Background: Hypertension is the leading cause of premature death globally, but the effectiveness of managing high blood pressure remains poor. There is a clear need for improved strategies for hypertension management. Researchers at the HRB Clinical Research Facility are developing an Artificial Intelligence Clinical Decision Support System (AI-CDSS) for Hypertension funded through the HRB Clinician Science Fellowship (CSF-2023-016).
Hypothesis: Hypertension treatment decisions from an AI-CDSS for hypertension are non-inferior to a gold standard panel of hypertension specialists.
Aim: The primary aim is to evaluate the accuracy of the AI-CDSS for hypertension in selecting the correct treatment decision in primary care patients with hypertension.
Research Design and Methods: This multi-center, prospective observational study will be conducted in primary care settings through the HRB Primary Care Clinical Trials Network. A total of 142 patients will be recruited from February 2024 to June 2025. The study involves no randomisation, as treatment decisions will follow routine clinical care. Historical GP visit data (up to one year before recruitment) will be included, along with data from subsequent visits during the study period. Data will be extracted from the Socrates electronic health record and stored in a Redcap research database. The AI-CDSS, developed using Python, TensorFlow, and Google Colab, will process each visit to generate treatment recommendations. These will be compared with decisions made by the gold-standard hypertension specialist panel in a blinded review process.
Objectives: 1) Validate the accuracy and reliability of data extraction from GP records. 2) Compare AI-CDSS treatment decisions with those of a gold-standard hypertension specialist panel.