Four Pillars of the Rising T1DE Alliance
Targeting and predicting outcomes
The Rising T1DE leverages advanced machine learning and natural language processing to proactively identify those patients most at risk for future negative and avoidable outcomes. This process uses available health data to predict negative clinical and operational outcomes like deteriorating glycemic control, hospital readmission for diabetic ketoacidosis, and short-term cancellation of clinic appointments.
Curating and evaluating interventions
The Rising T1DE team nominates and prioritizes candidate interventions which are then prioritized by a multidisciplinary Stakeholder Advisory Committee based on criteria including measurable outcomes, stakeholders affected, population and characteristics of that population affected by outcomes, current state of research, and determines if the outcome is a driver of glycemic control.
Rapid-cycle testing using implementation science
The Rising T1DE aims to improve predicted and non-predicted outcomes via iterative small tests of change using quality improvement and implementation science methods to guide deployment. Multiple intervention strategies are tested against any given outcome during Plan-Do-Study-Act (PDSA) cycles. PDSA cycles test for improvement by developing a plan to test an idea for change (Plan), carrying out the test (Do), observing and learning from the consequences (Study), and determining whether dissemination or iteration plus repeat testing comes next (Act).
Circulating to other health centers
The Rising T1DE disseminates learnings from our testing of data- and technology-driven quality-improvement interventions to clinicians and healthcare systems across the country. In doing so, we accelerate local and national change to get in front of problems, offering a clear, systematic approach to tracking and scaling QI projects so that other healthcare providers can successfully adopt these novel interventions.