Components, impacts and costs of dementia home support: a research programme including the DESCANT RCT
Clarkson P., Challis D., Hughes J., Roe B., Davies L., Russell I., Orrell M., Poland F., Jolley D., Kapur N., Robinson C., Chester H., Davies S., Sutcliffe C., Peconi J., Pitts R., Fegan G., Islam S., Gillan V., Entwistle C., Beresford R., Abendstern M., Giebel C., Ahmed S., Jasper R., Usman A., Malik B., Hayhurst K.
BackgroundOver half of people with dementia live at home. We know little about what home support could be clinically effective or cost-effective in enabling them to live well.ObjectivesWe aimed to (1) review evidence for components of home support, identify their presence in the literature and in services in England, and develop an appropriate economic model; (2) develop and test a practical memory support package in early-stage dementia, test the clinical effectiveness and cost-effectiveness of routine home support in later-stage dementia and design a toolkit based on this evidence; and (3) elicit the preferences of staff, carers and people with dementia for home support inputs and packages, and evaluate the cost-effectiveness of these approaches in early- and later-stage dementia.DesignWe undertook (1) an evidence synthesis, national surveys on the NHS and social care and an economic review; (2) a multicentre pragmatic randomised trial [Dementia Early Stage Cognitive Aids New Trial (DESCANT)] to estimate the clinical effectiveness and cost-effectiveness of providing memory aids and guidance to people with early-stage dementia (the DESCANT intervention), alongside process evaluation and qualitative analysis, an observational study of existing care packages in later-stage dementia along with qualitative analysis, and toolkit development to summarise this evidence; and (3) consultation with experts, staff and carers to explore the balance between informal and paid home support using case vignettes, discrete choice experiments to explore the preferences of people with dementia and carers between home support packages in early- and later-stage dementia, and cost–utility analysis building on trial and observational study.SettingThe national surveys described Community Mental Health Teams, memory clinics and social care services across England. Recruitment to the trial was through memory services in nine NHS trusts in England and one health board in Wales. Recruitment to the observational study was through social services in 17 local authorities in England. Recruitment for the vignette and preference studies was through memory services, community centres and carers’ organisations.ParticipantsPeople aged > 50 years with dementia within 1 year of first attendance at a memory clinic were eligible for the trial. People aged > 60 years with later-stage dementia within 3 months of a review of care needs were eligible for the observational study. We recruited staff, carers and people with dementia for the vignette and preference studies. All participants had to give written informed consent.Main outcome measuresThe trial and observational study used the Bristol Activities of Daily Living Scale as the primary outcome and also measured quality of life, capability, cognition, general psychological health and carers’ sense of competence.MethodsOwing to the heterogeneity of interventions, methods and outcome measures, our evidence and economic reviews both used narrative synthesis. The main source of economic studies was the NHS Economic Evaluation Database. We analysed the trial and observational study by linear mixed models. We analysed the trial by ‘treatment allocated’ and used propensity scores to minimise confounding in the observational study.ResultsOur reviews and surveys identified several home support approaches of potential benefit. In early-stage dementia, the DESCANT trial had 468 randomised participants (234 intervention participants and 234 control participants), with 347 participants analysed. We found no significant effect at the primary end point of 6 months of the DESCANT intervention on any of several participant outcome measures. The primary outcome was the Bristol Activities of Daily Living Scale, for which scores range from 0 to 60, with higher scores showing greater dependence. After adjustment for differences at baseline, the mean difference was 0.38, slightly but not significantly favouring the comparator group receiving treatment as usual. The 95% confidence interval ran from –0.89 to 1.65 (p = 0.56). There was no evidence that more intensive care packages in later-stage dementia were more effective than basic care. However, formal home care appeared to help keep people at home. Staff recommended informal care that cost 88% of formal care, but for informal carers this ratio was only 62%. People with dementia preferred social and recreational activities, and carers preferred respite care and regular home care. The DESCANT intervention is probably not cost-effective in early-stage dementia, and intensive care packages are probably not cost-effective in later-stage dementia. From the perspective of the third sector, intermediate intensity packages were cheaper but less effective. Certain elements may be driving these results, notably reduced use of carers’ groups.LimitationsOur chosen outcome measures may not reflect subtle outcomes valued by people with dementia.ConclusionsSeveral approaches preferred by people with dementia and their carers have potential. However, memory aids aiming to affect daily living activities in early-stage dementia or intensive packages compared with basic care in later-stage dementia were not clinically effective or cost-effective.Future workFurther work needs to identify what people with dementia and their carers prefer and develop more sensitive outcome measures.Study registrationCurrent Controlled Trials ISRCTN12591717. The evidence synthesis is registered as PROSPERO CRD42014008890.FundingThis project was funded by the National Institute for Health Research (NIHR) Programme Grants for Applied Research programme and will be published in full in Programme Grants for Applied Research; Vol. 9, No. 6. See the NIHR Journals Library website for further project information.