Impact of ageing on discourse types: a comparative study.

The study compared different people at one point in time rather than following the same people as they aged, so it cannot separate the effect of ageing from differences between generations. The authors themselves say the results should be interpreted with caution because of the small sample size. Discourse was rated only qualitatively; quantitative measures such as length of utterances, type-token ratio, pauses and fluency were not used. Only a limited picture task was used, and the authors suggest sequential pictures with a clear logical order might let older adults perform better. There was no monolingual comparison group, so the possibility that bilingualism helps these tasks could not be tested. Gender differences were found but not explored further, which the authors list as a limitation. Groups were formed on age and cognitive scores only; education, occupation and socio-economic background were not taken into account in forming them. Participants were recruited by convenience sampling, were all educated to at least secondary level, and were screened to exclude cognitive impairment, so the findings may not apply to less educated or cognitively impaired older adults. The paper does not report whether the person scoring the discourse samples knew which age group each recording came from, or how consistent the scoring was between raters.

The only declaration given is that there was no financial support to declare. The paper does not state whether the authors have any competing interests.

Do not read these lower scores as proof that discourse declines with age in your own patients. The study compared different people at a single point in time rather than following anyone over years, and everyone was screened as cognitively healthy and educated to at least secondary level, so the differences may reflect who was in each group as much as ageing itself.