Objective Historically, psychological processes are associated with disorders at the functional end of the gastro-oesophageal reflux disease (GERD) spectrum. However, recent research suggests that psychological symptoms are relevant across the entire GERD spectrum. We aim to investigate whether psychological symptoms are associated with reflux phenotype (True GERD, Borderline GERD, reflux hypersensitivity, functional heartburn) along the GERD spectrum in a cohort of refractory reflux patients.

Design Consecutive adult patients with refractory reflux symptoms underwent standard 24-hour pH-impedance monitoring and completed questionnaires assessing demographic, clinical and psychological information. Bayesian one-way analysis of variance assessed whether psychological variables differed across reflux phenotypes. Next, we applied multinomial and ordinal logistic regressions with clinical, demographic and psychological variables set as independent variables and reflux phenotype as the outcome variable. The complementary machine-learning approach entered all demographic, clinical and psychological variables into models, with reflux phenotype set nominally and ordinally. Cross-validated model performance was used to select the best model.

Results 393 participants (mean (SD) age=48.5 (14.1); 60% female) were included. The Bayesian analyses found no difference in psychological variables across reflux phenotypes. Similarly, age, gender and proton pump inhibitor use were the only significant variables in the multinomial logistic regression and body mass index was significant in both regressions. Machine-learning analyses revealed poorly performing models with high misclassification rates (67–68%) in both models.

Conclusion Psychological symptoms do not differ between nor predict reflux phenotype membership in refractory reflux patients. Findings suggest that psychological symptoms are relevant across the spectrum of GERD, rather than specific to functional oesophageal disorders.