Computerized Multiphasic Interactive Neurocognitive DualDisplay System

Comprehensive Computerized Alzheimer's Disease

Assessment and Characterization

Alzheimer’s Disease (AD) is characterized by decline or change in three distinct domains: cognitive function, activities of daily living and behavioral and psychiatric measures. In controlled clinical trials, evaluation within all three domains is commonly implemented. NeuroComp has developed an enterprise-scale, computerized solution that integrates all requisite elements for comprehensive AD characterization.

Computerization of the ADCS Core Battery

Recognizing the need to optimize AD assessment, the National Institute on Aging (NIA) sponsored the formation of the Alzheimer's Disease Cooperative Study (ADCS), based at the University of California San Diego. The mission of the ADCS is to advance research in the development of treatments in order to “ameliorate behavioral symptoms, improve cognition, slow the rate of decline, or delay the appearance of AD."

A mandate of the ADCS is to select and develop optimal assessment instruments and protocols for clinical trials and related research. Key contributions include development of: 1) an extended version of the Alzheimer’s Disease Assessment Scale (ADAS), 2) a Clinical Global Impression of Change (CGIC) scale and 3) an Activities of Daily Living Inventory (ADLI). These three form the “core” ADCS battery and are the de facto standards for nearly all controlled trials of candidate AD treatments.

In successive studies sponsored by the National Institute on Aging (NIA), NeuroComp Systems is the only firm to have computerized and validated the ADCS core battery. These instruments (and associated data acquisition and processing utilities) have been implemented and integrated on the CMINDS® and the ClinTabTM platforms which host a range of additional cognitive and behavioral instruments and scales, including the Neuropsychiatric Rating Scale (NPI) and the Clinical Dementia Rating (CDR) scale.

Computerized Alzheimer’s Disease Assessment Scale (cADAS®)

Computerized Activities of Daily Living Inventory (cADLITM)

Computerized Clinical Global Impression of Change (cCGICTM)

Additional Computerized AD Assessment Instruments

Alzheimer's Disease Assessment Scale (ADAS)

The Alzheimer's Disease Assessment Scale (ADAS) is the most commonly applied measure of cognition in AD clinical trials. It was introduced in 1983 by Mohs, Rosen, and Davis to assess the "severity of cognitive and noncognitive behavioral dysfunctions that are characteristic of persons with Alzheimer's disease." Thus, it was the first battery or scale specifically designed to quantify change in AD symptomatology and has become the "gold standard" for the evaluation of new treatments for AD.

Computerized Alzheimer's Disease Assessment Scale (cADAS®)*

Under an exclusive licensing agreement with the Mount Sinai School of Medicine (MSSM), NeuroComp CMINDS® hosts the only licensed, computerized version of the Alzheimer's Disease Assessment Scale (cADAS®). The development of the cADAS® was funded by the National Institute on Aging (#AGR43AG19572). This SBIR project included a concurrent validity study which was conducted under the direction of leading members of the Alzheimer's Disease Cooperative Study (ADCS), including Richard Mohs, Ph.D. (one of the primary authors of the ADAS), Pierre Tariot, M.D., Ph.D., David Salmon, Ph.D., and Lon Schneider, M.D.

Concurrent Validity of the cADAS® vs. the Standard, Paper-Based ADAS

The cADAS® was compared to the standard, paper-based ADAS in a sample of 88 subjects previously diagnosed with AD, in a counterbalanced, test-retest design. A summary of the key findings was presented as a poster¥ at the 4th Annual Meeting of the International Society for CNS Clinical Trials and Methodology (ISCTM) held in Washington D.C., February 25-27, 2008. The intraclass correlation (ICC) value for the total ADAS-Cog score was 0.96 (p < 0.001). This value is essentially the same or higher than the sADAS to itself (in published studies, ICCs range from 0.79 - 0.98). The range of ICCs for individual items was 0.78 - 0.93. All were significant at the 99% confidence level or higher. Test-retest reliability comparisons also yielded highly significant (p<.001) measures of absolute agreement on the ICC for the cADAS® Total Score and all cADAS® subtests for both short-term (four month) and long-term (one year) reliability comparisons. These values were actually higher than those obtained on the sADAS for both Total Scores and the majority of the subtest comparisons. Furthermore, a Paired Samples t-test was also conducted to compare the ICC reliability values between the two methods. This revealed that the mean ICC for the cADAS® (0.86) was significantly larger (t=3.01; p<.01) than the mean ICC for sADAS (0.80). Thus, the cADAS® displays significantly greater test-retest reliability than the sADAS.

*Copyright of the ADAS test content, and instructions are the intellectual property of the Mount Sinai School of Medicine (MSSM). NeuroComp Systems, Inc. holds an exclusive copyright license from the MSSM to all electronic and computerized forms of the ADAS assessment battery.

¥Kemp AS, Mohs R, Salmon D, Tariot P, Ismail MS, Sano M, Schneider L, O’Halloran JP (2008). Psychometric Comparison of Computerized and Standard Administration of the Alzheimer’s Disease Assessment Scale (ADAS) among Patients with Mild to Moderate Alzheimer’s Disease. 4th Annual Scientific Meeting of the International Society for CNS Clinical Trial Methodologies; Washington, DC; February 25-27.

