OBJECTIVES: Global mortality for noncommunicable diseases (NCDs) remains high and is increasing. The impact of major NCDs on developed European countries is disquieting. A risk predictor analysis of NCD outcome indicators is key to fight increasing global mortality and aid global monitoring. With this motivation, this study aimed to address the risk factors of NCD outcomes.

METHODS:

Data from the World Health Organization (WHO) were used to assess NCD country profiles in 194 countries for the year 2014. The proportional mortality (percentage of total deaths) from NCDs was used as a predictor variable, and tobacco consumption, alcohol consumption per capita, raised blood pressure, and obesity variables were used as covariates. A random forest plot was used to examine NCD-related mortality risk factors while performing a Classification and Regression Tree (CART) algorithm. Classification accuracy, specificity, sensitivity, and area under the receiver operating characteristic curve (AUC) were used as performance measures. A decision tree graph was used to visualize the risk predictors of NCD mortality.

RESULTS:

The CART algorithm showed classification accuracy, sensitivity, and specificity at values of 0.7425, 0.7889, and 0.7019, respectively. The AUC was 0.8285. Study results highlight that alcohol consumption per capita, in liters of pure alcohol, was the best predictor of NCD mortality. Raised blood pressure and obesity were also predictors of NCDs.

CONCLUSIONS:

The study results offer several insights for health policy makers in the fight against increasing global NCD mortality. Evidence-based strategies – helping adults change their drinking behaviors, promoting compliance with alcoholism treatment, limiting access to cheap alcohol, and encouraging networking and the exchange of experiences among countries – are essential to combat detrimental alcohol consumption. It is crucial that all stakeholders work together to develop holistic and sustainable policies that are opposed to alcohol consumption and NCD mortality.