Dear Editor-in-Chief

I read with great interest the article written by Heidi F. A. Moossdorff-Steinhauser et al. [1]. The aim of this study was to determine the prevalence of self-reported UI, level of experience of bother, and beliefs to gain a greater understanding of help-seeking behavior in adult pregnant women [1]. Although this study is an interesting and valuable research work, in order to reduce misinterpretation, a number of methodological points should be considered when interpreting the results, as follows:

  1. 1.

    In this study, considering that the questionnaire of help-seeking behavior is researcher-made and has no dimensions, it was much better to use exploratory factor analysis (EFA) to determine the dimensions and the questions were designed based on the Likert scale. In EFA, the researcher has no previous basis for categorizing questions into different dimensions. In fact, the questions are entered into the analysis and then the statistical analysis software itself decides the dimensions in which the questions should be placed. In other words, the application of EFA is greater in new research that does not have a preconceived theory [2, 3].

  2. 2.

    The second point in the present study is the lack of attention to collinearity between questions. Collinearity, indeed, correlation between predictor variables (or independent variables), such that they express a linear relationship in a regression model when predictor variables in the same regression model are correlated, cannot independently predict the value of the dependent variable [4, 5].

  3. 3.

    The third point that needs to be mentioned is not using univariate and multivariate regression models to control confounding variables. In regression models, in order to eliminate potential confounding variables, the variables with p value ≤0.20 in the univariate regression model were introduced into the multivariate regression model, simultaneously. However, in the present study, researchers have studied the relationship between variables as a univariate and have not paid attention to the control of confounding variables by the multivariate regression models [6].