Speed of Processing Training for Cognitive Deficits After Delirium in Older Adults

Overview

In this proposal, we will examine processing speed training (PST) as a potential intervention to reduce the risk of developing long-term cognitive impairment after a delirium episode.

The overall goal of this project is to measure longitudinal cognitive function among hospitalized older adults after an episode of delirium, and examine an intervention (processing speed training, or PST) to slow the rate of cognitive decline that occurs after an episode of delirium compared to an Internet-based control condition (INT). We will initially recruit a cohort of 136 subjects who have experienced delirium, provide PST or INT training to them, and assess them on measures of cognition and instrumental activities of daily living (IADLs) after training and every 6 months thereafter for 24 to 36 months. We will perform a preliminary determination of the efficacy of PST after these 136 subjects have completed training, testing whether subjects receiving PST have better cognitive test scores immediately after training compared to subjects receiving INT. Assuming this preliminary test of efficacy shows short-term benefits for PST, we will recruit an additional 280 subjects who have experienced delirium and provide PST training to them. We will follow all 416 subjects for 24 to 36 months after their delirium episode to test whether subjects receiving PST have slower rates of decline over time compared to subjects receiving INT on measures of cognition and IADLs. A more detailed timeline of study procedures follows. In Hospital (Month 0): We will initially identify subjects with delirium from the inpatient medical units at UAB Hospital. As part of UAB's Virtual ACE quality improvement project, nurses on the medical units administer the Nursing Delirium Screening Scale (NUDESC) to screen for delirium in all patients age ≥ 65 on admission and once per shift thereafter. After obtaining informed consent, we will collect a range of measures from the subject and family to make the diagnosis of delirium and rate its severity, as well as measures examining comorbidities potentially contributing to delirium. These measures for assessing delirium will be * the NUDESC ratings from the medical record; * the Confusion Assessment Method (CAM), a standard clinician-rated instrument for the diagnosis of delirium; * the Delirium Symptom Interview (DSI), an interview administered by a clinician to standardize the assessment of delirium symptoms for rating the CAM; * the Cognitive Test for Delirium (CTD), a general measure of cognition used to standardize the assessment of cognitive symptoms for rating the CAM; * the Family Confusion Assessment Method (FAMCAM), which assesses the family's observation of symptoms of delirium (although this will not be required for delirium diagnosis); * the chart-based review method for delirium (CHART-DEL), which collects information from the medical records about signs and symptoms that can indicate delirium; * the Delirium Etiology Rating Checklist (DERC), a clinician-rated instrument that categorizes the different potential causes of delirium for the subject, based on information in the medical record on other medical conditions, current medications, neuroimaging (CT or MRI, if available), abnormal laboratory values (complete blood count, serum electrolytes and chemistries, liver function tests, and urine drug screen, if available); * the Anticholinergic Burden Scale (ABS), which rates the contribution of the subject's medications to their episode of delirium. Additional measures collected on comorbidities and factors affecting delirium will be * demographics (age, race, gender); * the AD8 and the Modified Blessed Dementia Rating Scale, which assess the family's observations of symptoms of dementia that preceded the episode of delirium; * the Charlson Comorbidity Index (CCI), which assesses the presence of other medical conditions beyond delirium. After the initial in-hospital assessment, we will continue to follow the subject for the remainder of their stay, using the CAM and the DRS-R-98 to monitor the duration of delirium. We will also collect the NUDESC assessments given by nursing staff on a daily basis. Outpatient Baseline (Approximately Month 1, depending on length of hospitalization): Approximately 2 weeks after hospital discharge, we will contact subjects by telephone for follow-up. We will review the study procedures with the subject and use the CAM to see if their delirium has resolved. Once a subject's delirium has resolved, we will bring them to our offices for a comprehensive baseline assessment. Measures for this visit will examine cognition (particularly memory and executive function) and function (particularly activities of daily living and instrumental activities of daily living), as well as general health and mental health. Specific measures will be * the CAM; * the FAMCAM; * the CTD; * the AD8; * the Wechsler Test of Adult Reading (WTAR), a vocabulary test to estimate IQ; * the General Cognitive Performance (GCP) battery, a composite of the following neuropsychological tests * the Visual Search and Attention Test, measuring executive function and visuospatial skills; * the Hopkins Verbal Learning Test-Revised, measuring