Empirical Study and Statistical Analysis of Risk Factors for Cardiovascular Disease (CVD) in Sindh Province

OBJECTIVE: To gather information from Sindh's Civil Hospitals to discover crucial clinical risk factors for CVD in that state.METHODOLOGY: Data were collected between January and December 2022, focusing on 21 potential cardiovascular risk factors.Hypotheses were tested, and essential clinical risk factors were found using logistic regression.Additionally, this study uses algorithms such as Decision Tree (DT), Random Forest (RF), Logistic Regression (LR), and Support Vector Machine (SVM) for classification and prediction, splitting the data into training and validation sets.ROC curves are also utilized to evaluate these machine-learning classifiers.RESULTS: The study showed that 1358 (75.2%) of 1800 respondents had CVD.Moreover, logistic regression results for hypothesis testing showed that multiple variables are statistically significant risk factors for CVD patients at the = 0.05 level of significance for developing cardiovascular disease (CVD).Additionally, the Random Forest model outperforms DT, LR, and SVM as the most accurate predictor of CVD (77% accuracy).CONCLUSION: Based on the findings, Random Forest (RF) outperformed with 77% accuracy then the existing models are used in terms of classification and predictions.