A model to determine absolute risk for esophageal adenocarcinoma.

BACKGROUND & AIMS: Esophageal adenocarcinoma (EAC) develops rapidly and has a high mortality rate. We aimed to develop a prediction model to estimate the absolute 5-year risks, based on different profiles of factors, for developing EAC.
METHODS: We derived a risk model using epidemiologic data from 364 patients with incident EAC and 1580 population controls. Significant risk factors were fitted into an unconditional multiple logistic regression model. The final model was combined with age- and sex-specific EAC incidence data to estimate absolute 5-year risks for EAC. We performed a 10-fold cross-validation of the data to assess the relative performance of the model.
RESULTS: The final risk model included terms for highest level of education, body mass index, smoking status, frequency of gastroesophageal reflux symptoms and/or use of acid-suppressant medications, and frequency of nonsteroidal anti-inflammatory drug use. The population attributable risk for the model was 0.92. A 10-fold cross-validation produced an area under the receiver operating characteristic curve statistic of 0.75 (95% confidence interval, 0.66-0.84), indicating good discrimination. Adding data on alarm symptoms, frequency of symptoms of dysphagia, and unexplained weight loss to the model significantly improved discrimination (area under the receiver operating characteristic curve, 0.85; 95% confidence interval, 0.78-0.91).
CONCLUSIONS: Risk models can be used to identify people with a higher than average risk for developing EAC; these individuals might benefit from targeted cancer-prevention strategies.