Improving Cognitive Health in COVID-19 Survivors

Emerging evidence suggests a subgroup of survivors of COVID- 19 have residual difficulties with cognition and daily functioning. These deficits are pronounced in cognitive domains including attention, learning and executive skills, and may continue to impact quality of life after recovery from other COVID-19 symptoms. This study aims to investigate the efficacy of AKL-T01 (Akili Interactive), a remotely-delivered digital cognitive intervention, in targeting and improving cognition and functional outcomes in individuals recovering from COVID-19. The efficacy of the AKL-T01 intervention will be measured relative to a waitlist control group.

• AKL-T01 (Device): AKL-T01 will be administered as a 6-week intervention. It is an algorithmically delivered iPad-based video game designed to improve cognitive health by targeting attention and attentional control processes.

Primary outcomes

• Change in Cognitive Function, as Measured by Number Correct on the Digit Symbol Matching Task [Baseline and Post Treatment (6 weeks)]: Change in cognitive function in the experimental group vs. controls, measured by score on the Digit Symbol Matching Task, a timed measure of attention and processing speed. Participants are asked to correctly match pairs of shapes and numbers. Scores exceeding the normative mean number correct (mean = 65.79) reflect better task performance and scores below the normative mean reflect poorer performance.

Secondary outcomes

• Change in Daily Functioning, as Measured by the NeuroQOL Cognitive Function Scale [Baseline and Post Treatment (6 weeks)]: Change in scores on the NeuroQOL Cognitive Function scale in the experimental group vs. controls. The NeuroQOL Cognitive Function scale is a 28-item self-report measure of daily functioning. Higher scores, defined as those exceeding the population mean (mean (SD) = 50.09 (10.23)), reflect better self-reported daily cognitive abilities, while lower scores, defined as those falling below the population mean, reflect poorer self-reported daily cognitive abilities.
• Change in Cognitive Performance, as Measured by Proportion Correct on the Multiple Object Tracking Test [Baseline and Post Treatment (6 weeks)]: Change in scores on the Multiple Object Tracking task in the experimental group vs. controls This task measures sustained attention, cognitive control, and working memory, and requires participants to remember and track a set of target circles as they move around the screen, among a larger set of identical distractor circles. The outcome measure is the proportion of dots that a participant was able to track and identify successfully (on a scale of 0 to 1, with 0 being no dots and 1 being all of the dots). Higher proportion reflects better performance, while lower proportions reflect poorer task performance.
• Change in Cognitive Performance, as Measured by Digit Span on the Digit Span Backwards Test [Baseline and Post Treatment (6 weeks)]: Change in scores on the Test My Brain (TMT) Digit Span Backwards task in the experimental group vs. controls. This task measures sustained attention and working memory. Participants are shown a series of numbers and asked to reproduce them in reverse order. Higher digit spans, defined as those exceeding the population mean (M = 5.98) reflect better performance, while lower digit spans reflect poorer task performance.
• Change in Cognitive Performance, as Measured by the Simple Reaction Time Test [Baseline and Post Treatment (6 weeks)]: Change in scores on the Simple Reaction Time task in the experimental group vs. controls This task measures simple reaction time and psychomotor speed, and requires participants to press a key whenever a green square appears. Faster response times indicate better performance than slower response times.
• Change in Cognitive Performance, as Measured by the Choice Reaction Time Test [Baseline and Post Treatment (6 weeks)]: Change in scores on the Choice Reaction Time task in the experimental group vs. controls This task measures processing speed and cognitive control, and requires participants to indicate the direction of an arrow that is a different color from the rest. Faster response times indicate better performance than slower response times.
• Change in Cognitive Performance, as Measured by Proportion Correct on the Letter-Number Switching Test [Baseline and Post Treatment (6 weeks)]: This task measures sustained attention and set-shifting, and requires participants to switch between responses to letters and numbers. The outcome measure is proportion of correct switch trials, indicating that participants successfully switched from a letter response to a number response, or vice versa (on a scale of 0 to 1, with 0 being no correct switches and 1 being correct switches on all switch trials). Higher proportion correct for switch trials indicates better set-shifting performance, while lower proportion correct indicates worse set-shifting performance.
• Change in Cognitive Performance, as Measured by D-prime on the Gradual Onset Continuous Performance Test [Baseline and Post Treatment (6 weeks)]: This task measures sustained attention and response inhibition, and requires participants to respond to target stimuli and ignore distractors. Performance is measured by d-prime, a measure of memory sensitivity and discrimination. D-prime is based on the calculation of hits (i.e., responding to a target when a target is present), misses (i.e., not responding when a target is present), false alarms (i.e., responding to a target when a distractor is present), and correct rejections (i.e., not responding when a distractor is present). D-prime scores range from 0 (chance) to 4.65 (based on hit rate of 0.99 and false alarm rate of 0.01). Greater d-prime scores reflect better performance than lower d-prime scores.
• Change in Cognitive Performance, as Measured by Percent Correct on the Visual Paired Associates Task [Baseline and Post Treatment (6 weeks)]: Change in scores on the Visual Paired Associates Task in the experimental group vs. controls. This task measures visual memory, and requires participants to learn and identify image pairs. The outcome measure is proportion of image pairs successfully recalled (on a scale of 0 to 1, with 0 being no image pairs and 1 being all of the image pairs). A higher proportion of correct image pairs reflects better performance than a lower proportion of correct image pairs.
• Change in Overall Daily Functioning, as Measured by the World Health Organization Disability Assessment Scale (WHODAS) 2.0 [Baseline and Post Treatment (6 weeks)]: Change in total score on the 36-item WHODAS in the experimental group vs. controls. The WHODAS 2.0 assesses the following domains of functioning: cognition, mobility, getting along, self-care, participation, and life activities. Scores range from 0 (no disability) to 100 (maximum disability), with higher scores reflecting a greater degree of overall functional disability.

