The Scales of Cognitive and Communicative Ability for Neurorehabilitation (SCCAN): Development and basic psychometric properties

Abstract

The purpose of this research was to develop a new test, the Scales of Cognitive and Communicative Ability for Neurorehabilitation (SCCAN). The SCCAN is intended to be used in the hospital setting to make appropriate treatment and discharge recommendations, and to guide more in depth assessment in other contexts. The test was designed to: (1) provide an overview of cognitive and communicative abilities in adult patients diagnosed with a variety of neurological disorders; (2) provide a measure of both impairment and functional ability; and (3) be administered in approximately 30 minutes. 20 neurologically healthy young controls and 20 neurologically healthy older controls were administered the SCCAN to ensure that these groups could complete the test successfully. The SCCAN and six standardized measures of cognition and communication were administered to 51 patients diagnosed with either left hemisphere stroke, right hemisphere pathology, or probable Alzheimer's disease. The SCCAN was re-administered to the patient group after a delay of 7 days. Data was analyzed using methods from both classical test theory and item response theory (IRT). The SCCAN demonstrated sufficient reliability across test administrations (r = .96), items (r = .99), and raters (percent agreement = .97). Findings also support the conclusion that this test provides an overview of cognition and communication that is appropriate for a broad range of patients. First, the test adequately distinguished the performance of separate subject groups (sensitivity = 98%, specificity = 95%). In addition, test scores correlated significantly with external measures of the same constructs (r = .41 to .91). Furthermore, a wide range of difficulty was observed among test items (item p-values ranged from 0 to 1). The fact that the SCCAN correlated significantly with a measure of impairment (r = .90) and functional ability (r = .94) suggests that this test provides a valid measure of both impairment and function. Lastly, results support the claim that the test can be administered in approximately 30 minutes. This test differs from similar instruments in terms of its comprehensiveness, providing a measure of both impairment and function, and incorporating a psychometrically validated decision tree to reduce test administration time.