Ideally, the content of outcome measurement instruments will reflect what key stakeholders’ value, and subsequently evidence and changes in policies and practices will be responsive to what key stakeholders’ value. This is the ideal situation for all fields of research. Unfortunately, in the context of dementia care research for those still living at home, our work suggests this situation is the exception rather than the rule.
What and how to measure: Existing approaches
In the field of dementia there exists a heterogeneity of outcomes and measurement instruments. For example, however quality of life is defined, a recent review finds that there are 45 outcome measures that purport to measure it. The high degree of variation in outcomes and measures does not easily allow comparison for effectiveness and meta-analysis. It certainly makes the evidence base hard to interpret. There is a strong argument for researchers to agree on a minimum use of outcome measures and avoid inconsistency in the choice of measures. Those living with dementia also need to have a say in the matter.
As our systematic review of measurement instruments finds, there is even a great deal of variation and difference in some of the previous endeavours that have sought to attain consensus on outcomes and measures. The involvement of people living with dementia in these consensus exercises has tended to be either non-existent or tokenistic – perhaps reflective of wider societal issues. Unfortunately, in dementia research outcome measurement instruments do not reflect what people living with dementia consider to be important. Historically it has been those from professional groups (i.e. researchers and health professionals) who have dominated the development and choice of outcome measures. Until fairly recently there has been little role for people with lived experience when it comes to developing measures or choosing them. The “nothing about us, without us” hadn’t quite percolated through, to this area of study.
Starting at the beginning
In our core outcome set study we started at the beginning and shelved all existing assumptions about outcomes and how to measure them. We harnessed the participation of people living with dementia and care partners as co-researchers to design innovative and inclusive methods. Fifty-four outcomes were extracted during a literature review and identified from qualitative research with key stakeholders (people living with dementia, care partners, health and social care professionals, policy- makers and researchers). Over 300 people from these key stakeholder groups then participated in a two round Delphi survey and consensus workshop to attain agreement on the core outcomes. Of the fifty-four outcomes, 13 were regarded as core. These 13 outcome items have substantive overlap with the emergent concept of social health.
What to measure: Core outcome items for dementia research
|Domain, outcome (lay outcome term), lay description of outcome
Friendly Neighbourhood and Home
1. Importance of relationships – Continuing good relationships with people who are important to you
2. Communication – Being able to communicate with others
3. Feeling safe and secure – Feeling safe and secure at home
4. Feeling valued and respected by others – Feeling valued and respected by others
5. Meaningful activities – Being able to do things that you enjoy and want to keep doing
Self-Managing Dementia Symptoms
6. Apathy/Indifference (Losing interest) – Keeping interested in things you like
7. Alertness – Being aware of your surroundings indoors and outdoors
8. Understanding time and place (Knowing where you are) – Being able to find your way around a familiar place
Quality of Life
9. Hygiene and comfort (Personal hygiene & cleanliness) – Being as clean and comfortable as you would like
10. Stability (Falls) – Not falling at home or when out and about
11. Vision and hearing – Being able to see, hear and understand
12. A sense of who you are – Feeling able to keep your identity
13. Having a laugh – Feeling able to have a laugh with other people
How to measure
In the final part of the study we undertook a systematic review of relevant outcome measurement instruments to compare the content of measurement instruments to the scope and focus of the 13 core outcomes. The purpose of this was to recommend how the core outcome items can be measured. The research team compared all 13 core outcome items against the items in outcome measurement instruments. We identified 347 outcome measurement instruments through our searches, and 76 met the inclusion criteria for assessment.
We found an incredible level of dissonance between the 13 core outcomes and the contents of existing outcome measurement instruments. Many of the most frequently used and known measures faired particularly poorly. The ‘closest’ available outcome measure to the core outcome set, by some considerable distance, is the recently developed ‘Engagement and Independence in Dementia Questionnaire’ (EID-Q). However, on our assessments it only covers just over half of the 13 core outcome items.
Informing research, practice and policy agendas
The 13 outcome items distil the aspects of life considered most important. They emphasise the importance of living well with dementia, keeping communicating, keeping as active as possible and having a laugh. We would love these items not only to direct what gets measured but also what services get commissioned, supported and provided. There’s never been a better time to keep our focus on what matters most to people living with dementia and this needs to be translated into outcome measures that are used in research.
Our research highlights that existing outcome measurement instruments do not really reflect what key stakeholders’ value. Why have we got into this state of affairs? Is it because measures were developed by researchers who were targeting people living with dementia at a later stage of dementia? Or were they developed at a time when living well with dementia wasn’t an everyday expectation? If you do not measure what is important to people with lived experience, can the evidence base really claim to improve or enhance the lives of people living with dementia? These are uncomfortable questions. But one which researchers and funders need to consider. However, there is hope is on the horizon. Our research highlights the need for outcome measures to be looked at it light of what matters to people living with dementia. The development of an outcome measurement instrument based on the 13 core outcomes, and it’s use as a minimum in trials, has the potential to enhance comparisons for effectiveness and reconfigure the evidence base around what key stakeholders’ value. We think this would be a welcome paradigm shift for dementia research.
Dr Andrew Harding is a social scientist and Senior Research Associate at Lancaster University. His primary research interest is on information-giving practices of formal health and social services relating to ageing. Prior to his role at Lancaster University, Andrew worked at the Faculty of Health & Social Sciences at Bournemouth University, where he completed his PhD in the area of information-giving – a realist evaluation of a telephone housing options service relating to specialist housing for older people. In his spare time Andrew enjoys consuming craft ale, and maxing fancy drinks in pineapple shapes barware.
Professor Siobhan Reilly has a background is in health services research. Her research is primarily concerned with improving the evidence base for the delivery of health and social care services for people with people with dementia and older people with long term conditions and people with severe mental illness. Her work includes the policy, practice, organisation and delivery of health and social care services, particularly in relation to the integration and coordination of health and social care services, utilising mixed methods approaches. Siobhan recently left Lancaster University to return to the place where she began her academic career, University of Bradford, where she has taken the role of Director of the Centre for Applied Dementia Studies.