MEMORY ASSESSMENT IN MEMORY REHABILITATION

1
MEMORY ASSESSMENT IN MEMORY REHABILITATION Cardi
ff, September 17-19 2002 Narinder Kapur -
Southampton (n.kapur_at_soton.ac.uk) Veronica
Bradley - Haywards Heath, Sussex Jon Evans -
Cambridge
As in other branches of clinical or scientific
inquiry, accurate memory assessment depends on
asking the right question, in the right way.


2
Desert Island Test

• You are stranded on a desert island with a brain
  injury rehabilitation facility
• You are only allowed to take with you -
• Three questions you can ask a patient
• Three cognitive tests you can use
• Three treatment interventions you can employ
• Three outcome measures
• Which ones would you take?

3
OUTLINE OF TALK

• Overview of memory assessment in clinical
  settings
• Memory assessment procedures in clinical practice
• Future developments in memory assessment

4
OUTLINE OF TALK

• Overview of memory assessment in clinical
  settings - context, conceptual framework, sources
  of evidence
• Assessment procedures in clinical practice
• Future developments in memory assessment

5
MEMORY ASSESSMENT IN CONTEXT Memory assessment
needs to be considered in the context of other
features of the individual patient.

• Pre-morbid cognitive functioning, daily routine,
  etc.
• Other cognitive, emotional and behavioural
  functions - executive function, language,
  perception, attention, speed of processing, mood,
  temperament, insight, motivation, expectations,
  etc. Bermuda triangle of mood-concentration-memo
  ry.
• Medical condition - sensory/motor deficits,
  tiredness, somatic symptoms, medication, etc.

6
It is useful to have a theoretical/conceptual
framework of human memory when considering
selection of assessment procedures and
interpretation of memory performance. This can be
at several different levels -

• Anatomical
• Behavioural
• Cognitive

7
A. ANATOMICAL FRAMEWORK Hans Markowitsch
(1998). Neurocase.
Functional specialisation and integration.
Networks rule - OK
8
Distinctiveness
Sensory
Diversity

Cognitive
Significance
B. BEHAVIOURAL FRAMEWORK. Encoding Factors
Important for Long-Term Memory Consolidation
Exposure
Duration Spacing
Emotional

Significance
Predictability
Personal
Consequence
a
a
a
9
C. COGNITIVE FRAMEWORK Distinctive Human Memory
Systems

• Episodic (Experiential) Memory
• Semantic (Knowledge) Memory
• Working (Attentional / Short-term) Memory
• Procedural (Motor, Skills) Memory
• Perceptual Representational (Sensory) Memory
• Emotional Memory

10
A MODEL OF THEREHABILITATION PROCESS
Ideally, features of memory assessment should be
determined by components of such a rehabilitation
framework as they apply to an individual patient
(person-centred planning).
11
WHEN TO ASSESS MEMORY? Assessment should
ideally be as close as possible in time to onset
of programme of memory rehabilitation. While
assessments in the acute phase of a brain injury
may be useful to track early recovery of
function, such assessments will often be
compounded by extraneous variables, such as
general medical condition, side-effects of drugs,
etc., resulting in more noise/variability and
making it difficult to predict later levels of
recovery and the outcome of rehabilitation.
12
OUTLINE OF TALK

• Overview of memory assessment in clinical
  settings
• Memory assessment procedures in clinical practice
  - what are the key questions?
• Future developments in memory assessment

13
WHY ASSESS MEMORY? Most clinical investigations,
including assessment of memory functioning in
rehabilitation settings, are directly or
indirectly intended to answer one or more of the
following questions - 1. WHATS GOING ON? 2.
WHATS GOING TO HAPPEN? 3. WHAT WILL HELP? 4.
HAS MEMORY CHANGED?
Nature of memory impairment
Predict recovery, everday adjustment
Type of memory intervention.
Recovery. Effectiveness of memory intervention.
14
Three main sources of evidence for the
assessment of memory functioning in clinical
settings -

• Clinical interview - with patient, family and
  relevant health care professionals.
• 2. Behavioural Data - Questionnaires rating
  scales checklists memory behaviour recorded by
  patient, family or other health care staff in
  domestic, work and clinical settings. Such
  observations may be part of the treatment process
  itself.
• 3. Memory testing (flexible use of standard
  tests).

