Predictive Model on the Effect of Lifestyles in the Management of T2D

Type 2 Diabetes

T2D, CONSUMER DEVICES, SMARTWATCH, ARTIFICIAL INTELLIGENCE

This study aims to produce new evidence on the positive effects of physical activity and certain individual lifestyles in the control of type 2 diabetic disease. The goal is to build and evaluate the effectiveness of a new parsimonious risk prediction model based on the use of classical variables (blood exams), already used in other models for predicting the risk related to the disease, together with measures obtained from the use of wearable devices (steps count, sleep hours, heart rate).

Scientific literature has shown that the use of wearable devices can improve the control of type 2 diabetes mellitus. The aim of the study is to develop a statistical model for the control of type 2 diabetes mellitus (DMT2) based on the use of conventional clinical parameters and the use of unconventional data from wearable devices. The medical outcome variable that will be used to measure diabetes control and model performance will be the Glycated Hemoglobin (HbA1c). The participant will receive a Samsung smartwatch (Galaxy Watch 3) and a Samsung smartphone (A51), (the latter only in case the participant is not in possession of a compatible smartphone) which he/she must wear during his daily activities and also during the night to measure daily activity, sleep hours and heart rate. Casa Sollievo della Sofferenza Hospital previoulsy developted the ENFORCE calculator, an inexpensive and parsimonious prediction model of 2-year all cause mortality in real life patients with T2D. In this study the investigators aim to elaborate a standardized validated risk scores similar to what has been done for ENFORCE.

No related papers found yet.

Italy