As loneliness continues to affect seniors in long-term care amid the pandemic, researchers at the University of California, San Diego School of Medicine are using artificial intelligence to study the issue among residents.
The scientists harnessed natural language processing and machine learning to classify the sentiment and emotions of speech during interviews with 80 older adults aged 66 to 94 years, who resided in independent living at a seniors housing community in San Diego County. Using the tools allowed the team to identify degrees of loneliness among the residents, which could prove vital in helping society assess and address widespread loneliness, said senior author Ellen Lee, M.D., assistant professor of psychiatry at UC-San Diego School of Medicine.
“Most studies use either a direct question of, ‘How often do you feel lonely?’ which can lead to biased responses due to stigma associated with loneliness, or the UCLA Loneliness Scale, which does not explicitly use the word ‘lonely,’” Lee said. In addition, similar emotion analyses by humans often lack consistency and require extensive training to standardize.
Early findings said that “lonely speech” could be used to detect loneliness in older adults, improving assessment and treatment, especially during physical distancing and social isolation periods. The researchers found that lonely individuals had longer responses during interviews and expressed greater sadness to direct questions about loneliness, and men and women expressed their loneliness in different ways.
The UC-San Diego researchers also noted that the machine-learning models predicted qualitative loneliness with 94% accuracy. Full study findings are published in the American Journal of Geriatric Psychiatry.
From the November 2020 Issue of McKnight's Long-Term Care News
