association between duration of heartburn and patient characteristics: a quantile regression analysis

aim : to determine the associated factors of reflux duration, using quantile regression model. background : reflux is one of the most prevalent gastrointestinal disorders.  statistical tools are widely used to assess the associated factors on reflux severity and frequency. patients and methods : a door to door questionnaire has been used to evaluate gastrointestinal symptoms including reflux and heartburn between may and december 2006 in tehran. a total of 790 adult patients with gastrointestinal problems were selected using random sampling method. among them 208 persons were found to have heartburn. a quantile regression was employed to assess the associated factors and the results were compared to linear regression and cox semi parametric model. results : heartburn duration had dramatically high negative skewness. quantile regression analysis indicated significant increase in the given decile of heartburn duration associated with a unit increase in given patient characteristic containing sex, pain severity and pain after having some specific food. the results from linear regression and cox model showed poor reliability in comparison to quantile regression. conclusion : the results have demonstrated that using quantile regression leads to better interpretation and richer inference about association between heartburn duration and patient characteristics.