Machine learning evaluation of a hypertension screening program in a university workforce over five years

Abstract

Abstract The global prevalence of hypertension continues excessively elevated, especially among low- and middle-income nations. Workplaces provide tremendous opportunities as a unique, easily accessible and practical avenue for early diagnosis and treatment of hypertension among the workforce class. The evaluation of such a Workplace Screening Strategy can give insight into its possible effects. Innovative machine learning approaches like k-means clustering are underutilized for such assessments. We set out to use this technology to analyze the results of our university’s yearly health checkup of the employees for hypertension. An anonymized dataset including the demographics and blood pressure monitoring information gathered from workers in various departments/units of a learning organization. The overall amount of samples or data values is 1, 723, and the supplied dataset includes six attributes, such as year group (2018, 2019, 2021, 2022), Department/Unit (academic and non-academic), and gender (male and female), with the intended output being the blood pressure status (low, normal, and high). The dataset was analyzed using machine learning approaches. In this longitudinal study, it was discovered that the average age for the workforce is 42. Similarly, it was revealed that hypertension was common among employees over the age of 40, regardless of gender or occupational type (academic or nonacademic). The data also found that there was a consistent drop in the prevalence of hypertension from 2018 to 2022. According to the study findings, the use of machine learning algorithms for periodic evaluations of workplace health status monitoring initiatives (particularly for hypertension) is feasible, realistic, and sustainable in diagnosing and controlling hypertension among those in the workforce.