15
Consider questions in relation to these sets of
evidence

• Clinical interview
• Behavioural Data
• 3. Memory testing

16

• (i) Is there memory dysfunction?
• (ii) How severe is the memory impairment?
• (iii) Which memory domains are affected?
• (iv) Which memory systems are affected?
• (v) What is the aetiology of the memory
  dysfunction?
• (vi) What are the cognitive mechanisms?

17

• (i) Is there memory dysfunction?
• (ii) How severe is the memory impairment?
• (iii) Which memory domains are affected?
• (iv) Which memory systems are affected?
• (v) What is the aetiology of the memory
  dysfunction?
• (vi) What are the cognitive mechanisms?

18

• WHATS GOING ON?
• (i) Is there memory dysfunction?
• Clinical Interview/Behavioural Observations

• Ask the patient - however.
• Absence of insight - frontal dysfunction,
  amnesia.
• Excessive awareness - anxiety factors
• Generally low correlations between objective and
  subjective measures of memory (e.g. Kopelman et
  al., 1998, Cortex).
• Several memory questionnaires have been published
  (noted in Lezak book) - do not have the
  flexibility of the clinical interview in
  detecting the presence of memory dysfunction.

19
MEMORY INTERVIEW

• If acute brain insult, gather information on
  memory for events prior to and after the insult,
  as this may give an index of the severity of the
  brain illness/injury (in head injury, duration of
  post-traumatic amnesia).
• Cover range of everyday memory difficulties,
  including ones that are relevant to life-style of
  individual patient -
• Episodic / Experiential memory - messages,
  remembering to do things, retain reading
  material, follow film/soap on TV.
• Semantic / Knowledge memory - finding words in
  general conversation, navigating to familiar
  places.
• Informal, indirect assessment of memory - recent
  news events (e.g. whether recently deceased
  personalities are dead or alive) - check for
  media exposure, memory for features of clinical
  history (e.g. medication, dates of
  investigations, names of staff).

20
(No Transcript)
21

• WHATS GOING ON?
• (i) Is there memory dysfunction?
• Memory testing

Tests which are most sensitive to the general
presence of memory impairment include those with
a free recall component, a delayed retention
testing component, or a paired-associate learning
component - e.g. delayed recall of story or
complex figure from Adult Memory and Information
Processing Battery, California Verbal Learning
Test, and WMS-III verbal paired-associate
learning test. The pattern-location memory
subtest of the CANTAB (Cambridge computerised
cognitive testing battery) has been claimed to be
particularly sensitive to some neurological
conditions in the elderly.
22

• (i) Is there memory dysfunction?
• (ii) How severe is the memory impairment?
• (iii) Which memory domains are affected?
• (iv) Which memory systems are affected?
• (v) What is the aetiology of the memory
  dysfunction?
• (vi) What are the cognitive mechanisms?

23

• WHATS GOING ON?
• (ii) How severe is the memory impairment?
• Clinical Interview/Behavioural Observations

Some of the memory questionnaires reviewed by
Lezak (1995) give an overall score that reflects
subjective severity of memory impairment.
Disorientation for time (day of week, month,
year) usually associated with more marked memory
impairment. Estimates of severity of memory
symptoms and memory test deficits need to be seen
in the context of premorbid memory functioning.
One patients mild memory impairment may be
another patients marked memory
impairment. Confabulation may occur
spontaneously, but it can be elicited and
documented by a structured interview procedure,
along the lines of the one described by Mercer et
al. (Archives of Neurology, 1977), or by asking
for recall of episodes in response to cue words
(Moscovitch Melo, Neuropsychologia, 1997).
24

