Objectives A learning disabilities’ dementia battery was developed to assess cognitive abilities in individuals referred to the learning disabilities service because of concerns of possible dementia. The present study aimed to establish concurrent validity with previously validated measures of cognitive ability and its clinical effectiveness in detecting dementia in this population.

Methods Fifty five individuals aged 29 and over (range: 29 to 71), received a baseline and a follow-up assessment using the dementia battery between 2000 and 2010. Differences in performance between individuals allocated to ‘probable’, ‘unsure’ and ‘no’ dementia groupings were investigated at domain and subtest level, as well as overall performance. Results on the battery were compared with clinically relevant measures of dementia also included in the local assessment protocol.

Results Significant differences in overall performance were found between the ‘probable’ and ‘no’ dementia groups as well as cognitive domain-specific differences. No differences were found at subtest level. Good concurrent validity was found between the battery and comparable measures of change within the dementia assessment protocol, namely the VABS, DMR and BPVS II.

Conclusions The learning disabilities’ dementia battery appears to be a good measure, which can be used longitudinally, to detect change in individuals and help establish a diagnosis of dementia. It is also comparable with other measures of change incorporated in the dementia assessment protocol. Subtests included in the language domain appear to be the most relevant at detecting significant changes between baseline and follow up. Future studies should attempt to standardise this measure and establish cut-off scores.