a novel prediction model for all cause emergency department visits in ischemic heart disease

background: ischemic heart disease (ihd) is the main cause of morbidity and mortality worldwide, and a considerable part of these patients attend to emergency departments, which increases the burden to these busy departments. the aim of this study was to develop a prediction model enabling prediction of all cause emergency department (ed) visits in patients with documented coronary stenosis in a derivation set, and then to determine its accuracy in a validation set. methods: in a prospective study at outpatient setting of baqiyatallah hospital, tehran, iran, 502 patients with ihd were followed for 6 months for observing the outcome of ed visits for all causes. they were divided in two random groups of derivation set (n = 335) and validation set (n = 167). in the derivation set, to achieve an all cause ed visits prediction model, a prediction model was reached by entering demographic data, clinical variables, somatic comorbidity (ifudu index), level of anxiety and depression (hospital anxiety depression scale (hads) questionnaire), and angina grade (who rose angina) to a logistic regression. then in the validation set, the sensitivity, specificity, and the accuracy of that model was tested. results: a novel model for prediction of all cause ed visits in ihd patients in six months was presented with gender, anxiety, who angina grade and somatic comorbidity as inputs. sensitivity, specificity, and accuracy of the model were 63.0%, 68.6%, and 67.7%, respectively. conclusions: testing and using the achieved model is suggested to health care providers in other settings.