Background: Delirium, characterised by an acute and fluctuating change in awareness, attention and cognition, is often precipitated by acute illness. Delirium is associated with long-term cognitive decline and subsequent dementia. Delirium may therefore be a potential modifiable factor for dementia. However, the mechanism underlying the delirium-dementia relationship remains unclear. ORCHARD-PS, a bi-centre study, aims to investigate the pathophysiology of delirium and future dementia after acute illness, with acquisition of multimodal biomarker (clinical, blood and neuroimaging) and follow-up to 3 years. In an interim analysis, we determined the baseline characteristics of the ORCHARD-PS cohort from Oxford, including delirium occurrence, cognition and frailty.

Method: Acute medical patients aged >65 years without advanced dementia, terminal illness or nursing home residence were eligible. Consent/consultee agreement was required. Delirium was diagnosed according to DSM-V criteria and subsyndromal delirium (SSD) was considered if full criteria were not met. Montreal Cognitive Assessment (MoCA<26=abnormal) was performed as soon as possible after enrolment or deferred until feasible in patients who were too unwell or had delirium. Pre-admission functional ability was determined using the Clinical Frailty Scale (CFS>5=frail). Blood banking and neuroimaging were undertaken for biomarker analysis.

Results: Of 306 participants (mean age/SD=81.9/6.9 years, 152 [49.7%] female, 34 [11.1%] consultee declaration), 66 (22%) developed delirium, and 28 (9%) had subsyndromal delirium. Pre-existing dementia diagnosis was present in 20 (7%) participants (n=11 delirium/SSD and n=9 without delirium). Frailty was more common in delirium/SSD vs no delirium (69/94 [73%] vs 76/212 [36%], p<0.001). Full MoCA was completed in 261 (85%). MoCA was lower in delirium/SSD vs no delirium (median MoCA 18 [IQR 13-22] vs 23 [IQR 20-26], p<0.001). Blood banking was completed in 243 (79%) and neuroimaging was performed in 234 (77%) participants.

Conclusions: ORCHARD-PS participants comprise an older population with a high prevalence of physical and cognitive frailty, at overall high risk of dementia/accelerated cognitive decline. Delirium/SSD occurred in one-third of ORCHARD-PS participants. Our target sample of 620 will provide 190 delirium/SSD cases. Over three-quarters of participants completed blood banking and neuroimaging, enabling future multimodal biomarker studies of delirium and dementia in an under-represented population in whom data are currently limited.