verbal memory; * the WAIS Digit Span Forward and Backward, measuring attention; * Category Fluency, measuring executive function and language; * Phonemic Fluency, measuring executive function and language; * the Boston Naming Test, measuring naming and language; * the RBANS Digit Symbol Test, measuring executive function and visuospatial skills; * the DKEFS Trail Making Test, measuring executive function and visuospatial skills; * the Katz Activities of Daily Living (ADLs), which rates ability to perform basic activities for daily life; * the Lawton Instrumental Activities of Daily Living (IADLs), which rates ability to perform more complex activities for daily life; * the Timed IADLs, which measures speed at performing complex activities of daily life; * the Geriatric Depression Scale (GDS), a standard measure of depressive symptoms in older adults; * the Medical Outcomes Study 12-item Short Form (SF-12), a standard measure of overall health and quality of life; * healthcare utilization (rehospitalization or institutionalization) since discharge. PST or INT Training (Months 1 and 2): After the baseline assessment, we will randomize each subject to 10 sessions of PST or INT, which will be conducted in the outpatient setting at UAB for standardization. The PST training consists of computer-administered sessions in which a subject practices detection and discrimination of targets displayed for short periods of time. The INT training consists of sessions in which a subject practices general computer skills and use of Internet search engines. Post-Training Assessment (Month 2): After lab-based training, subjects will complete a follow-up assessment that will repeat measures of cognition, function, and general health * the GCP (neuropsychological testing); * the Katz ADLs (functional status); * the Lawton IADLs (functional status); * the Timed IADLs (functional status); * the GDS (mental health); * the SF-12 (general health); * the AD8 (dementia symptoms); * healthcare utilization (rehosopitalization or institutionalization) since baseline. Semiannual Assessments (Months 6, 12, 18, 24, 30, and 36): Subjects will continue to practice PST or INT at home with study-distributed tablet computers and supportive visits for the remainder of the study. Every 6 months, we will repeat measures of cognition, function, and general health * the GCP (neuropsychological testing); * the Katz ADLs (functional status); * the Lawton IADLs (functional status); * the Timed IADLs (functional status); * the GDS (mental health); * the SF-12 (general health); * the AD8 (dementia symptoms AD8); * healthcare utilization (rehospitalization or institutionalization) since previous visit; * incident dementia. For the determination of incident dementia, data on subjects scoring more than 1 standard deviation below the mean on 2 or more neuropsychological tests will be reviewed by a Clinical Consensus Panel, who will assign a diagnosis of mild cognitive impairment or dementia (if warranted) according to National Institute on Aging-Alzheimer's Association criteria. Preliminary Evaluation of Short-Term Treatment Efficacy This evaluation will be examine short-term differences between groups on cognition (GCP) and function (Timed IADLs) at 2 months, after all 136 subjects have completed lab-based PST or INT training. We will also evaluate the effects of pre-existing cognitive impairment by comparing the effects of PST to those without pre-existing cognitive impairment. Using ANOVA to compare treatment groups, a sample size of N=136 will have greater than 80% power to detect a 30% difference in scores on the GCP, which is smaller than the 30% difference considered meaningful in trials for interventions in prodromal Alzheimer's disease. A sample size of N=136 will also be sufficient to allow comparisons between subjects with and without pre-existing cognitive impairment. Recruitment of the Second Cohort Assuming the preliminary evaluation of treatment efficacy demonstrates positive effects for PST, we recruit 280 additional subjects while continuing to follow the original cohort of 136 subjects, so that our total cohort of 416 subjects will provide sufficient power to examine the effect of PST on cognitive and functional declines over time. Enrollment procedures and assessments for new subjects will be identical to those for the original cohort. After completion of PST or INT training, all subjects will be followed for 24 to 36 months (depending on the time of enrollment), with assessments every 6 months as described above. Evaluation of Long-Term Treatment Efficacy We will evaluate efficacy by examining differences in the slope, or rate of change, on cognition (the GCP) and function (Timed IADLs) between the PST and INT groups over the follow-up period. We will also evaluate whether effects of PST on cognitive outcomes differs subjects with pre-existing MCI or mild dementia compared to those without pre-existing cognitive impairment. Using mixed effects models to compare slopes, a sample size of N=416 will have greater than 80% power to detect a difference between treatment groups. This will also be sufficient to allow comparisons between subjects with and without pre-existing cognitive impairment.

Conditions

Interventions

Eligibility

Locations (1)

Outcomes