Anonymised data for the AQUEDUCT Trials and REDCap Code Book

Worldwide, people with dementia experience acute mental health crises requiring urgent hospital admission. In the UK, crisis teams support cases, but practices vary widely. The lack of a best practice model can lead to poor quality care and increased hospital admissions. We evaluated the effectiveness of a best practice Resource Kit (RK) designed in the AQUEDUCT program through a randomised controlled trial (RCT). We randomised 23 teams managing crisis in dementia (TMCD) across the country into intervention (RK plus usual care) and control (usual care only) arms. We found no significant difference in psychiatric hospital admissions between the intervention and control arms at six months (Incident Rate Ratio: 0.74; 95%CI: 0.37 to 1.48; p-value = 0.397). Results were robust across sensitivity analyses. Limitations included the failure to develop a ‘learning collaborative’ within crisis teams and unprecedented demands on the NHS during the COVID-19 pandemic. Although staff said the RK was useful for improving quality and practice, implementation remained limited.