ADL+: A Digital Toolkit for Multidomain Cognitive, Physical, and Nutritional Interventions to Prevent Cognitive Decline in Community-Dwelling Older Adults

Justin Chew, Zhiwei Zeng, Toh Hsiang Benny Tan, Pamela Chew, Noorhazlina Ali, Hao Wang, Nalini Singhal, Roslyn Raymond, Kalene Pek, Di Wang, Liang Zhang, Zhiqi Shen, Cyril Leung, Jing Jih Chin, Wee Shiong Lim, Chunyan MiaoInternational Journal of Environmental Research and Public Health 2024 · DOI

TL;DR

This paper introduces ADL+, an AI-powered digital toolkit designed to combat cognitive decline in older adults by integrating cognitive training, physical activity, and nutritional guidance. The study found that ADL+ significantly improved cognitive function, maintained activity levels, and enhanced quality of life in participants at risk of cognitive decline, demonstrating the potential of technology-driven, personalized interventions.

What Problem Does This Paper Address?

Cognitive decline and dementia pose a growing global health challenge, impacting individuals and healthcare systems. While research suggests that addressing multiple risk factors through multidomain interventions can delay or prevent cognitive impairment, implementing these interventions effectively in community-dwelling older adults is challenging. Factors like variable adherence, the need for personalized approaches, and the difficulty of integrating various health aspects limit current strategies. This paper addresses the need for accessible, engaging, and personalized technology-driven solutions to deliver these multidomain interventions and prevent cognitive decline.

Methods

The study employed a quasi-experimental design to evaluate the ADL+ toolkit. ADL+ is an AI-enabled platform that integrates several modules:

  • Smart Day Activity: An intelligent system for scheduling, guidance, and incentives.
  • Virtual ADL+ House: Gamified exercises simulating daily living activities.
  • Online Trail-Making Test (TMT): A Kinect-based adaptation for attention and executive function assessment.
  • Physio-Cognitive Ping Pong: A dual-tasking game combining physical and cognitive exercise.
  • Cognitive Stimulation: An app-based game for various cognitive exercises.
  • Diet Analysis and Recommendation: An AI-driven tool for tracking and analyzing food intake from meal photos.
  • Predictive Model: An AI component that uses collected data to assess cognitive functioning and personalize interventions.

Participants in the intervention group used the ADL+ toolkit for six months, while the control group received educational materials. Outcomes were assessed using neuropsychological tests (NTB composite and individual domains), the Frenchay Activities Index (FAI) for activity levels, and the EQ-5D questionnaire for quality of life. Adherence and usability were also measured.

Datasets

The study did not introduce new benchmark datasets. The Diet Analysis module utilized a local nutrition database from the Health Promotion Board, which included a dataset of over 500 dishes and 140,000 images for food recognition.

Key Insights and Ideas

  • AI-driven personalization is key: The ADL+ toolkit's AI component is crucial for adapting interventions to individual needs, which is a significant step beyond one-size-fits-all approaches.
  • Multidomain approach shows promise: Combining cognitive, physical, and nutritional interventions within a single platform appears more effective than isolated approaches for preventing cognitive decline.
  • Technology can enhance engagement: Gamification and interactive elements within ADL+ were well-received, suggesting technology can motivate older adults to participate in health-promoting activities.
  • Usability matters for adoption: While overall usability was rated as average, specific feedback on learning curves and technical issues highlights areas for improvement in future digital health tools for this demographic.
  • Sustained benefits are challenging but possible: While not all cognitive benefits were sustained post-intervention, improvements in processing speed and quality of life suggest long-term positive impacts.

Key Results

  • Cognitive Function: The intervention group showed a significant improvement in the NTB composite score (0.086) at six months, with a notable between-group difference of 0.17. Improvements were also seen in attention, processing speed, and memory domains. These benefits, particularly for processing speed, were sustained at nine months.
  • Activity Levels: While the control group's FAI scores declined significantly (-1.04) at six months, the intervention group's scores remained stable.
  • Quality of Life: The intervention group showed significant improvements in EQ-5D VAS scores, with mean differences of 4.06 at six months and 5.12 at nine months compared to the control group.
  • Adherence and Usability: Adherence to the ADL+ toolkit was high, with an overall median rate of 95%. The System Usability Scale (SUS) score was 70.7 (out of 100), indicating average usability. Participants particularly enjoyed the telephone mini-game, Online TMT, and Physio-Cognitive Ping Pong.

Limitations and Future Work

The authors acknowledge several limitations:

  • Quasi-experimental design: The study's design and baseline group imbalances may have introduced bias, despite statistical adjustments.
  • Limited follow-up: The six-month intervention and three-month post-intervention follow-up leave the long-term sustainability of effects unexplored.
  • Generalizability: The findings may not be generalizable to all older adults due to factors like technology acceptance, cultural context (Singapore), and varying educational levels.

Future work suggested includes:

  • Randomized controlled trials: To provide stronger evidence of efficacy.
  • Implementation research: To evaluate ADL+'s effectiveness in real-world settings beyond a research study.
  • Adaptations for different populations: Tailoring the intervention to diverse demographic and cultural groups.
  • Longer-term follow-up: To assess sustained effects and explore strategies like booster sessions.
  • Addressing technical challenges: Improving user interface design, clarity of instructions, and graphic element size to enhance usability.

Summary updated

Abstract

The ADL+ toolkit shows potential beneficial effects on cognitive function, activity levels, and quality of life for older adults at risk of cognitive decline. Findings will guide future randomized controlled trials and implementation efforts.
BibTeX
@article{chew2024adl,
  author = {Justin Chew and Zhiwei Zeng and Toh Hsiang Benny Tan and Pamela Chew and Noorhazlina Ali and Hao Wang and Nalini Singhal and Roslyn Raymond and Kalene Pek and Di Wang and Liang Zhang and Zhiqi Shen and Cyril Leung and Jing Jih Chin and Wee Shiong Lim and Chunyan Miao},
  title = {{ADL+: A Digital Toolkit for Multidomain Cognitive, Physical, and Nutritional Interventions to Prevent Cognitive Decline in Community-Dwelling Older Adults}},
  journal = {International Journal of Environmental Research and Public Health},
  year = {2024},
  doi = {10.3390/ijerph22010042},
}
Figure