Research conducted at the Florida State University College of Medicine has discovered a potential inexpensive method for predicting the risk of developing dementia. The study analyzed data from approximately 13,000 participants in a long-term aging study and found that an interviewer’s assessment of a cognitively healthy person’s memory was able to successfully predict the likelihood of developing dementia over a 15-year period. These findings will be published in the Journal of Alzheimer’s Disease.
The research author, Angelina Sutin, who is a professor of Behavioral Sciences and Social Medicine, stated, “Our findings demonstrate that interviewers were capable of detecting memory deficits in participants that indicated a higher risk of developing dementia over time, especially for those who performed well on objective memory tests.”
The results suggest that an interviewer’s rating of a person’s memory performance could be a valuable addition or alternative to other detection methods like self-reporting or cognitive testing.
Sutin’s team analyzed 15 years of data from nearly 13,000 people without cognitive impairment at the beginning of the study. These individuals participated in the University of Michigan Health and Retirement Study (HRS), wherein their memory was rated by the interviewer. The HRS is a longitudinal study that surveys a representative sample of adults aged 50 and older every two years about various aspects of their well-being.
To be included in Sutin’s sample, participants had to have been interviewed in 2006, scored within the normal range of cognitive function during their first interview, and had at least one follow-up assessment of cognition between 2008-2020.
The interviewers were trained research assistants working on the HRS, who conducted 2-3 hour interviews and rated the difficulty the participants had in remembering things on a scale of 1 (no difficulty) to 5 (could not do at all).
The interviewer-rated memory was used as a predictor for incident dementia over the 15-year follow-up period. The study found that each 1-point increase in poor memory (as rated by the interviewer on a 1-5 scale) was associated with a 40% increase in the risk of developing dementia. This association remained significant even after considering factors such as depression and poor hearing. The mode of interview (face-to-face or over the phone) did not affect the results.
Of particular interest is the fact that the association was even stronger among participants who had the best objective memory performance (remembering many words from a long list of words) and subjective memory (how well an individual perceives their memory). Both objective and subjective memory have been useful in identifying cognitive deficits that precede a dementia diagnosis.
Remarkably, the results were consistent even among participants who initially scored in the top quartile of memory function. Such scores on an objective memory test typically indicate good cognitive function and low risk of impairment.
Sutin commented, “This simple rating by an interviewer predicts who will develop dementia, especially when traditional measures of memory function fail to detect memory deficits. Furthermore, the association is similar across different age groups, genders, races, ethnicities, and educational backgrounds, suggesting that it may have broad predictive value across populations. In summary, we believe that interviewer-rated memory serves as a reliable marker for future dementia among the most cognitively healthy individuals.”
These findings support the increasing evidence for the importance of subjective memory ratings and extend the association to outside observers who may not be familiar with the person being assessed. Typically, clinicians rely on family members to gather information about cognitive function. However, this study suggests that interviews with strangers can also provide valuable insights into prospective cognitive health.
Based on the findings, a simple interviewer rating has long-term potential for predicting the development of dementia and could be beneficial for clinicians in delivering effective treatment.
Co-authors of the study include Professor Antonio Terracciano and Assistant Professor Martina Luchetti from Florida State University, Darmris Aschwanden from the University of Geneva in Switzerland, and Yannick Stephan from the University of Montpellier in France.
The National Institute on Aging of the National Institutes of Health provided support for this research.