Abstract

Objective: Valid definitions of dementia should discriminate dementia from other forms of cognitive impairment such as intellectual disability (ID). We aimed to evaluate the usefulness of criteria for dementia and mild cognitive impairment (MCI) in ID, including predictive validity, and inter-rater reliability. Method: We assessed 222 participants in a survey of older adults with ID without Down syndrome at two time points for dementia (T1 and T2). Mean follow-up period was 2.9 years. Dementia diagnoses were made according to International Classification of Diseases, Tenth Revision, Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition), Diagnostic criteria for psychiatric disorders for use with adults with learning disabilities (DC-LD) criteria. At follow-up (T2), raters were blind to initial diagnosis. Predictive validity was determined by comparing odds ratios (ORs) of death, or of having a "poor outcome" (i.e., either dying or being diagnosed with dementia at T2). Results: All dementia criteria showed substantial inter-rater reliability (k > 0.68) and high specificity (similar to 95%). Dementia cases at T1 were more likely to have died at T2 than those with no dementia (33.3% versus 14.9%; OR: 2.85; 95% confidence interval (95% CI): 1.12-7.22) and to have a "poor outcome" (77.8% versus 27.6%; OR: 9.18; 95% CI: 3.43-24.53). At least two dementia cases at T1 were false positives. Those with "MCI" at T1 were similar to "no dementia" cases in terms of poor outcomes at T2. Conclusion: Dementia diagnostic criteria show substantial reliability and satisfactory validity in ID. The diagnoses were, however, less stable than in the general population and some caution is advisable in those with more severe ID or additional sensory disability. MCI definitions require further consideration in the ID population.