Computerized Neurocognitive Test (CNT) in mild cognitive impairment and Alzheimer's disease

Currently, computerized batteries are of great value in detecting cognitive impairment. This aim of this review was to compare the computerized neurocognitive batteries used in most studies with cognitive decline over the last 10 years. Using the search words computerized cognitive assessment with: dementia, mild cognitive impairment, and Alzheimer's disease, the CogState, CNS Vital Sings, COGDRAS and Mindstreams batteries were retrieved.


INTRODUCTION
T he use of Computerized neurocognitive tests (CNT) to evaluate cognition has been widely studied and may be the most suitable tool for the early detection of impairments. 1 CNT are able to measure mild degrees of cognitive impairment and can gauge the effectiveness of an intervention. 2 Computerized batteries offer a number of advantages over paper-and-pencil type tests: they are precise, accurate and can be timed to the nearest millisecond. In addition they are easy to administer and score, have greater standardization and multiple parallel versions may also be available, known to reduce practice effects. 3 The advantages of a computerized battery are that it can be applied at bedside using a tablet device, results can be instant, and the tests applied by any person, eliminating exam-iner effects and providing increased reliability. The battery is also language independent and can be used in patients with mild aphasia offering consistent administration and scoring, while computerized tests can also generate alternative forms for repeated testing. 2,4,5 The disadvantages include the absence of the active participation of consulting neuropsychologists analyzing the qualitative performance, and that the limitations in the ability to understand and manipulate information technology can cause a negative effect because elderly tend not to be familiar with it. 2,6 The aim of this review was to compare the features of the computerized neuropsychological batteries used in most studies involving the cognitively impaired over the last 10 years, in order to verify which are most suitable for use in clinical practice within an outpatient clinic.

METHODS
We examined on-line articles published between 2004 and 2014 on the PubMed database or the references of these articles, which cited mainly reliability and validity studies prior to 2004, searching for the most used and most cited battery for dementia, Alzheimer's disease (AD) and Mild Cognitive Impairment (MCI). The descriptors computerized cognitive assessment with: dementia, mild cognitive impairment and Alzheimer's disease were used. Only studies available in English were included.
Exclusion criteria. Articles published before 2004, those addressing other dementias or computer programs used for rehabilitation, were excluded as were articles unavailable online.

