Generalized Anxiety Disorder (GAD) is a mental disorder characterized by persistent and excessive worry about everyday situations. In the area of mental health, machine learning has proven useful for classifying psychological disorders based on sociodemographic, personality and physiological variables. This article presents a model for the classification of GAD based on the Random Forest algorithm, complemented with the SHAP (SHapley Additive exPlanations) explanatory technique. A total of 280 records were collected using four questionnaires that assessed life history strategies, sociodemographic characteristics, and levels of worry. After preprocessing, feature selection and class balancing, different models were assessed. Random Forest model combined with the ADASYN technique achieved the best results, with an accuracy of 82% and sensitivity of 80%. The analysis of variable importance using SHAP identified gender, impulsivity, visuospatial ability, resilience to adverse situations and the K factor as the most relevant factors. These findings support the feasibility of integrating life history variables into machine learning models for the detection of GAD and provide valuable information for its understanding from a developmental and clinical perspective.