Online Program

Connection Between Physical Health and Mental Health: An Ensemble Learning Approach (305362)

Keywords: NHANES, Machine Learning, Ensemble Learning, Random Forest, Mental Health

Poor mental health is a complex public health problem which significantly affects vulnerable populations. Depression is also a risk factor for cardiovascular disease and medication nonadherence. Emotional distress and depression were recently identified as new risk factors for coronary artery disease, an important indicator of morbidity. Social determinants of health are critical to understanding how disadvantaged groups face barriers but importance is unclear. We proposed a machine learning (ML)-based system for predicting depression using a nationally representative sample from National Health and Nutrition Examination Survey (NHANES) data using social, physical, and behavioral risk factors. We assessed for depression by using the Patient Health Questionnaire-9. We applied several ensemble learning algorithms, in which the classifiers had a training-to-test split of 80% to 20%. In comparison to bagging, boosting (i.e. adaptive boosting and gradient boosting) was found to have superior accuracy to predict depression. Social predictors took precedence in relative importance in these models emphasizing the need to create models which further address health-related equity.