Contribution of an Intelligent Virtual Assistant to Healthy Ageing in Adults With Type 2 Diabetes

Abstract

This chapter describes the development of a theory-driven and evidence-based digital intervention to facilitate self-care in older adults with Type 2 Diabetes (T2D) and, additionally, its contribution to healthy aging and the individual care plan. T2D is highly prevalent in older adults. Difficulties in adopting and maintaining desirable self-care behaviors is associated with lack of glycemic control and subsequent complications, which significantly burden patients, their families, and the health system. The VASelfCare (Virtual Assistant Self-Care) intervention is a software application that provides an interface with a 3D anthropomorphic virtual assistant targeting three key self-care behaviors: medication-taking, physical activity, and a healthy diet. Other VASelfCare elements are intended for nurses providing diabetes consultations, including a web-based back-office with a patient data dashboard, which streamlines integration of care. The application prototype has been co-produced with older adults with T2D, primary care health professionals, and other stakeholders.

Usability Evaluation by Primary Care Providers of a Novel Digital Intervention for Type 2 Diabetes Self-Management in Older Adults

“…Input consists of options depicted in response buttons or recorded values. Dialogues resort to an artificial intelligence rule-based engine 14 …”

“…Vitoria uses evidence-based behavior change techniques from a widely accepted taxonomy, 15 operationalized according to tailoring principles, based on evaluation data and previous interactions 14,16,17 Figure 1. depicts the operationalization of the behavior change technique “4.1 Instruction on how to perform the behavior” in the healthy eating intervention component.…”

“…When Internet access is available, data are exported to a MySQL database hosted at an institutional server, powering the VASelfCare Web-based dashboard prototype for primary care providers. This prototype, which has been detailed elsewhere, 14 intends to facilitate integrated care by providing access to users' personal data and mobile application records transformed into indicators, such as percentage of self-reported medication adherence and number of steps over a time period (Figure 2). The Web-based dashboard is available through login with a password procedure; primary care providers can only access data of persons to whom they are providing care.…”

“…14 Vitoria uses evidence-based behavior change techniques from a widely accepted taxonomy, 15 operationalized according to tailoring principles, based on evaluation data and previous interactions. 14,16,17 The application stores all the interaction records of a user in a local database. When Internet access is available, data are exported to a MySQL database hosted at an institutional server, powering the VASelfCare Web-based dashboard prototype for primary care providers.…”

“…Dialogues resort to an artificial intelligence rule-based engine. 14 Vitoria uses evidence-based behavior change techniques from a widely accepted taxonomy, 15 operationalized according to tailoring principles, based on evaluation data and previous interactions. 14,16,17 The application stores all the interaction records of a user in a local database.…”