Developing a Dementia Registry: a descriptive case study from North Thames DeNDRoN and the EVIDEM programme

To describe the development of a dementia research registry, outlining the conceptual, practical and ethical challenges, and to report initial experiences of recruiting people with dementia to it from primary and secondary care.

Background

Women, the oldest old and ethnic minorities have been under-represented in clinical trials in dementia. Such under-representation biases estimates of absolute effect, absolute harm and cost-effectiveness. Research on dementia should include patient populations that more exactly reflect the population at risk. One of the impediments to this is the lack of a suitable tool for identification of patients suitable for studies.

Construction & contents

A technology development methodology was used to develop a registry of people with dementia and their carers. This involved phases of modelling and prototype creation, 'bench testing' the prototype with experts and then 'field testing' the refined prototype in exemplar sites. The evaluation of the field testing described here is based on a case study methodology.

Utility

This case study suggests that construction and population of a dementia research registry is feasible, but initial development is complex because of the ethical and organisational difficulties. Recruitment from primary care is particularly costly in terms of staff time and only identifies a very small number of people with dementia who were not already known to specialist services. Recruiting people with dementia through secondary care is a resource intensive process that takes up to six months to complete. Identifying the components of a minimum dataset was easy but its usefulness for pre-screening potential research populations has yet to be established. Acceptance rates are very high in the first clinic to recruit to the registry, but this may reflect the efforts of registry 'champions'.

Discussion and Conclusions

Easier recruitment may perpetuate potential selection biases and we are not yet able to assess the representativeness of the research-ready population recruited to the registry. The need to recruit from wider populations, through primary and social care, remains. The success of this registry will be measured by the proportion of people from it who are recruited to research projects, and its impact on overall accrual to studies.