Objective This study aims to identify patients with oesophageal cancer's level of distress, type of problems, and wish for referral prior to treatment. To identify the clinical relevance of patients with oesophageal cancer's level of distress and type of problems, we build models to predict elevated distress, wish for referral, and overall survival. Methods We implemented the Distress Thermometer and Problem List in daily clinical practice. A score of ≥5 on the Distress Thermometer reflected elevated distress. We first created an initial model including predictors based on the literature. We then added predictors to the initial model to create an extended model based on the sample data. We used the ‘least absolute shrinkage and selection operator’ to define our final model. Results We obtained data from 187 patients (47.9%, of 390 eligible patients with oesophageal cancer) which were similar to non‐respondents in their demographic and clinical characteristics. One‐hundred thirteen (60%) patients reported elevated distress. The five most frequently reported problems were as follows: eating, tension, weight change, fatigue, and pain. Most patients did not have a wish for referral. Predictors for elevated distress were as follows: being female, total number of practical, emotional, and physical problems, pain, and fatigue. For referral, we identified age, the total number of emotional problems, the level of distress, and fear. The level of distress added prognostic information in a model to predict overall survival. Conclusions Patients with oesophageal cancer report elevated distress and a myriad of problems yet do not have an explicit wish for referral prior to receiving their medical treatment plan. Copyright © 2016 John Wiley & Sons, Ltd.