Reminiscence groups for people with dementia and their family care-givers: pragmatic eight-centre trial of joint reminiscence and maintenance v usual treatment

The following will be assessed at baseline, 3 and 10 months after randomisation:

1. Quality of life for the person with dementia, self-assessed by the Quality of Life in Alzheimer's Disease (QoL-AD), which has been shown to be reliable and valid for people with mild and moderate degrees of dementia. The scale is completed in a structured interview with the person with dementia and covers 13 domains of life quality.

2. Care-giver's mental health, evaluated using the 28 item, self-report General Health Questionnaire GHQ-28 which has been widely used in care-giver research; the Likert scoring system 0-1-2-3 will be used. The scale includes indicators of anxiety, depression, insomnia, social dysfunction and somatic symptoms. This is preferred as the primary care-giver outcome to the Relatives' Stress Scale in this study, in view of its more general focus and wide usage.

The following will be assessed at baseline, 3 and 10 months after randomisation:

1. Autobiographical memory, assessed using an extended version of the Autobiographical Memory Interview (AMI). The extended AMI assesses recall of the person with dementia's personal memories relating to both factual (semantic) information for example, names of schools or teachers and specific incidents. In the trial platform, we validated an additional section on middle-age to retirement, to give systematic coverage to the life-span of our participants.

2. Measure of relationship quality, self-completed by both person with dementia and carer: Quality of the Care-giving Relationship (QCPR). Originally developed in the Netherlands this scale comprises 14 items (with 5 point Likert scales) designed to assess the warmth of the relationship and the absence of conflict and criticism.

3. Depression and anxiety for both people with dementia and carer:
3.1. Cornell Scale for Depression in Dementia (CSDD): A 19-item interviewer administered measure, using information from interview with the person with dementia and their carer. Signs and symptoms are described to the carer as they appear on the scale. Where there is a discrepancy between the carer and clinician's ratings the carer is re-interviewed before the interviewer makes the final judgment.
3.2. Rating for Anxiety In Dementia (RAID): An 18 item rating scale to measure anxiety in a person with dementia based on a structured interview with the carer and the person with dementia.
3.2. Hospital Anxiety and Depression Scale for carer: a 14-item, self-report well-validated scale, which provides an index of both anxiety and depression, and is suitable for use with adults of all ages.

4. Stress specific to the care-giving situation, assessed using the Relative's Stress Scale: self-report scale for the care-giver, contains 15 items rated on a 5-point Likert scale.

5. Quality of life of person with dementia, rated by the care-giver, assessed using the proxy version of the QoL-AD, identical in structure and content to the self-report version above.

6. Costs, using the validated Client Services Receipt Inventory (CSRI). The CSRI has been used extensively in studies of mental health and dementia care and comprehensively gathers data on accommodation, medication and services accepted. In this case, the data collected will reflect the previous 3 months (at baseline and post-treatment) and 7 months (at follow-up).

7. Quality of life of care giver and person with dementia will also be measured using EQ-5D. EQ-5D is a standardised instrument for use as a measure of health outcome, applicable to a wide range of health conditions and treatments. It provides a simple descriptive profile and a single index value for health status. EQ-5D was originally designed to complement other instruments but is now increasingly used as a 'stand alone' measure. EQ-5D is designed for self-completion by respondents and can be used in face-to-face interviews. It is cognitively simple, taking only a few minutes to complete. Instructions to respondents are included in the questionnaire. We did not include the EQ-5D originally, in view of concerns that use of a generic quality of life measure such as EQ-5D might not be sufficiently sensitive for use as the primary outcome measure with people with dementia. Our team has previously used the EQ-5D to evaluate the concurrent validity of the QoL-AD, and the two scales showed moderate correlation (0.54), but rather less of the sample of people with mild to moderate dementia were able to complete it, even though it was administered in an interview. Care-givers will be asked to complete the measure from their own perspective and for the person with dementia. The self-report of the person with dementia will also be obtained wherever possible.