The reliability and validity of the informant AD8 by comparison with a series of cognitive assessment tools in primary healthcare.

BACKGROUND: The validity and reliability of the informant AD8 in primary healthcare has not been established. Therefore, the present study examined the validity and reliability of the informant AD8 in government subsidized primary healthcare centers in Singapore.

METHODS: Eligible patients (≥60 years old) were recruited from primary healthcare centers and their informants received the AD8. Patient-informant dyads who agreed for further cognitive assessments received the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Clinical Dementia Rating (CDR), and a locally validated formal neuropsychological battery at a research center in a tertiary hospital.

RESULTS: 1,082 informants completed AD8 assessment at two primary healthcare centers. Of these, 309 patients-informant dyads were further assessed, of whom 243 (78.6%) were CDR = 0; 22 (7.1%) were CDR = 0.5; and 44 (14.2%) were CDR≥1. The mean administration time of the informant AD8 was 2.3 ± 1.0 minutes. The informant AD8 demonstrated good internal consistency (Cronbach's α = 0.85); inter-rater reliability (Intraclass Correlation Coefficient (ICC) = 0.85); and test-retest reliability (weighted κ = 0.80). Concurrent validity, as measured by the correlation between total AD8 scores and CDR global (R = 0.65, p < 0.001), CDR sum of boxes (R = 0.60, p < 0.001), MMSE (R = -0.39, p < 0.001), MoCA (R = -0.41, p < 0.001), as well as the formal neuropsychological battery (R = -0.46, p < 0.001), was good and consistent with previous studies. Construct validity, as measured by convergent validity (R ≥ 0.4) between individual items of AD8 with CDR and neuropsychological domains was acceptable.

CONCLUSIONS: The informant AD8 demonstrated good concurrent and construct validity and is a reliable measure to detect cognitive dysfunction in primary healthcare.