Canonical Correlation Analysis of Physiological and Anthropometric Variables in Hypertensive Patients

Abstract

The complex interplay between physiological and anthropometric characteristics in hypertensive patients remains poorly understood, despite its importance for diagnosis and management. This study uses canonical correlation analysis (CCA) to examine the multivariate association between physiological variables namely resting heart rate, systolic and diastolic blood pressure, vital lung capacity, and respiratory rate and anthropometric measures including age, height, weight, waist and wrist circumferences, forearm and arm lengths, sitting height, and shoulder width. A cohort of hypertensive patients was assessed, and canonical functions were derived to maximize the correlation between the two variable sets. The first canonical correlation was strong and statistically significant, suggesting a meaningful joint pattern linking physiological deregulation and bodily dimensions. The loadings and cross‑loadings revealed that certain anthropometric measures, such as wrist circumference and forearm length, contributed substantially to the canonical variates associated with key physiological outcomes. These results indicate that a small number of physical body‑size indicators may account for much of the shared variance with physiological stress markers in hypertension. While traditional CCA showed utility, the findings underscore the need to account for non‑normal and intercorrelated data dimensions commonly encountered in clinical settings. The insights provided by this multivariate approach have implications for more tailored assessments and interventions in hypertensive populations.