An ethnographic study to develop a taxonomy of lies for communicating with people with moderate to severe dementia.

Background: There is no definition of what constitutes a lie when working with people with moderate to severe dementia. Lies are often defined as therapeutic with no evidence of how therapeutic value is gauged. There is no previous research that observes lies being told or the impact the lies have on people with dementia.

Aim: The aim was to develop a taxonomy of lies for use when supporting people with moderate to severe dementia and then use this to develop a model, which could be used to explore the impact of lie telling.

Research Design: Ethnography was used, with the researcher observing and recording lies told to people with moderate to severe dementia, by professional caregivers. Data analysis was based on Burnard's thematic analysis.

Ethical Consideration: Ethical approval for the study was obtained from Northumbria University, the Research Ethics Committee (REC) and the Health Research Authority (HRA) in the United Kingdom. Study number 227508.

Findings: Six categories of lie were identified; blatant, avoidance, familiarity, props, banter and going along with. Spontaneous and planned lies were observed. Motivation, validation and genuineness were key in terms of how the person with dementia responded. These affective domains were used to develop the Lie ARM (affective reflective model).

Discussion: Planned and spontaneous lies are an effective intervention for people with moderate to severe dementia. Healthcare professionals need to be aware of their motivation for using lies, as well as the impact they have on patients.

Conclusion: The taxonomy is an objective way to categorise lies. It supports increased reflection, particularly if used in conjunction with the Lie ARM. The Lie ARM is a useful tool to help staff reflect on practice or to predict whether a proposed lie is likely to be therapeutic.