Classification accuracy of the English version of the Canadian Indigenous Cognitive Assessment (CICA) in a majority culture memory clinic sample
Journal Articles
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
AbstractBackgroundMost cognitive screening tools are culturally inappropriate for Indigenous peoples. The Kimberly Indigenous Cognitive Assessment (KICA) screening tool was collaboratively developed in Australia with, and for, Indigenous peoples. Using a similar community‐based method, the KICA was adapted to Canadian Indigenous peoples’ culture and translated into other languages, beginning in Ontario on Manitoulin Island, and was re‐named the Canadian Indigenous Cognitive Assessment (CICA). The classification accuracy of the CICA‐Ontario version, English translation, was explored in the current project in a memory clinic setting where the majority of patients were of Eurocentric origin.MethodOf 79 consecutive cases presenting to an interdisciplinary diagnostic memory clinic, 58 completed the CICA. Inclusion criteria included consent for participation in research and planned administration of the typical 2 hour long neuropsychological battery. CICA data were not used for any clinical process; consequently, diagnosis was independent. Diagnosis was made by consensus of the neurologist and neuropsychologist after patient and collateral informant interviews and nursing, neurological, medical (comprehensive blood work, neuroimaging), neuropsychological, and physical therapy evaluations.ResultPatients’ ages ranged between 38 and 90 (M= 65.3, SD= 12.3) and 52 percent were male. Diagnoses included no objective cognitive impairment despite cognitive concerns termed subjective cognitive impairment (SCI; n= 23; CICAM= 37.09; CICASD= 1.08), mild cognitive impairment (MCI; n= 14; CICAM= 34.93; CICASD= 2.46), dementia due to Alzheimer disease (AD; n= 9; CICAM= 31.44; CICASD= 3.00), and non‐AD dementias (n= 12; CICAM= 32.75; CICASD= 5.07). Classification accuracy as measured by the area under the receiver operating characteristic (ROC) curve (AUC) for groups with diagnoses of dementia (n= 21) vs no dementia (SCI or MCI; n= 37) was .78 (AUC confidence interval (CI) .66‐.92). Accuracy was high for cognitive impairment (dementia and MCI; n= 35) versus SCI (n= 23) (AUC=.84 CI .74‐.94) and lower for MCI (n= 14) vs dementia (n= 14)(AUC=.67; CI .49‐.85).ConclusionAlthough additional validity data from Indigenous peoples is needed, these data support the measurement properties of the CICA‐Ontario English‐version in a majority culture sample.
