Skip to main content

Research Repository

Advanced Search

A staff training intervention to improve communication between people living with dementia and health-care professionals in hospital: the VOICE mixed-methods development and evaluation study


Background: Twenty-five per cent of hospital beds are occupied by a person living with dementia. Dementia affects expressive communication and understanding. Health-care professionals report a lack of communication skills training.
Objectives: To identify teachable, effective strategies for communication between health-care professionals and people living with dementia, and to develop and evaluate a communication skills training course.
Design: We undertook a systematic literature review, video-recorded 41 encounters between staff and people with dementia, and used conversation analysis to investigate communication problems and solutions. We designed a communication skills training course using coproduction and multiple pedagogic approaches. We ran a pilot, followed by six courses for health-care professionals. We measured knowledge, confidence and communication behaviours before, immediately after and 1 month after the course, and undertook interviews with participants and managers. Behaviours were measured using blind-rated videos of simulations.
Setting: General hospital acute geriatric medical wards and two hospital clinical skills centres.
Participants: We video-recorded 26 people with dementia and 26 professionals. Ten experts in dementia care, education, simulation and communication contributed to intervention development. Six health-care professionals took part in a pilot course, and 45 took part in the training.
Results: The literature review identified 26 studies describing 10 communication strategies, with modest evidence of effectiveness. Health-care professional-initiated encounters followed a predictable phase structure. Problems were apparent in requests (with frequent refusals) and in closings. Success was more likely when requests were made directly, with high entitlement (authority to ask) and with lowered contingencies (made to sound less difficult, by minimising the extent or duration of the task, asking patients ‘to try’, offering help or proposing collaborative action). Closings were more successful if the health-care professional announced the end of the task, made a specific arrangement, body language matched talk, and through use of ‘closing idioms’. The training course comprised 2 days, 1 month apart, using experiential learning, including lectures, video workshops, small group discussions, simulations (with specially trained actors) and reflections. We emphasised the incorporation of previous expertise and commitment to person-centred care. Forty-four participants returned for the second training day and 43 provided complete evaluation data. Knowledge and confidence both increased. Some behaviours, especially relating to closings, were more commonly used after training. The course was rated highly in interviews, especially the use of simulations, real-life video clips and interdisciplinary learning. Participants reported that they found the methods useful in practice and were using them 1 month after the course finished.
Limitations: Data were from people with moderate to severe dementia, in an acute hospital, during health-care professional-initiated interactions. Analysis was limited to problems and solutions that were likely to be ‘trainable’. Actors required careful preparation to simulate people with dementia. Communication skills training course participants were volunteers, unlikely to be representative of the general workforce, who displayed high levels of baseline knowledge, confidence and skills. Before-and-after evaluations, and qualitative interviews, are prone to bias.
Conclusions: Requests and closings pose particular difficulties for professionals communicating with people with dementia. We identified solutions to these problems and incorporated them into communication skills training, which improved knowledge, confidence and some communication behaviours. Simulation was an effective training modality.
Future work: Further research should investigate a wider range of health, social care and family carers. Conversation analysis should be used to investigate other aspects of health-care communication.

Journal Article Type Article
Publication Date Dec 28, 2018
Print ISSN 2050-4349
Publisher NIHR Journals Library
Peer Reviewed Peer Reviewed
Volume 6
Issue 41
Pages 1-166
APA6 Citation Harwood, R., O'Brien, R., Goldberg, S., Allwood, R., Pilnick, A., Beeke, S., …Sartain, K. (2018). A staff training intervention to improve communication between people living with dementia and health-care professionals in hospital: the VOICE mixed-methods development and evaluation study. Health Services and Delivery Research, 6(41), 1-166.
Publisher URL

Downloadable Citations