• WHATS GOING ON?
• (ii) How severe is the memory impairment?
• Memory testing

Memory Quotients may in some cases provide a
useful index of the severity of memory loss, and
may be useful in grading those patients with a
global amnesic syndrome. However, in some cases
they may be misleading since they may mask
variability between subtests. Those memory test
batteries that provide an overall severity score
include the Wechsler Memory Scale-Revised, the
Wechsler Memory Scale III, and the Rivermead
Behavioural Memory Test.
25

• (i) Is there memory dysfunction?
• (ii) How severe is the memory impairment?
• (iii) Which memory domains are affected?
• (iv) Which memory systems are affected?
• (v) What is the aetiology of the memory
  dysfunction?
• (vi) What are the cognitive mechanisms?

26

• WHATS GOING ON?
• (iii) Which memory domains are affected?
• Clinical Interview/Behavioural Observations

Information relating to material-specific
symptoms may be obtained by questioning and
observations. Verbal memory symptoms may be
evident in memory lapses for phone messages, the
content of a conversation, what has been read,
etc. Nonverbal memory symptoms can be elicited
by asking about familiarity recognition memory
for faces.
27

• WHATS GOING ON?
• (iii) Which memory domains are affected?
• Memory Testing

Verbal memory - memory for word-lists,
word-pairs, short story, etc. Immediate story and
word-list recall - significant strategic
component that may be affected by executive
dysfunction all three tests may provoke some
degree of anxiety. Nonverbal, visuoperceptual
memory - faces subtests of the Recognition Memory
Test and the Wechsler Memory Scale III. Design
memory subtests of batteries such as the Adult
Memory and Information Processing Battery. Visual
Patterns Recall Test - recall checkerboard
pattern. Spatial memory - Location Learning Test
(Bucks et al., 2000). Pictorial memory - The
Contextual Memory Test (Toglia, 1993) - thematic
picture memory test with post-testing
questionnaire to tap awareness about how well the
patient thinks he/she has done on the
test. Face-name memory - components of face-name
memory in subtests from Rivermead Behavioural
Memory Test-Extended, Doors People Test, and
Memory Assessment Scales (Williams, 1991).
28

• (i) Is there memory dysfunction?
• (ii) How severe is the memory impairment?
• (iii) Which memory domains are affected?
• (iv) Which memory systems are affected?
• (v) What is the aetiology of the memory
  dysfunction?
• (vi) What are the cognitive mechanisms?

29

• WHATS GOING ON?
• (iv) Which memory systems are affected?
• Clinical Interview/Behavioural Observations

Long-term episodic memory may be tapped by
asking about memory for specific events, such as
holidays or personally experienced events
relating to interests/hobbies. May be selectively
impaired in some forms of temporal lobe
epilepsy. Working memory difficulties can be seen
in settings such as keeping track of a
conversation, especially where several people are
involved or where there is background noise,
doing two things at once e.g. answer phone while
watching TV. Semantic memory impairment may be
evident in word-finding symptoms.
Word-substitutions are more significant than
general tip-of-the-tongue symptoms.
30

• WHATS GOING ON?
• (iv) Which memory systems are affected?
• Memory Testing

• Long-term episodic memory - Autobiographical
  Memory Interview. The Crovitz procedure, where
  autobiographical incident memories are retrieved
  in response to a specific cue word.
• Semantic memory impairment - Graded Naming Test,
  the Boston Naming Test and the Pyramids and Palm
  Trees test.
• Verbal working/short-term memory - forward and
  backward digit span, and the Letter-Number
  Sequencing subtest of the WMS III.
• Nonverbal working/short-term memory - The block
  tapping subtest of the WMS-III can be seen to
  assess some aspects of nonverbal working memory.
• Loss of motor knowledge - tests of apraxia, such
  as that included in the Toronto-Kaplan
  Neuropsychological Battery.
• Implicit memory may be tapped by using tests that
  involve repeated identification of fragmented
  pictures over successive trials (Snodgrass
  pictures - available from Life Science
  Associates, USA).

31

• (i) Is there memory dysfunction?
• (ii) How severe is the memory impairment?
• (iii) Which memory domains are affected?
• (iv) Which memory systems are affected?
• (v) What is the aetiology of the memory
  dysfunction?
• (vi) What are the cognitive mechanisms?

