Development of a Computerized Memory Screen Shannon M' Foster, Dev Ashish, Michael A' Kisley, Hasker

1
Development of a Computerized Memory Screen
Shannon M. Foster, Dev Ashish, Michael A.
Kisley, Hasker P. Davis University of Colorado
at Colorado Springs
This work was supported by the Coleman Institute
for Cognitive Disabilities

• INTRODUCTION
• Recent statistics on aging have important
  economic and societal implications. Individuals
  are living longer and baby-boomers are nearing
  old age. Additionally, as individuals age their
  chance of developing Alzheimers disease (AD)
  increases. Some estimates suggest that by age 80
  an individual has a 50 chance of developing AD.
  There is a strong impetus for researchers to find
  an inexpensive and reliable method for detecting
  early cognitive decline. Several attempts have
  been made to define early symptoms of cognitive
  decline in older adults including,
  Age-Associated Memory Impairment (AAMI),
  Age-Associated Cognitive Decline (AACD), and Mild
  Cognitive Impairment (MCI). Previously our lab
  has reported that from a sample of approximately
  800 individuals over the age of 55, 30 met the
  criteria for at least one category of cognitive
  decline.
•  
• Over the last year, our lab has been working to
  develop a computerized memory screen to detect
  cognitive decline. We began this process by
  searching the literature to find measures that
  show sensitivity in detecting early signs of
  cognitive decline. This search indicated that
  delayed verbal recall and the Trails Making Test
  are sensitive measures of early decline. In
  addition to these measures, we also included
  measures sensitive to executive function decline,
  depression, anxiety, and alcoholism, four
  conditions common in old age and known to be
  associated with decreased cognitive ability.
•  
• After determining which measures to use, we set
  about the task of developing a touch screen,
  computerized program that would be short enough
  to administer in a doctors office, yet sensitive
  enough to capture early cognitive decline. To
  best balance between efficiency and sensitivity,
  we developed 4 different versions of the
  computerized memory screen. 
• MEASURES
• Memory Screening
• Rey Auditory Verbal Learning Task (RAVLT
  Recognition Version) In 2 versions of the screen
  there is one immediate trial and in the other 2
  versions there are five immediate trials. All
  versions contain a single 5-10 min delay trial.

MEMORY SCREEN RESULTS In order to streamline the
scoring and reporting of test results, we
programmed the computer to complete these
functions based on the individuals age and level
of education. In addition, the program is capable
of interpreting the results to determine whether
the individual meets the criteria for AAMI, AACD,
and/or MCI.  
Trails Making This task is administered in all 4
versions of the screen. In Trails A the
individual connects numbers in order and in
Trails B he/she alternates between connecting
numbers and letters. This task is sensitive to
deficits in attention, shifting ability, and
inhibition. Tower of London (TOL) This
executive function measure of planning ability is
included in 2 versions of the screen- 1 that
contains one immediate RAVLT trial and 1 that has
five immediate RAVLT trials.

• Convergent Validity
• Wisconsin Card Sorting Task Commonly used
  measure of executive functions requiring
  individuals to sort cards according to unknown
  categories. It is primarily sensitive to deficits
  in shifting ability and inhibition.
• RAVLT (Recall Version) The standard version of
  the RAVLT includes five immediate free recall
  trials, followed by a 20-min delayed free recall.
  Previous studies have indicated this test is
  highly sensitive in detecting age-related
  cognitive decline
• Divergent Validity
• Wechsler Abbreviated Scale of Intelligence (WASI)
  Vocabulary We expect a nonsignificant
  correlation between performance on the screen and
  vocabulary.
• Sensitivity
• Mini Mental Status Examination (MMSE) This test
  is the most frequently used test for the initial
  detection of Alzheimers disease. We expect to
  show that our computerized screen will better
  predict performance on our test being used to
  establish convergent validity.

FUTURE DIRECTIONS We are currently in the process
of validating the 4 memory screens. To do so, we
are recruiting individuals who have already
completed the University of Colorado Aging Clinic
(CUAC) Memory Clinic screen and asking them to
take 1 of the 4 computerized screens. When we
have collected an adequate sample of data we will
compare each screens sensitivity to the Mini
Mental Status Examination, a test frequently used
in primary care facilities. Convergent validity
of each measure will be assessed by comparing it
with delayed recall and the Wisconsin Card
Sorting Test. Divergent validity will be assessed
by comparing performance on a test of vocabulary
ability.   After a screen has been selected and
validation has been completed, the screen will be
available for use in doctors offices. This will
provide an efficient and reliable means for
detecting early signs of cognitive decline.
Detecting early cognitive decline is essential if
we wish to intervene when individuals are most
capable of effectively utilizing pharmacological
intervention or assistive technology.

• Depression, Anxiety, Drug/Alcohol screens These
  screens are included in all versions of the
  memory screen to detect emotional/abuse problems
  known to effect cognitive functioning.