Constrained econometric techniques hamper investigations of disease prevalence and income risks in the shrimp industry. We employ an econometric model and machine learning (ML) to reduce model restrictions and improve understanding of the influence of diseases and climate on income and disease risks. An interview of 534 farmers with the models enables the discernment of factors influencing shrimp income and disease risks. ML complemented the Just-Pope production model, and the partial dependency plots show nonlinear relationships between income, disease prevalence, and risk factors. Econometric and ML models generated complementary information to understand income and disease prevalence risk factors.