Socioeconomic, demographic and geographic factors affecting food purchase and acquisition decisions by the individuals and households in the United States are complex. Twenty two variables associated with socioeconomic, demographic and geographic factors affecting food purchase and acquisition decisions are studied in an algorithm that involve machine learning and causality structures to uncover complex causality patterns associated with such variables. Several publically available data sets as well as most recently collected USDA data, Food APS are used in this study. Preliminary analyses show that unemployment, poverty and race are direct causes of food insecurity, while income causes food insecurity via poverty. Unemployment is a common cause for both food insecurity and poverty. This modeling effort will help improve the nation’s path for effective nutrition and health-related policy interventions as stipulated by USDA-Economic Research Service Strategic Goals.