BMC Public Health. 2026 Jul 3. doi: 10.1186/s12889-026-27124-7. Online ahead of print.
ABSTRACT
BACKGROUND: Hypertension and obesity are the leading risk factors for cardiovascular disease (CVD), causing over ten million annual preventable deaths globally. Understanding individual differences in how blood pressure (BP) and weight changes over time can inform tailored prevention and treatment strategies.
METHODS: We used data from a longitudinal database of clinical and treatment records collected between 2015 and 2018 and linked to the Agincourt Health and Demographic Surveillance System (Agincourt HDSS-Clinic-Hospital link) in Mpumalanga Province, South Africa, to characterize long-term trends in BP and weight among hypertensive patients and assess the risk of death associated with homogenous group trajectories. K-means clustering was used to cluster the BP and weight trajectories. Multinomial logistic regression model was used to identify factors associated with emerging group trajectories. Finally, Cox Proportional hazard regression models were used to assess the mortality risk associated with each BP and weight group trajectory.
RESULTS: Three groups were identified among 1640 hypertensive patients and labelled according to the slopes of their BP and weight measurements. The first group was labelled as the steady BP and weight group (n = 760, 46% of patients) and consisted predominantly of patients from high socioeconomic status (SES) households, with a high level of education (secondary or higher education), and mostly women. The second group, labelled as the decreasing BP group (n = 264, 16% of patients), consisted predominantly of individuals from lower SES households and mostly older people of Mozambican ethnicity. The third group was labeled as the increasing BP and weight group (n = 616, 38% of patients) comprised of elderly men mostly from the middle and lower levels of household SES. The risk of dying was 1.82 times (95% CI [1.17-2.84]) higher for patients with increasing BP and weight as compared to patients with steady BP and weight.
CONCLUSION: Cluster-based characterization of hypertensive patients by BP and weight trajectory can facilitate personalized hypertension interventions and treatments.
PMID:42399830 | DOI:10.1186/s12889-026-27124-7