RESULTS
A total of 101 articles published between 2004 and 2014 were identified. Sixty-one articles were excluded for not assessing patients with AD and MCI or because the computerized test was cited only once. Some articles appeared more than once. Articles were divided by tests, population, mean age, diagnosis and main results ( Table  1) and organized according to the main characteristics of the tests ( Table 2). CogState. 7 Cogstate Research is a repeatable and sensitive computerized cognitive testing system designed specifically for use in research studies and can evaluate patients aged 6 to 106, which includes: attention deficit hyperactivity disorder (ADHD), fatigue and drug effects, post-operative cognitive dysfunction, MCI, early Alzheimer's disease and dementia, Schizophrenia and mood disorders. CogState was developed as a dementia screening instrument and for assessing concussion and has been shown to be valid, reliable, and sensitive for detecting cognitive impairment. 7,8 Subtests include measures of simple selection and complex reaction times, continuous monitoring, working memory, matching, incidental learning, and associative learning. The subtests are based on playing card formats and written instructions are presented on screen. Responses are made via a computer keyboard represented graphically on the screen, with responses using the "k" key for yes and "d" key for no. The battery requires 15 to 20 minutes to complete. 1 Specialized tasks can assess attention, memory, executive function, as well as language and socialemotional cognition if required. In addition to research studies, the batteries of tasks are used in commercial trials to determine the effect of drugs. Minimal learning effects ensure that participants can be tested repeatedly as often as needed, even multiple times in short periods (over a single day). 7,10 In a study that evaluated healthy older adults (n=105), amnestic MCI (n=48) and AD (n=42) patients over three months, the CogState battery showed high test-retest reliability and stability in all groups and was able to detect AD-related cognitive impairment. 9 Another study showed that in established dementia, the CogState tasks appeared to be sensitive for detecting cognitive impairment. Repeat administration also provided acceptable stability and test-retest reliability with minimal practice effects at short test-retest intervals even on the same day. 10 A study comparing dementia (AD, frontotemporal dementia, and dementia with Lewy bodies), MCI, and healthy controls, found the CogState was able to successfully differentiate dementia patients from control subjects, but achieved minimal differentiation between controls and MCI. Repeat administration also provided acceptable stability and testretest reliability with minimal practice effects at short test-retest intervals. 11 In a study following healthy older adults over a one-year period, CogState proved to be an instrument that can be used repetitively (at 3, 6, 9 and 12 months), differentiating individuals with risk of increased rates of cognitive decline in memory. 12 CNS Vital Signs (VS). 13 Developed as a brief clinical evaluation tool, its tests are familiar and well-established: verbal and visual memory, finger tapping, symbol digit coding, Stroop Test, a test of shifting attention and the continuous performance test. Gualtieri and Johnson 13 published a study of reliability and validity among 1069 individuals aged 7-90 years with psychometric characteristics. The reliability of the tests in the CNSVS battery are very similar to the characteristics of the conventional neuropsychological tests, and the battery was sensitive for detecting the most common causes of cognitive impairment, but should be used as a screening instrument and not as a substitute for formal neuropsychological testing. 13,14 Cognitive Drug Research Computerized Assessment System (COGDRAS). 15 COGDRAS was developed for use in neuropharmacological research. 15 A study with 152 older adults that investigated a combination of tests for the diagnosis of dementia using conventional tests and subtests of two computerized battery: the Poon-Baro-Wens (PBW) battery 16 and COGDRAS, found that computerized tests added very little diagnostic value. 17 This battery is widely used in clinical trials testing  Mindstreams (Neurotrax). 20 A computerized testing system for comprehensive clinical assessment of cognitive impairment, designed primarily for use in the elderly. The battery consists of nine subtests: verbal memory, nonverbal memory, Go-No Go response inhibition, Stroop interference, problem solving, visual spatial imagery, verbal rhyming, verbal naming, staged information processing speed, finger tapping, and visuomotor planning. 1 A study showed the capacity of this battery to discriminate individuals with MCI from cognitively healthy elderly. 20 Another study by the same author found that the Mindstreams battery provided detailed and distinct cognitive profiles of patients with moderate impairment. 21 In a study assessing 2888 patients, 83% rated the test as easy-to-use and patients were divided into nonusers of computers, patients older than 75, and poor performers. 22 In a study with a population of 161 older adults divided into healthy, MCI and mild dementia, the Mindstreams was able to discriminate between demented and non-demented individuals=AUC=0.886 (p<0.001) and between cognitively healthy individuals and noncognitively healthy individuals AUC=0.823 (p<0.001). 22 In another study, patients were divided into two groups, one assessed by the Global Depression Scale (GDS) 23 and another by the Cornell Scale for Depression in Dementia (CSDD). 24 Mindstreams discriminated among MCI, mild AD, and healthy control (HC) partici-pants following covariation for depression scale score in both cohorts, demonstrating that this battery is unaffected by depression. 25 A study was performed in Afro-Americans, comprising 27 MCI and 22 HC subjects. The MCI patients performed poorly compared to HC participants in all domains, with significant differences in memory (p=.003; d=0.96), executive function (p=.046; d=0.64), and overall battery performance (p=.041; d=0.63). 26 The Mindstreams battery was used in a study as a gold standard for a validation of the Hebrew version of the Montreal Cognitive Assessment (MoCA Test) as a screening instrument for the early detection of MCI. 27,28 DISCUSSION Some batteries are good for measuring reaction time (CODGRAS) or detecting AD (CogState), while others are for screening (CNS VS) and are easy-to-use and effective for discriminating between healthy controls, MCI and AD (Mindstreams).
Slow reaction time and memory impairment seem to be the main features of cognitive impairment, especially in AD, and clinical markers should be taken into account in the choice of test. 30 A review article about advances in design for AD in clinical trials compared the most widely used tests, namely, Automated Neuropsychological Assessment Metrics (ANAM), 31 Computer Assessment of Mild Cognitive Impairment (CAMCI), 32 CANS-MCI, 33 36 Mi-croCog and Mindstreams (Neurotrax). Strengths and weaknesses were detected for all tests, but the authors were emphatic in affirming that computerized assessment offers several advantages over pen-and-paper tests, in that they have a high degree of standardization