Optimal Timing of Computerized Cognitive Training for Older Intensive Care Unit Survivors (OPTIMIZE)

More than 60% of intensive care unit (ICU) patients are adults ages 60 and older, who are at high risk for ICU-acquired cognitive impairment. After ICU discharge, ICU survivors often experience sleep disturbances and inactivity, and almost 80% of ICU patients experience disturbances in circadian rhythm, which may affect cognitive function. Understanding the optimal, chronotherapeutic timing of cognitive interventions is crucial to promote circadian realignment and cognitive function, and may improve intervention feasibility, acceptability, and efficacy. Specific Aim 1 will determine feasibility, acceptability, and preliminary effect sizes for: 1) a morning session of a computerized cognitive training intervention [COG]; and 2) a late afternoon/early evening session of the COG intervention; compared to 3) standard inpatient care/usual care [UC]. Specific Aim 2 will examine circadian rhythm parameters to determine the optimal timing of the daily COG intervention. Exploratory Aim 3 will explore if the effects of the COG intervention on cognitive function are mediated by daytime activity, and explore if selected biological and clinical factors moderate intervention effects on cognitive function.

Over 60% of intensive care unit (ICU) patients are older adults (ages 60 and older). Up to 40% of ICU survivors experience cognitive impairment that is comparable in severity to moderate traumatic brain injury, while 25% have symptoms similar to mild Alzheimer's disease after hospital discharge. Older ICU survivors are at high risk for ICU-acquired cognitive impairment, often leading to a protracted recovery in a care facility. Several factors, including circadian misalignment (observed in about 75-80% of ICU patients), may decrease effects of interventions designed to improve cognitive function. Further, daytime activity is essential for recovery from critical illness and to promote circadian realignment, yet ICU survivors experience profound inactivity. The scientific premise of the proposed research is that identifying the optimal circadian timing of cognitive interventions for older ICU survivors may improve intervention feasibility, acceptability, and efficacy. Interventions targeting symptoms (i.e., disturbances in circadian rhythm and cognitive impairment) may deliver similar outcomes across conditions that require ICU admission (e.g., cancer, heart failure, pneumonia, hip fracture). To date, interventions have not been evaluated in older ICU survivors that simultaneously target circadian misalignment and cognitive impairment. Moreover, because circadian misalignment adversely affects cognitive function, understanding the optimal timing of cognitive interventions is crucial to promote both circadian realignment and cognitive function. Specific Aim 1 will determine feasibility, acceptability, and preliminary effect sizes for: 1) a morning session of a computerized cognitive training intervention [COG]; and 2) a late afternoon/early evening session of the COG intervention; compared to 3) standard inpatient care/usual care [UC]. Specific Aim 2 will examine circadian rhythm parameters to determine the optimal timing of the daily COG intervention. Exploratory Aim 3 will explore if the effects of the COG intervention on cognitive function are mediated by daytime activity, and explore if selected biological and clinical factors moderate intervention effects on cognitive function. The proposed early-stage clinical trial will be the first to evaluate chronotherapeutic timing of a computerized cognitive training intervention for hospitalized older ICU survivors after ICU discharge, to initiate early cognitive recovery on a post-ICU unit.

Study personnel evaluating cognitive function outcomes will use standardized measures from the NIH Toolbox Cognition Battery and will be blinded to participants' group assignments.

30-minute morning session of a computerized cognitive training intervention (Lumosity), delivered between the hours of 09:00 AM - 11:00 AM, in addition to UC

30-minute afternoon/evening session of a computerized cognitive training intervention (Lumosity), delivered between the hours of 16:00 PM - 18:00 PM, in addition to UC

Participants who are randomized to the COG-AM intervention group will be asked to complete 30-minute computerized cognitive training (Lumosity cognitive training program) sessions during morning hours (between 09:00-11:00), for a total of 7 days or until hospital discharge, whichever comes first. Participants will complete cognitive training sessions in each of the following subdomains: memory, attention, problem solving, cognitive flexibility, and processing speed.

Participants who are randomized to the COG-PM intervention group will be asked to complete 30-minute computerized cognitive training (Lumosity cognitive training program) sessions during late afternoon/early evening hours (between 16:00-18:00), for a total of 7 days or until hospital discharge, whichever comes first. Participants will complete cognitive training sessions in each of the following subdomains: memory, attention, problem solving, cognitive flexibility, and processing speed.

A retention rate of 80% will determine feasibility: the percentage of participants who complete all COG-AM sessions and cognitive outcome assessments. Adherence to the COG-AM intervention will be defined as the proportion of minutes spent training by the total minutes required per protocol (total minutes trained / [30 minutes x number of intervention days]).

A retention rate of 80% will determine feasibility: the percentage of participants who complete all COG-PM sessions and cognitive outcome assessments. Adherence to the COG-PM intervention will be defined as the proportion of minutes spent training by the total minutes required per protocol (total minutes trained / [30 minutes x number of intervention days]).

Likert-type surveys including open-ended questions will be collected regarding acceptability of the COG-AM intervention. To evaluate tolerability, reasons why participants were unable to complete COG-AM sessions in their entirety will be recorded.

Likert-type surveys including open-ended questions will be collected regarding acceptability of the COG-PM intervention. To evaluate tolerability, reasons why participants were unable to complete COG-PM sessions in their entirety will be recorded.

Fully-corrected T scores on selected assessments from the NIH Toolbox Cognition Battery (Flanker Inhibitory Control and Attention Test, Dimensional Change Card Sort Test, List Sorting Test, and Pattern Comparison Processing Speed Test) will be conducted to estimate effect sizes of the interventions on cognitive function.

Continuous body temperature will be recorded via a wearable sensor. Circadian rhythm parameters (amplitude, acrophase, mesor) of continuous body temperature will be examined using cosinor analyses to explore the optimal chronotherapeutic timing of the daily computerized cognitive training sessions.

Activity counts will be recorded via wrist actigraphy. Activity counts per minute during daytime hours will be calculated to explore the effect of the interventions on daytime activity.

Estimated total sleep time will be recorded via wrist actigraphy. Total sleep time in minutes during nighttime hours will be calculated to explore the effect of the interventions on nighttime sleep.

Inclusion Criteria: Age ≥ 60 years ICU length of stay ≥ 24 hours Active transfer order or expected discharge from ICU to a post-ICU unit Fluent in English Functional independence prior to hospital admission (Katz Index = 6) No suspicion of Alzheimer's disease/dementia Current hospitalization at University of Washington Medical Center Exclusion Criteria: Multiple ICU stays throughout present hospital admission Documented history or suspicion of Alzheimer's disease/dementia, or current prescription of anti-dementia medication Documented history of bipolar disorder or schizophrenia (due to the effect of medications on cognitive function and circadian rhythm) Documented acute stroke or acute traumatic brain injury Severe visual impairment (COG intervention requires iPad participation) Severe dominant arm paresis/paralysis (actigraphy measurement) Transfer from outside hospital, rehabilitation, or care facility