The objective of the study was to identify clinical characteristics that enable non-cardiac chest pain to be differentiated from coronary artery disease. An observational case-control study was carried out in 40 patients with non-cardiac chest pain and a control group of 40 patients with coronary artery disease. Sociodemographic, medical and psychological variables were assessed. There was no difference in personality except in "emotional control", which was less in patients with non-cardiac chest pain. These patients had greater alexithymia and more frequently used coping strategies, such as religion and seeking medical help. Multivariate analysis showed that a predictive model comprising four variables (i.e., alexithymia, quality of life, and coping based on religion and seeking medical help) had a sensitivity of 85.4% and a specificity of 80.0%. This predictive model could be used as a screening test to discriminate between the two conditions.