Objective To investigate the health literacy levels among hypertensive patients in mountainous areas of Anqing City and establish a risk prediction model. Methods From January to September 2025, adult hypertensive patients with established health records were selected from five administrative villages in Anqing""s mountainous areas using multistage random sampling. Data were collected via the National Health Literacy Monitoring Survey Questionnaire. Multiple Logistic regression analysis was used to identify the influencing factors and construct a prediction model, and its performance was evaluated by receiver operator characteristic (ROC) curve . Results The overall health literacy attainment rate among hypertensive patients in Anqing""s mountainous areas was 7.50% (9/120), with an average score of (32.88±13.05) points. Multiple Logistic regression analysis revealed that educational attainment, occupation, per capita annual household income, and self-rated health status were all independent factors influencing health literacy (all P＜0.05). The area under the curve (AUC) of the predictive model was 0.891 (95%CI: 0.819～0.963), with a sensitivity of 80.00% and specificity of 87.50%. Conclusion Health literacy levels among hypertensive patients in mountainous areas of Anqing City are generally low, influenced by factors such as educational attainment, occupation, income, and self-rated health status. The constructed risk prediction model demonstrates good predictive efficacy, providing scientific basis for early screening, stratified management and targeted interventions to improve health literacy among hypertensive patients in mountainous regions.