Predictive modeling of systolic blood pressure in elderly patients using multiple linear regression based on sleep duration, physical activity, and sodium intake
Computational Design: Journal of Modern Informatics · 30.09.2026 · 1 (2) · 151-157
DOI: 10.35730/z7ebpy34 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computational Design: Journal of Modern Informatics, online erschienen am 30.09.2026. Verfasst von Hezy Kurnia und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Hypertension among the elderly population represents a major global public health challenge, substantially elevating the risk of cardiovascular morbidity and mortality. Non-pharmacological lifestyle modifications remain the cornerstone of managing stage 1 hypertension in geriatric care. This study aims to quantify the simultaneous effects of nocturnal sleep duration, weekly aerobic physical activity, and daily dietary sodium intake on systolic blood pressure (SBP) in elderly patients using a multiple linear regression approach. A synthetic clinical dataset of 100 elderly patients aged 60 years and older was generated and analyzed using Python (scikit-learn and statsmodels libraries). The dataset was partitioned into 80% training data and 20% testing data to evaluate predictive performance, while Ordinary Least Squares (OLS) estimation was applied to the complete cohort to assess statistical significance. The multiple linear regression model demonstrated strong explanatory power, accounting for 77.4% of the variance in systolic blood pressure (R-squared = 0.774, Adjusted R-squared = 0.767, F = 109.7, p < 0.001) on the full dataset, and achieved an R-squared of 0.509 with a Mean Squared Error (MSE) of 33.63 mmHg squared on the unseen test set. All three lifestyle predictors were statistically significant (p < 0.001). Specifically, each additional hour of nocturnal sleep was associated with a 2.67 mmHg reduction in SBP (B = -2.6673), and each additional minute of weekly physical activity reduced SBP by 0.037 mmHg (B = -0.0372, equivalent to a 1.12 mmHg reduction per 30 minutes of exercise). Conversely, each additional gram of daily sodium intake increased SBP by 4.38 mmHg (B = 4.3764). Multiple linear regression provides an interpretable, quantitative framework for geriatricians to estimate personalized blood pressure improvements resulting from targeted lifestyle interventions.
Zitieren
APA
Kurnia, H., Kurniawan, I., & Ediana, D. (2026). Predictive modeling of systolic blood pressure in elderly patients using multiple linear regression based on sleep duration, physical activity, and sodium intake. Computational Design: Journal of Modern Informatics, 1(2), 151-157. https://doi.org/10.35730/z7ebpy34
Vancouver
Kurnia H, Kurniawan I, Ediana D. Predictive modeling of systolic blood pressure in elderly patients using multiple linear regression based on sleep duration, physical activity, and sodium intake. Computational Design: Journal of Modern Informatics. 2026;1(2):151-157. doi:10.35730/z7ebpy34
BibTeX
@article{Kurnia2026,
author = {Kurnia, Hezy and Kurniawan, Irfan and Ediana, Dina},
title = {Predictive modeling of systolic blood pressure in elderly patients using multiple linear regression based on sleep duration, physical activity, and sodium intake},
journal = {Computational Design: Journal of Modern Informatics},
year = {2026},
volume = {1},
number = {2},
pages = {151--157},
doi = {10.35730/z7ebpy34}
}
EUVIMED