HQLC Dementia Carers Instrument Development: DECIDE

The number of people living in the UK with dementia is rising. Friends and family provide much of the care needed. Having a caring role may compromise well-being, quality of life, and finances. The annual economic cost of supporting people with dementia is estimated at £26.3 billion, with almost half being attributed to the care provided by 'informal carers'. Without informal carers government costs would be increased greatly. Maintaining carer quality of life and finding a way of measuring it are important not only for the carers and those cared for, but also for society. Having an accurate and meaningful way of measuring carer quality of life is needed in three distinct areas.
1. Individual carer assessment: to identify carers most at need.
2. Development and evaluation of services.
3. Policy making: to inform decision making concerning government spending.
There are numerous quality of life questionnaires with many used in dementia carer assessment. However, some are too general and do not pick up subtle changes in carer quality of life, others are too long and some have unreliable ways of working out scores. Some questionnaires (e.g.EQ-5D) have been used by organisations such as NICE (National Institute for Health and Care Excellence) to make recommendations on how money is spent by working out the 'value' the general population put on different health related quality of life situations (health states). The EQ-5D may be useful to value 'health states' but less useful in valuing 'carer states' since these go beyond health. Furthermore the 'value' the general population put on health/carer states may be different to the 'value' carers put on these states.
Therefore, the aims of this study, undertaken in two workstreams, are to:
-develop a quality of life questionnaire for carers of people with dementia (Carer-QoL) for use as described above in points 1-3,
-test how justifiable the underlying rules used by NICE are in 'valuing' benefits to carers,
-estimate preliminary 'values' for Carer-QoL.
Workstream 1: developing the new questionnaire (3 phases)
To capture the 'carer' experience the questionnaire items will be obtained directly from carers (phase 1 and 2 with 72 carers). The questionnaire scoring will be worked out in phase 3 in which 400 carers will be asked to complete the new questionnaire items and to complete the EQ-5D. The Carer-QoL will be created from the longer list of questions (without losing score accuracy and range) using an approved method called Rasch analysis.
Workstream 2: valuation and calibration (3 studies)
In study one, 1,000 people (general population, carers of people with dementia and other carers) will value different EQ-5D health states. These health states will include their own (real) as well as states previously reported by other carers (hypothetical) in national surveys. Values for real and hypothetical health states will be compared across each group. In the second study the scores from the EQ-5D and the Carer-QoL, collected in Workstream 1, will be jointly analysed to calibrate the Carer-QoL against EQ-5D scores. In the third study carer profiles will be generated using responses collected using the Carer-QoL in Workstream 1. Another 1,000 people (general population and carers) will be asked to select which carer profile they prefer from pairs of carer profiles presented in a web-based survey. From this it will be possible to place a value on the Carer-QoL scores from both a carer and a general population view point.
Beneficiaries from the research outcomes include: carers, people being cared for, service providers, policy makers and other researchers. This research will inform the debate on the value of health and quality of life and could contribute new empirical evidence bearing on the issue of perspective in determining that value. This will benefit decision-making bodies such as NICE as well as those submitting evidence on effectiveness and cost-effectiveness.

DECIDE has 3 objectives:
1. develop a quality of life measure for carers of people with dementia (Carer-QoL)
2. test key assumptions embedded in the current approach to measuring the benefits of healthcare interventions and
3. estimate preliminary values for the Carer-QoL.

The project comprises 2 workstreams (psychometric and valuation) running in parallel.

Workstream 1 psychometric (3 phases). Phase 1, item pool generation employing a needs-based approach through interviews of 42 purposively sampled carers. Phase 2, Item refinement following cognitive interviews with 24 carers. Phase 3, Psychometric evaluation employing Rasch methods to ensure the final item set is unidimensional and satisfies requirements necessary to enable transformation to interval scaling. Carers (n=400) will complete the item pool and the EQ-5D with a subset re-completing the item pool for test-retest reliability and sensitivity to change analyses.

Workstream 2 valuation (3 studies). Study 1 compares health state values (EQ-5D derived) in carers and non-carers (n=1,000) using online surveying of self-reported health status using visual analogue scale rating and presenting descriptions of 8 "hypothetical" health states selected from the most frequently recorded EQ-5D health states. Study 2 will estimate the relationship between a draft Carer-QoL (Workstream 1 generated) and the utility-weighted index form of EQ-5D. Study 3 will quantify Carer-QoL values through direct valuation using an online Discrete Choice Experiment. Carer profiles will be generated from the draft carer-QoL items. Carers and non-carers (n=1,000) will select carer profile preference from a list of paired carer profiles. From this it will be possible to place a value on the Carer-QoL scores from both the carer and general population perspective.

DECIDE outputs and data will be available for other researchers, clinical, social and voluntary service organisations, practitioners and government agencies.

This research is expected to generate the following outputs:
1. New, evidence-based insights about how carers of people with dementia describe their experience as a carer.

2. A new, brief, validated quality of life measure (Carer-QoL) designed using modern psychometric methods derived directly from the carers' experiences.

3. An item bank for use in developing:
(a) parallel short forms of the quality of life measure,
(b) a short-form of the measure that generates the best estimates of EQ-5D using social preference weights and 'carer' adjusted weights,
(c) computer adaptive testing.

4. New knowledge concerning how value (experienced and hypothetical) assigned by carers of people with dementia converge with societal value assigned to health state by the general population.

We expect these outputs to impact on the following individuals/organisations.

Service users
1. Carers: by providing services with a measure relevant to carers which is brief and easy to complete (low burden) resulting in potential early identification of diminished or changing quality of life, more timely intervention and provision of support leading to improved carer well-being.

2. People with dementia: by providing the means for services to undertake easy, accurate carer assessment leading to review of support for the carer and person being cared for, potentially resulting in sustained, supported continuity of care in their own home.

Service commissioners/providers
Through provision of a brief, free, reliable and valid carer assessment measure for use in:
- supportive care practice, with the Carer-QoL administered on paper or electronically for one-off assessment or for assessment over time, to aid effective carer assessment and support, leading to improved, targeted service delivery;
- service planning and development and evaluation to provide insight into interventions/services which are of most benefit to carers and to inform future service planning, possibly including economic evaluation.

The providers who would benefit include:
1. NHS funded providers such as primary care, community and acute commissioners. In particular those working on the integration of health and care shared records, including Chief Clinical Information Officer's (CCIOs) and NHS England transformational change leads for carers and dementia.

2. Local Authority commissioned health and care providers - 3rd sector, private sector and public health.

3. Other, regional and national Third Sector organisations (e.g. Alzheimer's Society, Carers Trust, Ageing Networks, Dementia Action Alliance members) and those not directly involved but influencers such as INVOLVE and the National Council for Voluntary Organisations (NCVO).


Policy makers
Though provision of:
- a calibrated measurement tool which has the potential to be used in economic -evaluation of service development and delivery;
- knowledge for challenging key assumptions embedded in the current approach to valuing benefits.

The policy makers who would benefit include the:
1. National Institute for Health and Care Excellence (NICE)
2. Care Quality Commission (CQC)
3. Department for Work and Pensions (DWP)
4. Department of Health (DH)
5. NHS England (NHSE)

Pharmaceutical companies or other companies testing interventions for dementia
Through provision of knowledge and an assessment measure which could be used for testing the effects of treatment for dementia, including assessing carer spill-over impacts.

The research community
Through publications, access to the quality of life measure for future research, new methodology regarding value assignment, collaborations leading to potential increased research capacity.