# The two-stage exploratory association between hypertension and physical examinations: a community-based biobank study

## TL;DR

This study explores how clinical exams and health behaviors relate to hypertension using data from the Taiwan Biobank.

## Contribution

The study identifies key risk factors and develops a predictive model for hypertension using clinical and lifestyle data.

## Key findings

- Hyperlipidemia and diabetes are significant risk factors for hypertension.
- Clinical biochemistry indicators predict hypertension with high accuracy (AUROC 0.8769).
- Age, heart rate, and lipid levels are strong predictors of hypertension.

## Abstract

Hypertension is one of the most critical public health problems in developing countries and a leading cause of mortality and disability. Relevant extrapolations show that the global adult population with high blood pressure will increase significantly. This study utilizes partial data from the Taiwan Biobank to explore the association between clinical examinations and hypertension-related morbidities.

Data for this study were sourced from the Taiwan Biobank, which has collected health information since 2012 and serves as a partial source for scientific research into lifestyle and health trends of cardiovascular-related diseases in the general population. This study focused on distinguishing the correlation between blood pressure changes and stability in various age groups and exploring the relationship between environmental exposure factors and health behaviors through stratified analysis.

The comorbidities identified as significant risk factors for hypertension include hyperlipidemia [odds ratio (OR), 4.0504] and diabetes (OR, 2.1871). Clinical biochemistry examinations also indicated classifiers for hypertension, such as age, heart rate, triglycerides, high-density lipoprotein cholesterol, hyperlipidemia, and diabetes, which were represented as explanatory indicators [the area under the receiver operating characteristic curve (AUROC), 0.8769]. These findings underscore the potential of clinical examinations to predict and prevent hypertension.

This study suggested the possibility of developing a risk assessment tool based on these classifiers and investigated the generalizability of these findings using biobank resources. The findings could aid in informing clinical decision-making, enhancing digital health education, and reducing the burden of hypertension.