Sex: All, ages: 18 Years to 89 Years

Inclusion Criteria:

* Male or female 18-89 years of age
* Documentation of a deficit in cognitive function (score > 1 standard deviation below normal range) compared to age-adjusted normative data) on at least one screening measure of attention and executive function (Oral Trail Making Test, Stroop Test, or FrSBe)
* Previous diagnosis of COVID-19 confirmed via SARS-CoV-2 polymerase chain reaction (PCR) test (or reported experience of COVID-19 symptoms with a documented positive antibody test or clinical diagnosis based on symptoms and accompanying physician's note) documented in the electronic medical record or in other existing medical records
* Access to and self-report of ability to connect wireless devices to a functional wireless network.
* Ability to follow written and verbal instructions (English) as assessed by the PI and/or coinvestigator.
* Able to comply with all testing and study requirements and willingness to participate in the full study duration

Exclusion Criteria:

* History of neurologic disorder prior to COVID-19 diagnosis, such as Parkinson's disease, multiple sclerosis, Alzheimer's disease, stroke, brain tumor, or dementia.
* History of severe mental illness (e.g., schizophrenia, psychosis, history of suicide attempt in the last year) or substance use disorder, recent history (in the past year) of symptoms of psychosis
* Participant is currently considered at risk for attempting suicide by the Investigator, has made a suicide attempt within the past year, or is currently demonstrating active suicidal ideation or self-injurious behavior.
* Motor condition (e.g., physical deformity of the hands/arms) that prevents game playing as reported by the participant or observed by the Investigator
* Recent history (within 6 months prior to screening/baseline) of substance use disorder
* History of seizures (excluding febrile seizures), a tic disorder, significant tics, a current diagnosis of Tourette's Disorder.
* Color blindness as determined by self-report
* Regular use of psychoactive drugs other than antidepressants or benzodiazepines, including stimulants that in the opinion of the Investigator may confound study data/assessments.
* Any other acute medical condition that may interfere with participation or interpretation of the results
* Previous exposure to AKL-T01.

Participant flow

Outcome results

• Change in Cognitive Function, as Measured by Number Correct on the Digit Symbol Matching Task (Primary), Least squares mean, number of correct matches: AKL-T01 Intervention 63.0 (61.1 to 65.0); Waitlist Control 63.5 (61.6 to 65.5) (p 0.54)
• Change in Daily Functioning, as Measured by the NeuroQOL Cognitive Function Scale (Secondary), Least squares mean, score on a scale: AKL-T01 Intervention 25.4 (24.3 to 26.6); Waitlist Control 25.7 (24.5 to 26.9) (p 0.04)
• Change in Cognitive Performance, as Measured by Proportion Correct on the Multiple Object Tracking Test (Secondary), Least squares mean, proportion of dots tracked: AKL-T01 Intervention 0.72 (0.71 to 74); Waitlist Control 0.73 (0.72 to 0.75) (p 0.79)
• Change in Cognitive Performance, as Measured by Digit Span on the Digit Span Backwards Test (Secondary), Least squares mean, digit span: AKL-T01 Intervention 4.7 (4.4 to 5.1); Waitlist Control 4.9 (4.5 to 5.2) (p 0.45)
• Change in Cognitive Performance, as Measured by the Simple Reaction Time Test (Secondary), Least squares mean, millisecond: AKL-T01 Intervention 450 (431 to 470); Waitlist Control 434 (414 to 453) (p 0.02)
• Change in Cognitive Performance, as Measured by the Choice Reaction Time Test (Secondary), Least squares mean, millisecond: AKL-T01 Intervention 1183 (1135 to 1230); Waitlist Control 1164 (1118 to 1211) (p 0.96)
• Change in Cognitive Performance, as Measured by Proportion Correct on the Letter-Number Switching Test (Secondary), Least squares mean, proportion of correct switch trials: AKL-T01 Intervention 0.93 (0.92 to 0.95); Waitlist Control 0.95 (0.94 to 0.96) (p 0.04)
• Change in Cognitive Performance, as Measured by D-prime on the Gradual Onset Continuous Performance Test (Secondary), Least squares mean, d-prime: AKL-T01 Intervention 2.3 (2.2 to 2.4); Waitlist Control 2.3 (2.1 to 2.4) (p 0.60)
• Change in Cognitive Performance, as Measured by Percent Correct on the Visual Paired Associates Task (Secondary), Least squares mean, proportion of image pairs: AKL-T01 Intervention 0.56 (0.52 to 0.59); Waitlist Control 0.56 (0.52 to 0.59) (p 0.06)
• Change in Overall Daily Functioning, as Measured by the World Health Organization Disability Assessment Scale (WHODAS) 2.0 (Secondary), Least squares mean, score on a scale: AKL-T01 Intervention 70.4 (67.4 to 73.4); Waitlist Control 69.8 (67.0 to 72.7) (p 0.04)

Adverse events

Time frame: Adverse event data were collected in all participants for the duration of enrollment. Adverse event data were collected for 10 weeks in: participants in the intervention arm and participants in the waitlist control arm, including control participants who did not start the intervention. Adverse event data were collected for 6 additional weeks (16 weeks total) in: control participants who started the intervention after the waitlist control arm.

Other adverse events by term

• Study Protocol and Statistical Analysis Plan (PDF, 2022-11-18)

• Faith Gunning, PhD, Principal investigator, Weill Medical College of Cornell University