Characteristics of utterances produced by persons with dementia living in a long-term care facility: Lexical diversity, message clarity, and influence of conversational partners

The language of persons with dementia has been studied in the past, but rarely in the natural context of conversation. The purpose of this study was to examine lexical diversity and message clarity of language produced by persons with dementia during conversational interactions and the influence of conversational partners, if any, on their language characteristics. Six case studies were used to describe the language of people with dementia. Participants were also divided into two groups based on whether or not their communication partner had been trained. Quantitative and qualitative analyses were performed. The following lexical diversity variables were measures: noun rate, pronoun rate, adjective rate, and verb rate. Additionally, Brunét's index, Honoré's statistic, and type-token ratio were also calculated to quantify vocabulary richness. The following message clarity variables were measured in percentages: utterances with a verbal maze, unfinished utterances, fully unintelligible utterances, partially unintelligible utterances, utterances with an inappropriate topic shift, and pronouns with a clear referent. Results revealed that persons with dementia produced more pronouns compared to other content words. Some participants produced language that was lexically rich while others did not. Measures of message clarity varied across participants. There were no significant differences between trained and untrained conversation partner groups on any of the measures. Qualitative analyses revealed some communication partner techniques that seemed to enhance language production in persons with dementia. This study identified some lexical diversity and message clarity measure that could be obtained from a brief conversational language sample and used to assist in the diagnosis and treatment of persons with dementia. Additionally, the study provided insight regarding the contribution of conversational partners to the language of persons with dementia.