus consensus 5Non-consensus Delphi 5Types of Deplhi 6How has the Delphi evolved? 6Sampling and the use of experts 7Defining ‘expert’ 7Employing an expert panel 8Size of the expert panel 8Valid opinion 9Anonymity 9Quasi-anonymity 9Group dynamics 10Delphi rounds 10Round 1 11Subsequent rounds 11Response rates 12Enhancing response rate 12Consensus 13Does consensus exist in expert panels? 14Concept of consensus 14Increasing popularity in nursing and health research 14Comparison of the Delphi with other consensus methods 15Nominal group technique 15Consensus conference 16Key learning points 16Recommended further reading 172 Debates, Criticisms and Limitations of the Delphi 18Introduction 18The qualitative-quantitative debate – which paradigm does the Delphi belong to? 18Criticisms of the Delphi technique 20Lack of universal guidelines 20Size of expert panel 21Implications of lack of anonymity 22Expert ‘opinion’ 23Level of consensus 27Limitations of the Delphi 28Pressures of conformity 28Demanding nature of the technique 29Key learning points 30Recommended further reading 313 Applications of the Delphi in Nursing and Health Research 32Introduction 32Historical application of the Delphi technique in nursing 32Identification of clinical nursing research priorities 36Trends of the Delphi in nursing 40Key learning points 41Recommended further reading 424 How to Get Started with the Delphi Technique 43Introduction 43Preparation and practicalities 43Suitability of the Delphi 43Availability of resources 44Level of consensus 45Identifying target sample – panel of experts 46Who is the target population? 46How do you select your experts? 47Sampling criteria 47What size does the sample have to be? 48Response rate and attrition 53Deciding on nature and delivery of the first round 55Gaining names and addresses 56Deciding on a ‘return by’ date 56Administration 57Invitation to participate 57Explicit cover letter outlining the working of the Delphi 59Design of questionnaire 59Administration systems 63Mailing 64Maximising response rate 64Content analysis 65Process 66Key learning points 67Recommended further reading 675 Conducting the Research Using the Delphi Technique 69Introduction 69First round 69Classical Delphi 69Modified Delphi 70Return of first round 71Consideration of sample size and number of items generated 71Bootstrapping 72Content analysis 72Subsequent rounds 73Round 2 74Cover letter explaining Round 2 74Instructions for Round 2 75Designing Round 2 questionnaires 76Likert scales 77Round 2 analysis 77Sample motivation 78Follow-ups/reminders 78Round 3 78Individual and group feedback 78Round 3 analysis 81Number of rounds 81When to stop 82When not to stop 82Reaching consensus 82Key learning points 83Recommended further reading 836 Analysing Data from a Delphi and Reporting Results 84Introduction 84Analysis of Round 1 85Content analysis – the practical aspects 85Demographics for sample profiling 86Analysis of middle rounds (Round 2) 86Consensus level 86Statistical analysis 86Statistical feedback to panel 87Exclusion of items with consensus 88Analysis of end round (Round 3) 89Determining the end of the process 89Statistical analysis 89Items that have gained consensus 90Items that have not gained consensus 90Stability of responses 91Examples of statistical analysis used in recent Delphi studies 92Reporting of results from a Delphi study 92Key learning points 95Recommended further reading 957 Reliability and Validity 96Introduction 96Reliability 96Criteria to assess rigour 99Response rates as a measure of rigour 99Is the definitive answer reached? 99Validity 100Content validity 100Criterion-related validity 100Threats to validity 101Key learning points 104Recommended further reading 1048 Ethical Considerations 105Introduction 105Ethical principles 105Respect for human dignity 106Justice 106Principle of beneficence 108Principle of non-maleficence 108The role of the researcher 108Ethics documentation 109Key learning points 113Recommended further reading 1139 A Classical Delphi Design Case Study 114Introduction 114Background 114Methods 115Initial considerations 116Enhancing response rates 117Identifying and accessing the sample 117Round 1 118Round 2 119Discussion 122Lessons learned 123Conclusion 123Acknowledgements 123Further information 124Publications 12410 A Modified Delphi Case Study 125Introduction 125Aims of the study 126Methodology 126Expert panel 126Round 1 – focus groups 127Delphi Round 2 – postal round 127Round 3 – postal round 128Consensus conference 129Results 129Commissioning of health and social care 129Leadership 131Generic and specialist roles 131Clinical governance 132Teamwork 133Public involvement 133Education 134Practical training versus academia 135Attitudes to nurse training and education 135Multidisciplinary education in nurse training 135Communication 135Conclusion 136Recommendations 137Commissioning of health and social care 137Leadership 137Generic and specialist roles 138Clinical governance 138Teamwork 139Public involvement 139Education 140Communication 140Reflections on the modified Delphi 140Acknowledgements 141Further information 141Publications 14111 e-Delphi Case Study 142Introduction 142Sample 142Setting a consensus level 143Theoretical framework 143Data collection and analysis 144Design of instrument 144Pilot study 144Round 1 144Round 2 145Round 3 145Ethical considerations 145Results 146Round 1 146Round 2 146Round 3 147Discussion 147Conclusion 148Reflections on the e-Delphi 149Acknowledgements 150Further information 150Publications 150Annotated Bibliography 151References 164Index 193

"This book is easy to understand and is excellent for learning the Delphi technique. It is a must for anyone wanting to use this technique." (Doody's, 9 September 2011)