Neuropsychological Differential Diagnosis
Impairment in cognitive functioning is a feature of many different psychiatric and neurological conditions. These impairments are often evaluated with neuropsychological tests, and the number of different tests available is increasing by the month. Many articles published in psychiatric journals such as this one examine profiles of cognitive impairment in different conditions and the correlates of those impairments, such as deficits in adaptive life functioning. It is not hard to find statistically significant differences between different psychiatric populations and healthy individuals, but there are substantial methodological problems associated with many of these research studies. In fact, it is often difficult to find tests on which patients with schizophrenia or dementia do not differ significantly from healthy subjects because of these patients’ overall pattern of impaired performance. Some research careers have been made simply by the exposition of methodological problems associated with the assessment of differential cognitive deficits in psychiatric populations.
The authors of Neuropsychological Differential Diagnosis take a different perspective, explicitly rejecting the concept that tests of statistical significance provide useful information on their own in terms of meaningful differences in classification between subjects with neuropsychiatric disorders and healthy individuals. Significant differences in cognitive test performance between two groups of subjects do not mean that there is much actual difference between the two groups in their overall distribution of performance scores.
An effect size of 0.6 standard deviations in the difference of two means, by convention a large effect and easy to detect in samples as small as 20 subjects per group, is associated with 62% overlap between two samples. In order to be able to tell with 90% certainty that an individual’s test score is consistent with a psychiatric or neurological diagnosis and not part of the lower end of the distribution of healthy scores, an average difference of about 2.5 standard deviations between healthy and impaired samples is required.
To make their case, the authors performed an enormously ambitious meta-analysis of the differential performance deficits of patients with several common and well-characterized neuropsychiatric conditions, including schizophrenia, major depressive disorder, Alzheimer’s disease, frontotemporal dementia, several forms of subcortical dementia, white matter diseases, obsessive-compulsive disorder, and mild traumatic brain injury. Their meta-analysis includes all of the research published on neuropsychological test differences between healthy comparison subjects and each of the target populations during the years 1980–1997. They review the performance of patients and healthy comparison subjects across a multitude of standard neuropsychological measures and present the effect sizes of the differences between the samples, as well as the variance across studies in the effect size of the differences and the resulting overlap between healthy comparison subjects and the patient populations. As a result, there is a wealth of detail on how much information each of these neuropsychological tests provides for test-based differential diagnosis of the target populations compared with healthy comparison subjects.
An intriguing result of this meta-analysis is that many of the tests often described as capturing fundamental characteristics of illnesses such as schizophrenia fare relatively poorly when evaluated with these standards. For instance, the Wisconsin Card Sorting Test, a multidimensional test of executive functioning, is associated with 40% overlap between the performance of patients and healthy comparison subjects. In schizophrenia, in fact, the top five discriminators, all associated with 20% or less overlap, are in the domains of verbal and visuospatial memory. In the domain of chronic multiple sclerosis, only one test is associated with less than 25% overlap between healthy individuals and multiple sclerosis patients, but many of the tests are associated with about 50% overlap between multiple sclerosis patients and healthy comparison subjects. These tests would provide essentially no data useful for differential diagnosis.
One factor not considered with this type of analysis is the importance of cognitive performance for within-illness variation. In an illness such as schizophrenia or multiple sclerosis, heterogeneity in course, symptoms, and outcome is much greater than in an illness such as Alzheimer’s disease. Certain cognitive impairments that do not help in differential diagnosis may provide substantial information regarding the likelihood that a specific individual may be able to live independently or sustain employment. Therefore, in an illness such as schizophrenia, where there is considerable variation in outcome across patients, a test that identified every patient with the illness would not predict the substantial variation across patients in functional skills.
This book is not about gloom, doom, and the uselessness of cognitive assessment. In the domain of Alzheimer’s disease there are 15 different tests, all of memory, that are associated with less than 5% overlap between healthy comparison subjects and patients with Alzheimer’s disease, and the state of affairs for progressive supranuclear palsy is almost as good, considering that many fewer studies have been conducted for this disorder. For each of these conditions, a detailed hierarchy of the level of relative impairment across tests of affected patients compared with that of healthy subjects is presented in detail. These data are extremely valuable for researchers or clinicians who want to identify a battery of assessment measures that are likely to be highly discriminating.
For instance, in the domain of Alzheimer’s disease, the variation in discriminative usefulness across different memory measures is substantial. The California Verbal Learning Test long delay free recall measure was 98% useful for discrimination of healthy individuals and Alzheimer’s disease patients, but the California Verbal Learning Test free recall intrusions measure was associated with 50% overlap between the samples. Clinical neuropsychological conventional wisdom characterizes free recall intrusions as a specific sign of cortical dementia, a belief that is completely contradicted by these data.
This book is a very useful reference for researchers studying cognitive impairments in different neuropsychiatric conditions. It will also be an excellent reference for trainees who are about to embark on a career that involves either administering or interpreting the results of neuropsychological tests. There is no other comparable reference, and it is an excellent supplement to previous books that provide detailed descriptions of neuropsychological tests and normative standards. The reference list is outstanding, and the authors provide suggestions for further reading at the end of the text.
