About 30 million adults in the United States have diabetes and it is estimated that nearly 90 million have pre-diabetes with an elevated risk of developing diabetes. Current models to predict changes in glycemic control perform poorly and do not take into account daily health behaviors. Automated monitoring of these behaviors could inform enhanced risk prediction models that can detect patients that are likely to have a decline in their glycemic control. If these patients could be identified, effective approaches could be better targeted towards them to improve outcomes in a more timely manner.
This study will aim to compare different methods to use data on daily health behaviors collected by wearable devices to enhance risk prediction models. Participants will be randomly assigned to use a waist-worn or wrist-worn wearable device for 6 months.
This study is complete and results are forthcoming.
Center for Health Incentives & Behavioral Economics, Penn LDI