Birmingham Cognitive Screen: Brain Behavior Analysis

DESCRIPTION

The Birmingham Cognitive Screen: Brain Behavior Analysis (BCoS) is a cognitive screening instrument designed to provide a cognitive profile indicating whether an examinee has a clinical impairment relative to norms. The BCoS is comprised of 38 scores in five cognitive domains: Attention and Executive Function, Language, Memory, Number Skills, and Praxis and Action. The purpose of the BCoS is to provide a broad analysis of cognition measuring several attentional and executive functions (controlled, selective and sustained attention, working memory, rule following and concept switching). The BCoS was designed to be inclusive for individuals with common problems after brain lesions. The BCoS can be administered in a single session requiring approximately 1 hour; however, it is possible to break administration into two sessions. The BCoS was developed for stroke survivors and individuals with brain injury or dementia.

DEVELOPMENT

Test domains were reportedly selected to represent those where significant problems have a direct impact on activities of daily living. The construction of BCoS tests was guided by four key philosophies:

• Sensitive: Item difficulty reflects the ability to pick up a problem if one exists.
• Informative: Tests are designed to give direct information about the nature of the problem and to rule out contributions from irrelevant factors.
• Inclusive: Tests are designed to be as inclusive as possible for individuals with common problems after brain lesions.
• Time Efficient: Several measures can be derived from a single test when possible.

The BCoS test developers indicate they created tasks to assess cognitive domains they believe are most related to everyday life; however, they do not provide any review of the literature to support their selection of such tasks. The test authors state that the tests “were chosen on the basis of current models of information processing”; however, the manual lacks references to any clearly articulated theoretical models.

TECHNICAL DETAILS

The standardization sample for the BCoS included 100 male and female participants from the United Kingdom ranging in age from 50 to 75 or above divided into three age groups: 50 to 64, 65 to 74, and 75 or above. Age group 50 to 64 comprised 34 participants, and the remaining two age groups included 33 participants each. The test manual describes attempts to ensure normative sample representation to the U.K. population based on age, gender, and education. No information on ethnocultural background was reported. Cutoff scores based on the 5th percentile are reported as indicators of cognitive impairment.

INTERNAL CONSISTENCY RELIABILITY

No internal consistency reliability evidence was reported for the BCoS.

VALIDITY

Content evidence for the BCoS is reportedly based on tests being chosen according to current models of information processing; however, the manual is devoid of a clearly articulated theoretical model. Exploratory factor analysis was used to investigate construct validity. BCoS tests reportedly clustered into seven categories: spatial attention, long term memory, language, general orientation and comprehension, controlled attention/working memory, praxis/sequential processing, and other. Total variance explained is reported at 62%. The test manual reports evidence of convergent and divergent validity of the BCoS gathered from studies of the relationship between BCoS subtests and various established external measures of attention, language, memory, number skills, and praxis (e.g., Boston Diagnostic Aphasia Examination, Brixton Test of Executive Function, Wechsler Memory Scale-Revised). The correlation coefficient ranged from .41 to .94. Additional validation using clinical scans (voxel-based morphometry) was reportedly conducted, but no supporting evidence is provided in the test manual.

COMMENTARY

The BCoS is a cognitive screening instrument that is fairly quick and relatively easy to administer. The standardization sample of 100 is somewhat small, and the age group intervals are somewhat large. A larger sample and smaller interval (e.g., 5 years) would be more consistent with increased sensitivity to mild impairment. A major oversight is the absence of information regarding the ethnicity of the standardization sample. Content evidence for the BCoS is supported by recognizable tasks that make up the battery. The subtests are similar in style and format to other established assessment instruments (e.g., Wechsler Memory Scale, Fist-Edge-Palm Test, Rey Complex Figure). No theoretical model was articulated by the test authors for the test’s development, although test selection was reportedly based on current models of information processing; therefore the user must possess knowledge of current models of information processing, assessment, and identification of critical issues. Cutoff scores are used to determine cognitive impairment. Although cutoff scores are commonly used, this approach does not necessarily provide an empirical estimation of the relevant cutoff value for abnormal performance. The attention given by the BCoS test developers to minimizing irrelevant factors that can affect performance and to designing tasks that are neglect, aphasia, and hemiplegia friendly is laudable. In addition, the ability to assess multiple cognitive domains relatively briefly is a clear strength of the BCoS. However, the discussion of the BCoS test development process is unacceptably limited. The standardization sample is reasonably representative but very small. The temporal stability and interscorer reliability data are acceptable. The diagnostic agreement across all patients ranged from 62.5% to 100%, but there is no discussion of a standard by which to evaluate the percentage agreement data. Finally, no information regarding the type of brain injury or neurological damage the members of the reliability samples had sustained is provided in the test manual, and there is no information provided regarding where and how the test authors obtained their brain-injured samples. The sample sizes for the reliability studies are small. Although a reasonable amount of validity evidence is provided in the test manual, the authors do not specifically discuss how their evidence supports the inferences to be made from the BCoS scores. Further discussion of how the reliability and validity data collected support the utility of the BCoS is needed, and more evidence of the reliability and validity of BCoS test scores is needed with older samples. In addition, some discussion of the training required for proper administration and interpretation of the BCoS is critical.

SUMMARY

The BCoS appears to be conceptualized as a broad but brief screening tool for clinical assessment that contributes to the diagnostic process as opposed to being a diagnostic tool itself. The BCoS is easy to administer and score; however, its documented psychometric soundness is incomplete. Although its shortcomings preclude routine clinical use, the BCoS may be a useful instrument for trained neuropsychologists and others for generating hypotheses regarding cognitive functioning.

REVIEWER’S REFERENCES

Information about the reviewer’s references was not available.