Author
BROOKER Dawn;
Title
Dementia care mapping: a review of the research literature.
Journal citation/publication details
Gerontologist, 45 (Supplement), October 2005, pp.11-18.
Summary
This review of 34 studies on dementia care mapping (DCM) as a research and evaluation/practice development tool comes to generally positive conclusions, although identifying many areas in which improvements need to be made. The review was conducted specifically to inform the future development of DCM8, which was launched in late 2005.
Context
DCM is an observational tool used in formal dementia care settings for more than a decade for developing person-centred care practice, and as a tool in quality of life research. Essentially, a mapper observes a group of people with dementia continuously for a representative daytime period, coding at five minute intervals for behaviour (24 domains), well- (or ill)-being value (6-point scale), ‘personal detractions’ (staff behaviours likely to adversely affect personhood), and ‘positive events (that enhance personhood). A review of the literature on DCM was published in 2002, covering studies published up to 2001, and this review aims to update what is known and to inform the future development of the tool.
Methods
What sources were used?
The University of Bradford maintains a bibliographic database (10pp Word document) of all known publications on DCM (see http://www.bradford.ac.uk/health/dementia/dcm/publications/index.php) which is updated annually by the Bradford Dementia Group through searches of Medline, CINAHL (Cumulative Index to Nursing and Allied Health Literature) and PsycINFO using the terms DCM and dementia care mapping. Other sources, including personal correspondence with practitioners and researchers, are also used as sources for material to be included in the database. The author of this review is responsible for the database.
What search terms/strategies were used?
The database is not searchable electronically. Relevant material was extracted manually and covered the period 1993 to March 2005.
What criteria were used to decide on which studies to include?
Eligible studies were all those that specifically examined the efficacy of DCM, or in which DCM was used as a main measure. There were no exclusions on the grounds of methodological design, but purely descriptive papers and dissertations were excluded.
Who decided on their relevance and quality?
All aspects of the review were carried out by the author.
How many studies were included and where were they from?
Thirty-four papers met the inclusion criteria. Geographical origins are not specified in detail, but they do include several UK studies.
How were the study findings combined?
Data were extracted on: settings and size; aims of the study; length of time mapped; sample selection and characteristics; study design; version of DCM used; inter-rater reliability; DCM outcomes; statistical tests; and level of significance. The tables are accessible at http://www.bradford.ac.uk/health/dementia/dcm/DCMLitReviewTables.pdf. Each study was then assigned to one of five categories according to its basic purpose in using DCM.
Findings of the review
The studies used DCM for various purposes: in cross-sectional surveys in a variety of settings (11); to evaluate the impact of interventions on the lives of people with dementia (10); in repeated evaluations to change care practice (6); and in multi-method qualitative evaluations (3). The remaining four studies looked directly at some of the psychometric properties of DCM. The author reports that, despite the variety of studies, ‘there is consistency in what they report in terms of DCM data’, and they are thus able to provide answers, ‘at least in part’, to common questions about DCM.
Does DCM measure quality of care and/or quality of life?
In terms of concurrent validity with other measures, there is some evidence that DCM is related to indicators of quality of care (for example, one study reports a relationship between lower DCM scores and an increase in pressure sores); two proxy quality-of-life measures (Quality of Life AD-Staff, and Alzheimer’s Disease-Related Quality of Life); and the Affect Rating Scale. Although DCM is measures something similar to proxy measures and other observation measures of quality of care and quality of life, it differs from them in that ‘it attempts to measure elements of both’.
Can different mappers use DCM reliably?
Although many of the studies demonstrate that ‘it is perfectly possible to achieve acceptable interrater reliability’, there is a danger of ‘drifts in coding’. This has significant training implications. One way forward is to provide regular checks for all mappers with a ‘gold standard mapper’, while changes need to be made to reduce complexity and ambiguity in codes for the forthcoming DCM8 version.
Does DCM show representative reliability across all people with dementia?
There is evidence to suggest that level of dependency is correlated with DCM scores, specifically that low well-being scores are associated with high dependency levels, although this relationship is not universally demonstrated. However, it is sufficiently strong to suggest that a measure of dependency should be routinely taken alongside DCM evaluations, and the nature of this measure needs to be agreed,
Does DCM change care practice?
The six studies on the use of DCM in repeated evaluations confirm that it can change care by improving measured levels of well-being and reducing the incidence of ‘personal detractions’. However, ‘the mix of papers in this review cannot be taken as a reflection of the way DCM is used generally’ because many practitioners may not publish the results of their activity. To provide more reliable data it will be necessary to test DCM as a tool for practice development in a longitudinal study in which other quality of life measures (i.e. not DCM scores) are used to measure outcomes.
Is DCM a suitable tool for research?
DCM was not designed as a research tool, and there are few studies of reliability and validity. Existing findings on inter-rater reliability, concurrent validity with other proxy measures of quality of life, internal consistency and test-retest reliability need to be replicated. The issue of the impact of dependency and diagnosis on scores also needs to be investigated. From the studies reviewed here, it seems that DCM is particularly suitable for smaller scale within-subjects or group comparison evaluations, and it also has value in enriching data derived from proxy and service-user interviews and focus groups. However, the rules of coding mean that it is likely to under-estimate the occurrence of socially passive and withdrawn behaviour, and researchers may be better advised to use another tool for this purpose.
What do DCM scores mean in terms of benchmarking?
The data from the review suggest that scores are generally higher in day care than long stay care, but it is not clear how this is confounded by different dependency levels. Work is currently underway to develop an international database of DCM results that could be used for benchmarking purposes, and the quality of this needs to be safeguarded by accepting only data that has been verified by a gold standard mapper.
What is a significant change in scores?
Few studies have reported data on the statistical significance of changes in DCM scores. ‘Further research is needed to clarify what constitutes a clinically significant change.’
How long should a map be?
Six hours is the current guidance in DCM training, and most studies (with the exception of those using DCM for practice development where longer periods are involved) use this time period. However, there is some evidence that useful insights can be gained from shorter periods, such as two hours. This is another are requiring further investigation.
Authors' conclusions
‘These studies report evidence that DCM has a role to play in practice development and research within the broad aim of improving the quality of the lived experience for people with dementia.’ The author recommends a controlled, longitudinal study to evaluate fully its impact on practice development and quality of care.
Implications for policy or practice
DCM is already being widely used, and the results of this review will feed into the process of improving its reliability and utility. The author emphasises the fact that it provides ‘a shared language, and focus across professional disciplines, care staff and management teams. It is seen as a valid measure by frontline staff as well as those responsible for managing and commissioning care’. Many of the studies cited in the review were undertaken at the behest of practitioners, and DCM thus provides a major opportunity for dialogue between research and practice in social care.