April 2-4, 2020 • Hyatt Regency • Lexington, KY
Intersectionality and Interdisciplinarity in Health Communication Research
Intersectionality and Interdisciplinarity in Health Communication Research
Abstract: "You Know You Are Getting Old When…": Improving Adherence to Mobile-Based Cognitive Assessments Using AI-Based Tailored Reminders
Declines in cognitive and perceptual abilities are generally part of the normal aging process. This becomes worrisome when these changes become severe enough to limit one’s ability to live independently and to do common everyday tasks. In the US, the early detection of cognitive declines is particularly critical with the rapidly increasing number of older adults and the increasing prevalence of age-related neurodegenerative dementias. In 2018, about 16% of Americans was 65 years or older [1]. This is projected to nearly double from 52 million in 2018 to 95 million by 2060. About 1 in 6 Americans 65 years and older have mild cognitive impairment. Moreover, the Alzheimer’s Association [2] estimates that 5.7 million Americans suffer from dementia. Of these, 60-70% are thought to suffer from Alzheimer’s disease. Alzheimer’s and other dementias are recognized as a major source of disease burden in the US, costing an estimated $277 million in healthcare costs for adults >65 years [2].
While there is still no cure for dementia or Alzheimer’s, current diagnosis relies on documenting mental decline, at which point it is usually too late to benefit from cognitive interventions. Early detection through regular neuropsychological tests can help individuals and their doctors identify interventions or lifestyle changes that may help with preserving cognitive functioning or identify potential risk factors that may lead to dementia. Fortunately, there are valid and reliable cognitive assessments that can be self-administered through mobile technology allowing for continuous remote monitoring of changes in cognition [3]. Since the sensitivity of these tests to detect major cognitive declines require periodic and regular assessment over time, long-term adherence is typically poor. We collected pilot data from older adults (N=120) engaged in home-based cognitive training over a 12-week period. By the end of the 12th week, only half of all participants engaged in the training.
The Adherence Promotion with Person-centered Technology (APPT) project aims to promote early detection and treatment of age-related cognitive decline through an artificial intelligence-based reminder system to help ensure long-term engagement with home-based cognitive assessments. Machine learning approaches will be used to predict patterns of adherence based on objective and subjective measures of adherence, technology proficiency, self-efficacy, behavioral intention to engage in training, attitudes toward cognitive training, and cognitive training history. This information will be used to create and schedule tailored reminders adapted to the daily routine of users and their context. The problem of understanding and promoting technology based-adherence requires a multi-disciplinary approach with a team that has extensive experience in cognitive training and assessment, older adult technology use, intervention design, tailored interventions, health data mining and analytics, aging and individual differences, and machine learning. This diverse knowledge base is crucial for the success of this multi-year NIH-funded project. This paper will describe progress on the development, pilot testing and refinement of the AI-based reminder system. This project is supported by NIA R01 AG064529.
While there is still no cure for dementia or Alzheimer’s, current diagnosis relies on documenting mental decline, at which point it is usually too late to benefit from cognitive interventions. Early detection through regular neuropsychological tests can help individuals and their doctors identify interventions or lifestyle changes that may help with preserving cognitive functioning or identify potential risk factors that may lead to dementia. Fortunately, there are valid and reliable cognitive assessments that can be self-administered through mobile technology allowing for continuous remote monitoring of changes in cognition [3]. Since the sensitivity of these tests to detect major cognitive declines require periodic and regular assessment over time, long-term adherence is typically poor. We collected pilot data from older adults (N=120) engaged in home-based cognitive training over a 12-week period. By the end of the 12th week, only half of all participants engaged in the training.
The Adherence Promotion with Person-centered Technology (APPT) project aims to promote early detection and treatment of age-related cognitive decline through an artificial intelligence-based reminder system to help ensure long-term engagement with home-based cognitive assessments. Machine learning approaches will be used to predict patterns of adherence based on objective and subjective measures of adherence, technology proficiency, self-efficacy, behavioral intention to engage in training, attitudes toward cognitive training, and cognitive training history. This information will be used to create and schedule tailored reminders adapted to the daily routine of users and their context. The problem of understanding and promoting technology based-adherence requires a multi-disciplinary approach with a team that has extensive experience in cognitive training and assessment, older adult technology use, intervention design, tailored interventions, health data mining and analytics, aging and individual differences, and machine learning. This diverse knowledge base is crucial for the success of this multi-year NIH-funded project. This paper will describe progress on the development, pilot testing and refinement of the AI-based reminder system. This project is supported by NIA R01 AG064529.