The ADCS Activities of Daily Living Inventory (ADLI)

The development of the ADCS-ADLI was motivated by the need for an improved measure of ADL in clinical trials of new treatments. Inventory content was created by a subcommittee of the ADCS, comprised of clinicians with extensive expertise in dementia assessment. A multicenter study determined the structure of the final version.

The ADCS-ADLI relies on purely observed activities and avoids introducing judgment of hypothetical behaviors or opinion. The contribution of examiner judgment or interpretation has also been minimized or eliminated by judicious construction and wording of item response options. Each question or item has distinctive, detailed response options that precisely describe and include or exclude certain functional behaviors. This important characteristic facilitates self administration.

Computerized ADCS-ADLI

The computerized ADLI (cADLITM) supports both examiner-administered (EA) and self-administered (SA) modes. This allows the cADLI to be given in a manner similar to the standard ADLI (sADLI) by an examiner, or in an automated mode. The examiner-administered computerized ADLI (EA-cADLI) is intended to be given to informants who need assistance or may prefer not to directly interact with a computer. The self-administered computerized ADLI (SA-cADLI) mode enables a caregiver to directly answer the ADLI queries, without an examiner being present. The purpose of the SA-cADLI is to reduce staff involvement and consequently, the cost of administration.

The cADLI was evaluated in an NIA-funded study at the UC San Diego ADRC in a group of 54 informants of patients enrolled in the ADRC program.

The concurrent validity of the cADLI forms with the sADLI was assessed using the intraclass correlation coefficient (ICC) and Pearson’s R. Table 2 gives the results of correlations of each computerized form with the sADLI and also a measure of short-term reliability of the computerized forms across a one month period.

The Clinical Global Impression of Change (CGIC) Scale

The CGIC was specifically designed to establish clinically meaningful change, as opposed to any change, in behavioral, cognitive and functional domains over the course of AD and in response to treatments. The CGIC differs from more structured assessments, including most global severity scales, in that the clinician is free to engage the patient or informant on a variety of topics and probe areas of interest over the course of the interview. In practice, interviewers are instructed to assign a rating indicating change when “clinically meaningful and distinct change” is perceived. Development of the ADCS-CGIC was motivated by FDA guidelines specifying CGIC assessment in AD clinical trials and by an urgent need for standardization of CGIC-type instruments. The CGIC was the product of contributions by a subcommittee of the ADCS composed of 37 researchers and clinicians with an average of more than 10 years experience in AD assessment in research and clinical trials. The ADCS-CGIC thus represents the consensus of a group of recognized authorities in the field of AD assessment.

Computerization and Validation of the ADCS-CGIC

The format of the computerized CGIC (cCGICTM) is identical to the standard CGIC (sCGIC) paper forms, using the same page formatting, section spacing, indentation, proportions and dimensions. Importantly, the space for writing and the high resolution of the electronic pen result in an overall look and feel that very closely resembles the paper forms and permits the user to write on and use the electronic forms in a functionally identical fashion. However, the electronic page differs from its paper counterpart in several ways. First, the electronic pen is only able to write in areas or cells intended for writing. Areas such as frames, borders and text labels are insensitive to the pen. Second, only one of the check boxes used to indicate information sources and assign change ratings on the “Summary Sheet” form is selectable at a time (by a single pen tap), preventing multiple (i.e.,ambiguous) entries. Third, the program requires the user to make a selection (i.e., complete the form) before allowing the file/session to be closed. At the conclusion of each interview, the user selects a “Finish” button on the last form page.

The cCGIC was compared to the sCGIC in a sample of experienced clinicians in a study performed at the University of Southern California, Keck School of Medicine (Geriatric Psychiatry Clinic) and at NeuroComp Systems. All participants viewed videotaped standard baseline and six month clinical interviews of 2 mock patients and their respective caregivers, at each of two visits, approximately 7 days apart. Each participant rated each interview with both the ClinTab and traditional paper forms, with the order of administration counterbalanced across participants and between visits. Results: Intraclass Correlation Coefficient (ICC) Pearson's r and Paired t-test comparisons between the number of words yielded highly significant levels of absolute agreement (Table 1), with no significant mean differences. These results provide clear support for the concurrent validity of the ClinTab platform and highlight its potential for paperless data collection in clinical trials.

Additional Computerized Alzheimer's Disease Assessment Measures

CMINDS® hosts a range of additional measures of cognition and behavioral functions. Specifically, NeuroComp has been funded to implement a range of commonly applied instruments for the characterization of Mild Cognitive Impairment (MCI) and other dementia subtypes that necessitate the use of much more sensitive assessments for accurate diagnostic specificity than anything currently employed on a routine basis.

CMINDS® Mental Status Module

The CMINDS® Mental Status Module (MSM) is comprised of a set of items and queries that are common to numerous brief mental status exams and scales. Administration of the CMINDS® MSM enables extraction of substantially equivalent scores for several commonly applied mental status measures. Areas assessed include: orientation to time and place, memory, spatial ability, simple praxis tasks and attention.