Cognitive Performance Test (CPT)

Overview and Research

Minneapolis VA Geriatric Research, Education and Clinical Center (GRECC)

The Cognitive Performance Test (CPT)

The Cognitive Performance Test (CPT) is a standardized, performance-based assessment instrument, originally designed for the objective evaluation of function in Alzheimer’s Disease. This instrument, based on Allen Cognitive Disability Theory, uses six common activities of daily living (ADL) tasks, for which the information-processing requirements can be systematically varied to assess ordinal levels of functional capacity. Six tasks, titled DRESS, SHOP, TOAST, PHONE, WASH and TRAVEL comprise the test. For each task, standard equipment, set-up and methods of administration are required. A gross level score is determined for each of the six tasks; these scores are then added for a total score and averaged (divided by 6) to determine the functional level and mode. The CPT was initially developed as a research instrument, to be used in longitudinal study of functional change and for serial assessment to detect change in response to a pharmacologic or environmental intervention. It currently serves as the functional assessment for the GRECC diagnostic work-up and has proved to be useful in the assessment of patients, with a variety of diagnoses, to predict and explain capabilities to function in various contexts.

Allen Theory

Allen Cognitive Disability Theory is an Occupational Therapy frame of reference that addresses the functional consequences of cognitive impairment. In this theory, function is organized into 6 ordinal levels of global function ranging from normal (level 6) to profoundly disabled (level 1). Modes of performance within each level further qualify behavior variations and give a more precise measure of the person’s capacity. Allen theory emphasizes how information-processing deficits interfere with or prevent safe and effective occupational performance. The person’s level is determined by evaluating sensori-motor behavior, including the types of sensory information that can be processed and the resulting motor or task behavior. Sensory cues are ordered from internal cues (proprioception), to external concrete cues (tactile, visual, verbal), to increasingly abstract cues (related visual cues, verbal hypotheticals, symbols and ideas). Motor performance is also ordinal, beginning with reflexive actions that appear in response to internal cues, to planned actions that reflect processing of tactile, visual and then abstract cues. At each higher cognitive level, the sensory cues used in performance are more complex resulting in behavior that is more organized and complex.

Validity and Reliability Studies

Studies of the CPT were initiated in 1991 at the GRECC as part of a NIH longitudinal study of Alzheimer’s Disease (AD). Seventy-seven patients (56 male, 21 female) with mild to moderate AD and 15 neurologically normal elderly controls (8 male, 7 female) were administered the CPT. The average age of patients was 67.8 years, and of controls, 65.2 years. Subsets of the AD patients were assessed again at 4 weeks and at 1, 2, and 3 years following the initial evaluation. Internal consistency of the CPT estimated by alpha was .84. Intraclass correlation for interrater reliability was .91 and for test-retest reliability at 4 weeks, .89. CPT scores were significantly correlated with Mini-Mental State Examination scores (r=.67) and two measures of caregiver-rated ADL (Instrumental Activities of Daily Living, r=.64; Physical Self-Maintenance Scale, r=.49). Longitudinal testing (N=64) demonstrated significant decline in mean CPT scores with disease progression, and in contrast to the MMSE, initial CPT scores predicted the risk of institutionalization over a four-year follow-up period.

Bares (1998) retrospective study of AD patients who were evaluated in the GRECC found significant relationships between performance on the CPT and on neuropsychological measures. The sample included 100 mostly male, Caucasian patients aged 59 or older with mild to moderate stage disease. The average age was 74.9 years. In a hierarchical regression analysis of neuropsychological variables predicting function as measured by the CPT, significant predictors were neuropsychological measures that involved psychomotor skill with a planning, sequencing, and attentional component, while measures of memory, language, background variables and comorbidity were not predictive of function. Neuropsychological predictors of performance on the CPT were characterized under the rubric of executive function.

Other analyses showed the CPT tasks were highly related to each other. Pearson correlations among the six tasks ranged from .20 (travel and dress) to .51 (travel and phone). Burns et al. found slightly larger correlations among the six tasks ranging from .36 (travel and dress) to .68 (wash and toast). The CPT was found to have high internal consistency reliability (a = .76), which is comparable to although somewhat lower than the previous finding of .84. Factor analysis of the six CPT subtests supported the Burns et al. conclusion that the tasks are nonspecific, and reflect a single construct characterized as global functional status rather than discrete functional living skills. Findings support the conclusion that the CPT total score, not the individual tasks should be used.

Jennings-Pikey (2001) conducted similar studies of the relationship between neuropsychometrics and the CPT. One hundred eleven inpatient records from an inpatient psychiatric hospital in the Upper Midwest were accessed for study through archival records. The sample was mostly Caucasian, 81% female (N=91) and 19% male (N=20) 58 years of age and older. The average age was 79.6 years. Sixty-four and a half percent (N=71) of the subjects received a primary diagnosis of dementia or memory loss. There was a significant difference on the CPT between the means of the group that was given a memory loss diagnosis and those who did not receive this diagnosis, demonstrating that people with a memory loss diagnosis functioned at a lower level on the CPT. The validity of the CPT was supported to the extent that the test showed significant correlations with measures known to be sensitive to cognitive functioning in older adults. Convergent validity was demonstrated with significant correlation to the Global Assessment of Functioning Scale and the internal structure of the CPT to measure competency to carry out independent living skills was upheld.

CPT Administration

Administration of the CPT is based on occupational therapy principles of task analysis and adaptation. The test involves the sequential elimination or inclusion of sensory cues as difficulty with performance is observed. For example, in administering the PHONE task, if difficulty is observed with locating a phone number (use of symbolic cues), the phone book is removed and the number is given. If difficulty initiating dialing follows (use of visual cues), the number is removed and non-specific dialing is demonstrated (inclusion of manual cues) for imitation. The therapist must be able to determine whether or not the difficulties are due to information processing deficits or other factors such as low education, cultural bias, or physical impairment. The focus of the assessment is on the degree to which particular deficits in information processing compromise common functional activities. The specific tasks that comprise the CPT, while having face validity, are less important than the manner in which patients respond to the demands of varying complexity.

References

Burns T, Mortimer J. A. and Merchak P. The Cognitive Performance Test: A new approach to functional assessment in Alzheimer's disease. The Journal of Geriatric Psychiatry and Neurology. Vol. 7, 1994.

Bares K. Neuropsychological predictors of functional level in Alzheimer’s Disease. Unpublished doctoral dissertation. University of Minnesota, 1998.

Jennings-Pikey M. A validation study of the Cognitive Performance Test. Unpublished doctoral dissertation. Wheaton College, IL, 2001.