Implementation of the Cognitive Daisy (COG-D) for improving care planning and delivery for residents with dementia in care homes: results of a feasibility randomised controlled trial.

Background: Many residents in care homes for older adults live with dementia. Understanding the unique profiles of cognitive impairments for each resident is important for person-centred care, yet information about specific cognitive problems is limited, and knowledge varies. This study explored the feasibility of implementing the Cognitive Daisy (COG-D) intervention, which provide a visual summary in the shape of a 15-petal flower derived from the scores on a neuropsychological assessment battery, in care homes for older adults.

Methods: A parallel-group feasibility cluster randomised controlled trial (cRCT) was conducted over 24 months. Eight care homes were randomised in a 1:1 ratio to either usual care plus the Cognitive Daisy intervention (COG-D) or usual care (control). Care staff were trained on how to use Cognitive Daisies and/or on how to conduct the COG-D assessments with residents. Cognitive Daisies were displayed in residents' rooms and included in care plans. COG-D assessments were repeated after 6 months. The primary objective was to explore areas of uncertainty for a future large-scale trial including recruitment rates and intervention implementation and adherence. Secondary objectives were to explore signals of effects in candidate outcome measures for residents and staff, obtained at baseline and 6- and 9-month post-randomisation. A process evaluation explored barriers and facilitators to intervention implementation through care-plan audits (to explore recommendations in response to COG-D assessments), interviews and focus groups with staff, residents and relatives.

Results: Resident recruitment (n = 115) and staff recruitment (n = 99) in 8 care homes exceeded targets (100 and 50, respectively, in 8-10 care homes). Staff training was perceived positively with high completion rates (77.1% and 83.3% for basic and advanced training, respectively). Completion rates were also high for COG-D assessments (75.5% for assessment 1, and 72.5% of these residents completed assessment 2), and COG-D scores remained stable across the two assessment points. No clear signals of effects were found for candidate outcome measures. Number of recommendations in care plans varied across care homes, and interviews/focus groups highlighted several barriers to staff's use of the Cognitive Daisies in daily practice.

Conclusion: Findings indicate trial delivery was feasible. However, the COG-D requires modification if it is to be feasibly implemented in care home settings.

Trial registration: This trial was registered on (date) (ISRCTN15208844).