The use of Glucagon-like peptide-1 receptor agonists (GLP-1s) has seen a significant increase in recent years, with 1.7% of Americans being prescribed the drugs in 2023, compared to a 40-fold increase over the past five years. However, the lack of expertise in healthcare organizations has posed challenges for clinicians in determining the suitability of these medications for their patients.
To address this issue, leaders at Emory University have developed a digital twin technology called Tacit Object Modeler (TOM) to provide clinical decision support for GLP-1 prescribing. TOM works by modeling the decision-making process of expert clinicians, such as Dr. Caroline Collins, to guide appropriate prescribing decisions for patients. The tool takes into consideration factors such as lifestyle medicine and primary care expertise to ensure the accuracy of clinical decision support.
The process of developing TOM involves interviewing the expert clinician to determine key variables and measures that influence their decision-making. By creating unique clinical scenarios based on these variables, the tool can learn how the clinician weighs each factor to make prescribing decisions. Generalizing the model to a wide variety of patients is crucial for its effectiveness, and efforts are made to mitigate potential biases by focusing on core variables informed by the expert.
The use of digital twins like TOM has the potential to transform medical professionals’ distribution of knowledge and streamline decision-making processes. By harnessing experts’ wisdom and lived experience, these tools aim to support clinicians in providing optimal care to their patients amidst rapidly evolving healthcare landscapes.
Source
Photo credit www.techtarget.com