Application of Discriminant Analysis for Blood Pressure Classification Based on Vital Signs: Evidence from a Regional Hospital in Ghana

Background and AimsHypertension and diabetes mellitus are major non-communicable diseases and leading contributors to cardiovascular morbidity and mortality, particularly in low and middle-income countries. Both conditions frequently coexist and share common risk factors, including obesity and advancing age. Early identification of abnormal blood pressure status using routinely collected clinical data may enhance timely intervention and reduce complications associated with hypertension and diabetes. This study aimed to develop and validate a discriminant model that classifies patients as hypotensive, normotensive, or hypertensive using common vital sign indicators, and to evaluate the predictive contribution of individual variables within the broader context of cardio metabolic risk. MethodsThis retrospective observational study analyzed secondary data from 1,000 adult patients at a regional hospital in Ghana. Linear discriminant analysis (LDA) was applied using age, heart rate, body temperature, and body weight as predictors of systolic blood pressure classification. Model performance was assessed using cross-validation and classification matrices. Receiver operating characteristic (ROC) analysis was conducted to evaluate the discriminatory ability of individual predictors. ConclusionsRoutinely collected vital sign data, particularly body weight, can accurately classify blood pressure status. The high classification accuracy observed supports the feasibility of data-driven risk stratification in clinical settings. These findings further underscore the importance of weight management in the prevention and control of hypertension and related cardio metabolic conditions, including diabetes.