CREATING A CROSS-WALK BETWEEN TWO COMMONLY USED PROXY-BASED MEASURES OF COGNITIVE IMPAIRMENT: IQCODE AND AD8. (16th November 2018)

• Record Type: Journal Article
• Authors: Malloy, S; Zhou, W; Ngo, L; Gallagher, J; Husser, E; Inouye, S; Fick, D; Marcantonio, E
• Is Part Of: Innovation in aging. Volume 2(2018)Supplement 1
• Page Start: 897
• Page End: 897
• Publication Date: 2018-11-16
• Subjects: Aging -- Periodicals; Gerontology -- Periodicals; 612.67
• DOI: 10.1093/geroni/igy031.3343
• Languages: English
• ISSNs: 2399-5300

Abstract

Abstract: The Informant Questionnaire on Cognitive Decline in Elderly (IQCODE) and the Alzheimer's Disease 8 (AD8) are proxy-based assessments for mild cognitive impairment (MCI) and dementia that are particularly useful when direct cognitive testing of patients is not possible, such as during delirium. The IQCODE consists of 16 Likert questions, while the AD8 consists of 8 yes/no questions and is briefer to administer. We aimed to develop a crosswalk of scores between these measures, and to use established IQCODE cutoffs for MCI and for dementia to define similar cutoffs for AD8. We administered both the IQCODE and AD8 to proxies of enrolled hospitalized patients from two medical centers. We used equipercentile equating to develop the crosswalk of scores between the two instruments. The study sample consisted of 151 patients, with mean age 79 ± 11 years, 55% female. Of the proxies, 47% lived with the patient, and included 34% spouses, 50% children/children-in-law, and 16% other. We equated AD8 and IQCODE scores with good precision (standard error=0.04) throughout the range of cognitive decline (3.0–5.0 for IQCODE, 0–8 for AD8). Previously established cut-off scores of 3.2 for MCI and 3.5 for dementia on the IQCODE equated to scores of 2 for MCI and 4 for dementia on the AD8. Our results suggest that the previously published AD8 cut-off score of 2 for dementia might be overly sensitive, and a cut-off of 4 should be considered. Our results will enable researchers and clinicians to more directly compare studies using these two proxy-based cognitive measures.