Methodological difficulties when adhering to the emerging consensus criteria for assessment of cogni

1
Methodological difficulties when adhering to the
emerging consensus criteria for assessment of
cognitive and behavioural components of
Amyotrophic Lateral Sclerosis (ALS) in
practice. Norah Jordan1, Julie Phukan2,3, Orla
Hardiman2,4 and Niall Pender1. 1Department of
Psychology, Beaumont Hospital, Dublin 9, Ireland,
2 Department of Neurology, Beaumont Hospital,
Dublin 9, Ireland, 3 Trinity College Institute of
Neuroscience, Dublin 2, Ireland, 4 Trinity
College Dublin, Dublin 2, Ireland,

• INTRODUCTION
• We are currently conducting a longitudinal study
  in patients with ALS in order to characterise the
  cognitive and behavioural features of the disease
  in the Irish Population.
• In doing so we are characterising the phenotypes
  of cognitive impairments in this patient group
  but have encountered significant methodological
  limitations of the current consensus criteria and
  recommended assessment processes.
• AIMS
• To discuss the methodological difficulties
  encountered during the process of the study when
  attempting to apply the current ALS-FTD, ALSci
  and ALSbi consensus criteria in practice.
• METHODS
• We reviewed the issues and difficulties under the
  headings noted in Table 1.
• 1. Psychometrics
• a) Use of self-report measures
• The use of self-report measures increases the
  risk of a number of errors which might result in
  an over-estimate or under-estimate of
  cognitive-behavioural dysfunction.

• c) Issues with solely using tests of executive
  functioning to categorise ALSci
• Cognitive impairment criteria may be too narrow
  and theoretically invalid the frontal-striatal
  system is implicated in a broader range of
  cognitive/behavioural difficulties.
• 3. Utility of Tests
• There are differences in ALS bulbar onset
  patients alternate measures may be necessary.
• Patients with bulbar onset may have a different
  profile but cannot complete many of the tests we
  used for executive function, therefore
  mis-representing a proportion of patients with
  bulbar onset, e.g.
• Stroop
• Verbal fluency
• Category fluency
• Digit span and digit span backwards
• Development of computer assessments for such
  cases may be prudent.
• SUMMARY/CONCLUSIONS
• We would like to generate an awareness of
  methodological issues with cognitive and
  behavioural assessment in ALS patients.

Table 1 Areas of concern in the current ALS FTD
assessment methodology

• A further issue emerges when one uses the FrSBe
  patient-rated form. Such ratings might be
  unreliable due to the presence of cognitive
  deficit or reduced insight in the patient group.
• b) Statistical properties of measures
• Detailed psychometric data are scarce for the
  FrSBe in ALS. Conclusions about behavioural
  change and cognitive deterioration are difficult
• and limited given the lack of extensive data on
  the form.
• 2. Criteria (Strong et al., 2008)
• a) Global indices
• Development of global index measures (GIM) may
  not be the most appropriate use of
  neuro-psychological data.
• When trying to categorise patients by
  neuropsychological assessment, it is possible
  that we should be looking at deficits on
  individual neuropsychological test rather than
  trying to create global index measures.
• GIMs may be unreliable and mask important
  individual patterns and dissociation. As far back
  as 1988, Lezak criticised the use of Global
  Indices for investigating IQ (Lezak, 1988). The
  same principal also applies for assessing
  deficits in ALS. Attending to individual
  assessments would give a more detailed picture of
  the deficits that may appear with a dementing
  process in ALS patients.
• b) Behavioural descriptors
• The FrSBe may be more sensitive to frontocortical
  areas and may miss behaviours associated with FTL
  degeneration like semantic dementia/ compulsively
  stereotypic behaviours