Construct validity and reliability of the Dementia Test for People with Intellectual Disability: neuropsychological...

People with an intellectual disability are vulnerable to additional disorders such as dementia. Psychometrically sound and specific instruments are needed for assessment of cognitive functioning in cases of suspected dementia.

To evaluate the construct and item validity, internal consistency and test-retest reliability of a new neuropsychological test battery, the Dementia Test for People with Intellectual Disability (DTIM).

The DTIM was applied to 107 individuals with intellectual disability with (n = 16) and without (n = 91) dementia. The psychometric properties of the DTIM were assessed in a prospective study. The assessors were blinded to the diagnostic assignment.

Confirmatory factor analysis at the scale level showed that a one-factor model fitted the data well (root mean square error of approximation < 0.06, standardised root mean square residual < 0.08, comparative fit index > 0.9). At the domain level, one-factor models showed reasonable-to-good fit index for five of seven domains. Internal consistency indicated excellent reliability of the overall scale (Cronbach's α: 0.94 for dementia and 0.95 for controls). Item analysis revealed a wide range of difficulties (0.19-0.75 for dementia, 0.31-0.87 for controls), with minimal floor and ceiling effects. Eleven items (26%) had discrimination value ≤ 0.50. Test-retest reliability (n = 82) was high, with intraclass correlations of 0.95 (total score) and 0.69-0.96 (domains).

The DTIM fits a one-factor model and demonstrates internal and test-retest reliability; thus, it is suitable for use in cases of suspected dementia in people with various intellectual disabilities.