Identifying individuals at risk of esophageal adenocarcinoma: challenges, existing tools and future steps

Abstract

Purpose of review This review aims to discuss some of the clinical and epidemiological challenges of risk prediction models; summarize the evidence supporting existing models; and highlight the translational requirements. Recent findings A variety of risk prediction models exist to identify prevalent Barrett's esophagus or predict future esophageal adenocarcinoma. External validation studies have investigated performance of these models in a variety of settings. These models appear to be more predictive than use of symptoms alone, but the added complexity means that implementation challenges may require investigation. Summary Risk prediction models could be useful for identifying individuals at an increased risk of esophageal adenocarcinoma, which could assist screening decisions. However, risk prediction models must be implemented with care. Implementation science to assist the translation of existing models into practice may be an important next step.