To improve medication safety (AMTS), prediction models are being developed to identify patients at increased risk of specific hospitalizations at an early stage. These models complement traditional AMTS indicators, as not all preventable causes of adverse drug events (ADEs) are covered by standard criteria.
Based on routine data from statutory health insurance (GKV), the models aim to predict the risk of hospital admissions with certain diagnoses, such as gastrointestinal bleeding, fractures, or heart failure - regardless of the documented cause. In the future, this risk profile will be linked to current medication data from the electronic patient record (ePA). Particular need for action arises when both a high risk and the prescription of high-risk medications are present (e.g., NSAIDs or antithrombotic agents in patients at risk for GI bleeding).
Contact at StaBLab: Johannes Piller
Project partner: Prof Dr. Tobias Dreischulte (LMU University Hospital, Department of General Practice)