<![CDATA[Hepatitis is an inflammatory liver disease caused by viral infection and remains a major global public health concern, responsible for approximately 1.4 million deaths annually. Egypt is among the countries with the highest prevalence of Hepatitis C. To address this issue and support Goal 3 of the Sustainable Development Goals (SDGs), this study applies a quantitative approach using secondary data to analyze factors influencing Hepatitis C infection in Egypt. Two statistical models Binary Logistic Regression and Multivariate Adaptive Regression Splines (MARS) were compared, with the SMOTE method implemented to correct class imbalance. The dataset consisted of 608 patient observations, initially imbalanced at a ratio of 86.5:13.5, and were balanced to 52.6:47.4 after SMOTE application. The results revealed that the MARS model demonstrated superior predictive performance compared to binary logistic regression. All independent variables were found statistically significant (p < 0.05), except sex. Additionally, all odds ratios were less than 1, indicating a lower probability of Hepatitis C infection relative to non-infection. These findings highlight the relevance of statistical modeling and data-driven strategies in supporting preventive health measures. ]]>
Copyrights © 2026
