PROJECT SUMMARY Delirium is a form of acute brain failure that occurs in 17% of older emergency department (ED) and 25% of older hospitalized patients. Numerous studies have reported that delirium is significantly associated with accelerated cognitive decline, particularly in patients with Alzheimer’s disease and related dementias (ADRD). Even in cognitively intact older patients, 50% will develop incident ADRD at 6-months after an episode of delirium. While interventions that efficaciously prevent delirium exist, there is no universally accepted intervention for delirium after it has developed. Most studies have demonstrated that they have minimal effect on outcomes including cognition. We posit that the lack of efficacy exists because these interventions modify risk factors for delirium development rather than the adverse outcome itself. Additionally, these interventions are uniformly administered to all delirious patients without taking into account this syndrome’s heterogeneity. Many interventions are also initiated 24 to 48 hours after admission, further limiting their efficacy. To preserve long-term cognition after delirium has developed, we must develop a novel and innovative approach to delirium management that emphasizes early (<24 hours) identification of delirious patients at higher risk for poorer long-term cognition and early modification of risk factors that are associated with accelerated cognitive decline. Unfortunately, the data needed to develop such an approach is lacking. To address this dearth in data, we completed several pilot studies that demonstrated that initiating cognitive training during hospitalization and following up with cognitive rehabilitation for 12 weeks after hospital discharge can be feasibly implemented and may improve cognitive outcomes. We also identified that delirious patients with normal arousal, more severe symptoms, metabolic, infectious, and central nervous system etiologies, and higher serum biomarkers of systemic inflammation, endothelial dysfunction, and blood brain barrier injury were also more likely to have poorer 6-month cognition. It is possible that performing the cognitive intervention in delirious patients at higher risk for poorer cognition may improve its efficacy and efficiency. To build upon our pilot studies, we propose this R01 with the following specific aims: (1) Using a randomized control trial design, determine if early (<24 hours) cognitive training performed twice daily during hospitalization and cognitive rehabilitation performed weekly for 12-weeks post-hospital discharge are associated with improved 6-month global cognition in older delirious patients with and without ADRD. (2) Determine whether the cognitive training / rehabilitation intervention is associated with improved network connectivity in the frontoparietal cortex at 6-months as determined by functional magnetic resonance imaging (fMRI) compared with controls. (3) Perform latent class analysis to identify delirium phenotypes and determine if they modify the relationship between early cognitive training / rehabilitation and 6-month global cognition. To complete these specific aims, this R01 will enroll a total of 336 patients and of these, approximately 235 will have pre-existing ADRD.

NARRATIVE Most delirium treatments developed to date have been ineffective in improving outcomes such as accelerated cognitive decline, because they modify risk factors for delirium development and are uniformly administered to all patients 24 to 48 hours after admission. We propose this randomized control trial to determine if early (<24 hours) cognitive training and rehabilitation improve 6-month global cognition in older delirious patients with and without Alzheimer’s disease and related dementias. To identify patients who may benefit more from the cognitive intervention, we will also perform latent class analysis to identify novel delirium phenotypes, and determine if they modify the relationship between the cognitive intervention and 6-month global cognition.