32

• WHATS GOING ON?
• (v) What is the aetiology of the memory
  dysfunction - ?Psychogenic Basis
• Clinical Interview/Behavioural Observations

• Psychogenic variables may range from anxiety to
  depression to hysteria to malingering. Do not
  rely on single feature. Note clinical context of
  feature.
• Secondary gain/compensation factors, ?psychiatric
  history, ?alcohol/drug abuse
• Loss of personal identity, forgets how to sign
  name, inability to recognize family members.
• Dense, life-long autobiographical amnesia in the
  absence of correspondingly severe anterograde
  amnesia.
• Memory symptoms related to absentmindedness.
• Concentration perceived to be more impaired than
  memory.
• Significant variability in memory difficulties.
• Patient complains of memory difficulties yet
  enjoys reading books or can readily follow a film.

33

• WHATS GOING ON?
• (v) What is the aetiology of the memory
  dysfunction?
• Memory Testing

Anxiety - Recognition memory tests invoke less
anxiety than recall tests, and recognition memory
impairment may therefore be somewhat more
reliable. Malingering - Chance level or poor
scores on easy memory tests. Some malingering
tests reviewed by Lezak (1995) and by Spreen
Strauss (1998). Recent standardized tests include
- Test of Neuropsychological Malingering
Victoria Symptom Validity Tests - recognition
memory for digit sequences. Test of Memory
Malingering - picture recognition memory. Other
tests - Picture Memory subtest of the Camden
Battery, and forward digit span. Coin-in-the-hand
test, fragmented picture learning, and the
modified Rey test.
34
MODIFIED REY TEST (Kapur, unpublished) Present
set of items for 10s, immediate recall. 12
numbers and letters to remember
1 2 3 3 2 1 A B C C B A
35

• (i) Is there memory dysfunction?
• (ii) How severe is the memory impairment?
• (iii) Which memory domains are affected?
• (iv) Which memory systems are affected?
• (v) What is the aetiology of the memory
  dysfunction?
• (vi) What are the cognitive mechanisms?

36

• WHATS GOING ON?
• (vi) What are the cognitive mechanisms?
• Clinical Interview/Behavioural Observations

Some memory symptoms may be more susceptible to
variability in attention, such as forgetting
where something has been put or going into a room
and forgetting what one went in for. Repeatedly
forgetting the names of family members, or giving
the month when asked the year, may reflect
retrieval difficulties that could be due to a
mild dysphasia. Observe behaviour during memory
testing debrief on strategies used, ask about
concentration during a test that was performed
badly, etc.
37

• WHATS GOING ON?
• (vi) What are the cognitive mechanisms?
• Memory Testing

Some word-list learning tests, such as the
Buschke Selective Reminding Test purport to
provide distinct measures of storage and
retrieval capacity. The California Verbal
Learning Test, which presents items in the form
of a shopping list to remember, can provide
several measures of potential theoretical value -
recall consistency across trails susceptibility
to proactive and retroactive interference
comparisons between free recall-cued recall-
recognition and data on intrusions,
perseverations and false-positive responses.
38

• WHATS GOING TO HAPPEN?
• (i) Likely level of recovery of memory at later
  time.
• (ii) Functioning in everyday settings -
  domestic, social, education, work, etc.

39

• WHATS GOING TO HAPPEN?
• (i) Likely level of recovery of memory at later
  time.
• (ii) Functioning in everyday settings -
  domestic, social, education, work, etc.

