Association of Cardiovascular Disease and Life Habits: Smoking, Physical Activities and Alcohol Usage
In this study, a survey data collected by WHO with 69269 records was used to analyze the relationships between life habits and cardiovascular disease. The logistic regression model can be applied to fit the data with binary outcome and 6 covariates including age, BMI, gender, smoking status, alcoholism status, regular activities indicator. Based on the BIC criteria and backward model selection process, a final logistic regression model containing all covariates of interest was fitted. The model diagnostics and comparison methods showed that the logistic regression model fits data well with less high influential points and outliers. The conclusions are robust and consistent in different logistic regression models. The results suggested that non-smoker (OR = 0:93; P = 0:0388), regular physical activities (OR = 0:84; P < 0:001) are significantly associated with the decreased odds of presence of cardiovascular diseases. However, the alcoholic intake (OR = 0:94; p = 0:0875 > 0:05) shows no significant associations for the cardiovascular diseases.