Social isolation is a major public health epidemic. According to Statistics Canada, 1.4 million elderly Canadians report feeling lonely. Almost 30% of Canadians live alone, and up to 50% of people over the age of 60 are at risk of social isolation due to factors such as: outliving family and friends, disability, life-threatening illness, caregiving, and low income. Loneliness and isolation are significant risk factors for depression, anxiety, dementia, heart disease, diabetes, early all-cause mortality, and lower quality of life. People experiencing structurally inequality and health inequity are more greatly impacted. This situation is now greatly exacerbated by the COVID 19 pandemic. Responsive solutions are critical to the health of individuals, and to pandemic recovery. It has been shown that early preventative intervention methods are more beneficial than curative methods to address the problem of social isolation in older adults. Current healthcare models are inadequate to address such issues. Determining who is at risk and implementing interventions to solve this issue can be challenging as the problem of social isolation is multidimensional. When socially isolated individuals are identified, interventions at various levels can be implemented to decrease the negative impact of social isolation on the health of older adults and other vulnerable groups.

This research project aims to develop and test a Minimally Viable technology platform (MVP) to alert communities to social isolation among seniors, suggest matching recommendations and other options to address, and track and generate feedback on changes and results.

Our proposed solution is enabled through an artificially intelligent (AI) analytics platform that will reorient communities to the whole person, value-based care. It will consist of a core computational engine and multiple analytical tools that enable it to be used by a wide range of users (end-users, community and health providers, scientists). This platform is designed to: (1) detect individuals at risk of isolation or poor health in a community using AI, survey and social network analysis techniques, (2) to recommend and match them to a range of community resources, support groups, volunteers, and education/skills based on their preferences, and (3) to track changes in loneliness, quality of life, and health utilization as a result of these matches, feeding this data back to the community to prompt continuous monitoring and improvement.