Researchers at Yale University have developed a machine-learning-based clinical decision support tool to provide personalized recommendations on whether to pursue intensive or standard blood pressure treatment goals.
Through a data-driven approach, the tool described in The Lancet Digital Health earlier this week facilitates shared decision-making between providers and patients with hypertension. It is a leading cause of heart disease and mortality and is defined as sustained blood pressure greater than 140/90 mm Hg.
Researchers in the study noted that lowering blood pressure is crucial to reducing these risks, but the extent to which blood pressure should be lowered has been disputed. According to the study, aggressive blood pressure control for patients with Type 2 diabetes has been inconclusive in clinical trials.
For patients with and without diabetes, the research team developed a ML-based model to personalize blood pressure management treatment goals.
Firstly, the researchers collected data from two randomized clinical trials: the Systolic Blood Pressure Intervention Trial (SPRINT), which did not include diabetic patients, and the Action to Control Cardiovascular Risk in Diabetes Blood Pressure (ACCORD BP), which only included diabetic patients. Each trial randomized patients to an intensive or routine systolic blood pressure goal of 120 mm Hg or 140 mm Hg.
In comparison, the ACCORD BP trial showed that intensive blood pressure treatment was ineffective, while SPRINT showed the value of lowering blood pressure. In order to develop an ML model that identifies the characteristics of patients most likely to benefit from intensive blood pressure lowering, the team used data from SPRINT to identify 59 variables, including kidney function, smoking, and statin or aspirin use.
When PRECISION was applied to ACCORD BP, researchers found that it could identify patients with diabetes who could benefit from aggressive blood pressure management.
In a press release, Rohan Khera, MD, assistant professor at Yale School of Medicine and director of the Cardiovascular Data Science (CarDS) Lab, said that it is difficult to determine the appropriate blood pressure targets and treatment course for hypertension and diabetes patients. “Here, we used machine learning to enhance inference from two landmark clinical trials in assessing a personalized cardiovascular benefit of intensive blood pressure control. The key finding is that the benefit profile derived in patients without diabetes seems to define those with diabetes that benefit from such a treatment strategy.”
PRECISION appears to provide reliable, practical insights to inform decisions regarding intensive versus standard systolic blood pressure management among diabetic patients, according to the researchers. A variety of factors contribute to the risks and benefits of an intensive blood pressure-lowering strategy, they noted, but additional testing is needed.
This research comes amid concerns about blood pressure control and rising chronic disease prevalence.