Never give a prognosis, and - if you have to -
make it damn vague Sir Thomas Lewis, cardiac
physician
40

• WHATS GOING TO HAPPEN?
• (i) Likely level of recovery of memory at later
  time.
• Clinical Interview/Behavioural Observations

The severity of post-traumatic amnesia, either
assessed retrospectively or concurrently, is
often used to help make judgments about long-term
prognosis. It is an imperfect measure, especially
if it is made retrospectively, and may in some
cases be compounded by post-injury events such as
sedation, surgery, etc. If patients are testable
at an early stage in recovery, this may be a
useful sign in itself of a promising prognosis.
41

• WHATS GOING TO HAPPEN?
• (i) Likely level of recovery of memory at later
  time.
• Memory Testing

Two concurrent measures of post-traumatic
amnesia (PTA) include the Galveston Orientation
and Amnesia Test, and the Westmead Post Traumatic
Amnesia Scale. Some items from the Wessex Head
Injury Matrix, which mainly deals with
behavioural observations relating to emergence
from coma, may be used to assess orientation and
memory in the period of time when PTA is
beginning to resolve.
42

• WHATS GOING TO HAPPEN?
• (i) Likely level of recovery of memory at later
  time.
• (ii) Functioning in everyday settings -
  domestic, social, education, work, etc.

43

• WHATS GOING TO HAPPEN?
• (ii) Functioning in everyday settings -
  domestic,
• social, education, work, etc.
• Clinical Interview/Behavioural Observations

The prediction of memory functioning in everyday
settings may not be readily possible from
clinical interview alone. Observations of
performance in occupational therapy settings that
mimic domestic, social or work settings may be
useful in helping to predict adjustment in
everyday situations outside the clinical
environment.
44

• WHATS GOING TO HAPPEN?
• (ii) Functioning in everyday settings -
  domestic,
• social, education, work, etc.
• Memory Testing

Memory tests that mimic retention scenarios in
the real world include the Rivermead
Behavioural Memory Test. In its earlier edition,
it had four forms, and in the more sensitive
Extended edition it has two forms. Tests in this
battery, such as those which assess prospective
memory, may be particularly useful in helping to
predict adjustment in everyday settings. The
flexible use of other tests is worth keeping in
mind according to the particular everyday memory
demands - e.g. the California Verbal Learning
Test uses a shopping list analogy to present
word-lists.
45

• WHAT WILL HELP?
• Type of intervention - memory aids, cognitive
  strategies, environmental support, advice to
  carer.

46

• WHAT WILL HELP?
• Type of intervention - Memory aids, cognitive
  strategies, environmental support, advice to
  carer.
• Clinical Interview/Behavioural Observations

Clinical interview with the patient is important
to establish - Detailed listing of memory
symptoms, using memory questionnaires where
appropriate, with an indication of which ones are
a priority and how they relate to patients
overall goals and feelings of well-being Obtain
similar information from family/care
staff Pre-morbid use of memory strategies and
memory aids Likely demands on memory - e.g. in
work or educational settings Insight into memory
loss Motivation to improve memory Expectations
of memory intervention Family support - e.g. in
use of memory aids Support from work
colleagues Structure of home and work
environment.
47
Use this page to draw rough sketches of the
layout of the rooms in your house/flat that you
use frequently Use this page to draw rough
sketches of the layout of your workplace.
48

• WHAT WILL HELP?
• Type of intervention - Memory aids, cognitive
  strategies, environmental support, advice to
  carer.
• Memory Testing

If patients have mild to moderate memory
impairment, in the context of good executive
function and minimal additional cognitive
deficits, this is a favourable sign for many
forms of memory intervention, especially those
which involve concentration and new learning on
the part of the patient, and the use of more
complex memory aids. Global indices of memory
impairment that fall in the amnesic range, and
where there is significant executive dysfunction,
are more compatible with interventions that
involve very simple memory aids, changes in
environment and in routine, and advice to carer
or care-staff. Presence of material-specific
deficits may inform certain memory rehabilitation
procedures - e.g. use of visual imagery or
spatial strategies where selective verbal memory
deficits are present.
49

• HAS MEMORY CHANGED?
• Recovery/progression of memory deficit
• evaluate effects of intervention.

