Downloads: 0
Research Paper | Information Technology | Volume 15 Issue 7, July 2026 | Pages: 816 - 820 | India
Predicting Lifestyle-Related Health Risks in Adolescents Using Biological and Behavioural Data: A Critical Review and Conceptual Framework for Early, Data-Driven Risk Prediction in Youth (Ages 10-19)
Abstract: Adolescence (ages 10-19) is a formative window during which transient behaviours consolidate into enduring health trajectories, making it a decisive stage for preventing non-communicable disease. This paper examines how the integration of biological indicators- body mass index, resting blood pressure, heart rate, sleep duration, and family medical history- with behavioural signals such as physical activity, dietary quality, screen and social-media use, substance exposure, and perceived stress can enable the early prediction of lifestyle-related health risks through predictive analytics, artificial intelligence (AI), and machine learning (ML). Drawing on a structured narrative synthesis of literature published between 2019 and 2025, the study advances a biobehavioural predictive framework and critically evaluates supervised learning approaches- logistic regression, ensemble tree methods, and neural networks- for adolescent risk stratification. The analysis contends that although ensemble and deep-learning models achieve superior discrimination (reported AUCs of 0.80-0.92), their genuine clinical value depends on data quality, interpretability, equity, and ethical safeguards rather than headline accuracy. The paper concludes that responsibly governed, explainable models embedded within school and primary-care ecosystems represent the most credible route toward preventive adolescent health.
Keywords: adolescent health, predictive analytics, machine learning, lifestyle risk, behavioural data, preventive medicine
How to Cite?: Saanvi Bhambhani, Raghu Raja Mehra, "Predicting Lifestyle-Related Health Risks in Adolescents Using Biological and Behavioural Data: A Critical Review and Conceptual Framework for Early, Data-Driven Risk Prediction in Youth (Ages 10-19)", Volume 15 Issue 7, July 2026, International Journal of Science and Research (IJSR), Pages: 816-820, https://www.ijsr.net/getabstract.php?paperid=SR26710210732, DOI: https://dx.doi.org/10.21275/SR26710210732