Remember to turn off the stove! : prospective memory in dementia : a thesis presented in partial fulfilment of the requirements for the degree in Master of Arts in Psychology at Massey University, Wellington, New Zealand

Abstract

Dementia predominately affects older people, progressively affecting activities of daily living. Research shows prospective memory, the memory for future intentions, to be a sensitive indicator of dementia. Prospective memory is not routinely assessed in older people, yet testing prospective remembering could result in early diagnosis of dementia creating opportunity for early intervention. A retrospective analysis of the prospective memory subtest scores on the Rivermead Behavioural Memory Test (RBMT), a test of everyday memory, was completed for a group of older adults diagnosed with either vascular dementia (n=35) or Alzheimer’s disease (n=39) aged 60-89 years. These individuals participated in a study by Glass (1998) exploring the possibility of discrimination between vascular dementia and non-vascular dementia using the RBMT. Glass’s findings indicated that a combination of four subtests, two of which assessed prospective memory, were able to classify a case as vascular or non-vascular with an error rate of 2.7% out of the 74 cases analysed. The question that rose from that data was, what would be the predictive validity of the 3 prospective memory subtests if the individual scoring components were analysed separately? Glass’s data was reviewed and analysed using nonparametric and Chi Square statistical analysis. Analysis indicated that the components that assessed the prospective element of prospective memory were more predictive of vascular dementia (VAD) and dementia of the Alzheimer’s type (DAT) than the retrospective element. Individual subtest scoring components indicated significant differences between VaD and DAT on ‘delayed message’, ‘delayed location’ and ‘spontaneous appointment request’. Unexpectedly, there was also a significant gender difference within the DAT group favouring males on the scoring component ‘delayed message’. It was also significant that the PM performance of VaD females exceeded that of the DAT females on the subtests ‘delayed message’ and ‘appointment’. Differences in prospective memory performance between VaD and DAT were not substantial enough to support the use of PM performance as a discriminator between VaD and DAT in a clinical setting but may be useful as an additional marker in differential diagnosis.