50

• HAS MEMORY CHANGED?
• Recovery/progression of memory deficit
• evaluate effects of intervention.
• Clinical Interview/Behavioural Observations

Measures such as the Galveston Orientation and
Amnesia Test, and the Westmead Battery, may help
to monitor the early stages of recovery from
acute memory loss following conditions such as
severe head injury or subarachnoid haemorrhage.
In the more chronic phase, electronic monitoring
is possible for memory domains such as
remembering to take medication.
51
REMEMBERING TO TAKE MEDICATION - ASSESSING DRUG
COMPLIANCE
Electronic device in bottle cap registers when
pill box has been opened. Data can be output from
reader. Number on top of cap shows number of
times bottle has been opened that day. Aardex
Ltd, Switzerland.
52

• HAS MEMORY CHANGED?
• Recovery/progression of memory deficit
• evaluate effects of intervention.
• Clinical Interview/Behavioural Observations
• Measures such as the Galveston Orientation and
  Amnesia Test, and the Westmead Battery, may help
  to monitor recovery from acute memory loss
  following conditions such as severe head injury
  or subarachnoid haemorrhage.
• Electronic monitoring is possible for events such
  as remembering to open pill-bottles and take
  medication.

Subjective memory ratings are important to guage
how patients feel their memory in everyday
situations has improved, whether they are more
self-confident in settings where there is a
demand on their memory, and how well they now
cope with memory failures. Diary-based recording
of memory lapses, may be helpful in assessing the
effectiveness of treatment intervention.
53
Example of tailor-made behavioural
memory measures used in a memory rehabilitation
study that examined the effectiveness of an
electronic pager (Wilson et al., 2001, JNNP).

• Daily observations recorded by patient for
  following memory activities -
• Did you unlock your door this morning for the
  carer?
• Did you take purse, keys, diary and coat when you
  went out?
• Did you check your door was locked last night?
• Did you remember what time the carer started and
  finished today?
• Last night, did you get clean clothes out for the
  morning?

54

• HAS MEMORY CHANGED?
• Recovery/progression of memory deficit
• evaluate effects of intervention.
• Memory Testing

Commonly used, standardized tests with parallel
forms that may be useful in assessing change
include the original version of the Rivermead
Behavioural Memory Test (four forms), the Hopkins
Verbal Learning Test-Revised (six forms), the
Brief Visuospatial Memory Test-Revised (six
forms) and computerised batteries that are often
used in dementia drug trials (e.g. CANTAB,
batteries developed by Crook et al., Wesnes et
al.). Practice effects vary according to tests -
e.g. Wilson et al. (2001, JINS) found that digit
span, doors recognition memory and word list
recall showed little practice effects with repeat
testing, whereas verbal fluency did show practice
effects. Behavioural measures have minimal
practice effects, and can thus be given much more
frequently than cognitive tests.
55
Results of memory assessment should be
communicated in ways that are meaningful to
person who is going to use that information

• Use non-technical language.
• 2. Describe behaviours/abilities rather than
  tests and test scores.
• 3. Highlight strengths as well as weaknesses.

56
OUTLINE OF TALK

• Overview of memory assessment in clinical
  settings
• Assessment procedures in practice
• Future developments in memory assessment

57
FUTURE DEVELOPMENTS

• More ecologically valid memory tests - e.g. tests
  of prospective memory.
• Tests which take particular advantage of computer
  technology - e.g. virtual reality memory tests.
• Memory tests which incorporate advances in
  communication technology - e.g. internet-based or
  videophone based memory assessment.
• Memory tests specifically designed to monitor the
  effects of therapeutic intervention, with matched
  parallel forms and data on practice effects.
• Memory testing that is combined with functional
  brain imaging, and which may shed light on
  compensatory mechanisms underlying memory test
  performance, perhaps using upright MR scanners.

58
UPRIGHT MR SCANNER - Nakada et al (2001,
Neurology)
59
SUMMARY

• Accurate memory assessment is critical before
  embarking on memory rehabilitation.
• It is important to have some theoretical/conceptua
  l framework relating to memory and
  rehabilitation.
• Memory assessment should ideally be informed by a
  combination of clinical interview data,
  behavioural observations, and formal memory
  testing.
• Technological developments are likely to have an
  increasing impact on memory test procedures.