The objective of this study was to determine the main patterns and factors influencing food transition in riverine people in the Brazilian Amazon. Through interviews with a semi-structured questionnaire, we inferred their food habits and provide information about general demographic, socioeconomic, resource use and environmental context. Data from the questionnaires were categorized and analyzed using a logistic regression model to assess the relative influence of socioeconomic and environment factors on the local diet. Based on a logistic regression data analysis, it was found a greater consumption of processed food significantly associated with multiple factors such as market participation, sex (female and male), government aid to forest conservation and environment context (upland and wetland). Although the local diet is composed mainly of local resources such as fish and cassava flour, increasing incomes due to direct government subsidy programs and marketing of cassava flour have influenced these local practices and habits. Through the analysis of factors influencing food transition, it was possible to evaluate those having the greatest effect on this Amazon region and propose an alternative method to subsidy food policy grounded in local opinion surveys.