Many patients have low levels of health literacy and their difficulties in using health information places them at risk. Low levels of health literacy may also be related to health disparities. Further, the increasing complexity of the healthcare system may present even greater challenges to individuals with chronic health conditions who must assume responsibility for many aspects of their healthcare. One solution to these problems is to give these individuals a tailored intervention focused on developing their chronic disease self-management skills in a way appropriate to their level of health literacy but identifying who has low health literacy continues to be a challenge. In this study, we present a model that can be used to identify persons with low levels of health literacy. It includes brief automated assessment of health literacy and readily available patient information and can easily be integrated into electronic health records. We used data from a study of a computer-administered health literacy measure with 280 adults 50 years of age or older (mean age 66; 180 women; 227 white and 53 black; 141 English- and 139 Spanish-speaking). Using latent profile analysis of their performance on questions that varied in their cognitive demand characteristics, we identified individuals likely to benefit from specific tailored interventions for chronic disease self-management, such as low literacy text and audio narration. Predictive analytic models (polytomous logistic regression) using demographic variables (age, gender, race, education, and language) and performance on a 10-item health literacy screen were used to identify individuals at one of four functional literacy levels (below basic, basic, intermediate, or proficient). The model predicting membership in one of the four levels had an accuracy of 74%. For the difference between those with below basic and basic compared to higher levels, the accuracy was much higher (93%; pseudo R2 = 0.85). This method to accurately identify individuals with low levels of health literacy can be used to tailor interventions delivered to those with low health literacy, and is currently being developed in a study of a mobile app for chronic disease self-management. We conclude that including a brief health literacy measure that can readily be included in electronic health records in combination with basic demographic information may allow quick and accurate identification of individuals who may need additional support in understanding written health information.