Background: The attitude towards lactation is one of the best predictors of breastfeeding. Iowa Infant Feeding Attitudes Scale (IIFAS) is used to measure the attitude toward lactation. IIFAS is a valid and reliable tool but factorial structure of this tool was reported various in different studies. The aim of this study is to assess factorial structure of IIFAS. Materials and Methods: An extensive search was done in databases of databases of Medline, EMBASE, Web of Science, Scopus, Cochrane Library, and CINAHL until May 2018. Two independent researchers screened articles and in the next step, full texts of probably relevant articles were read and summarized. The quality of studies was performed by COSMIN checklist. The following keywords were used: (Iowa Infant Feeding Attitude Scale OR IIFAS) AND (Factor Analysis OR exploratory factor analysis OR confirmatory factor analysis OR Validity OR psychometric). Results: Six studies were assessed in systematic review. In Spanish version, single- factor solutions with 9 items in sample of 1,294 pregnancy women was tested and showed a satisfactory fit to the data. In Japanese version, authors provided single-factor- model with 16 items. Factors loading were ranged from -0.06 to 0.68. Arabic version, EFA identified 6 factors with eigenvalues more than 1 explained 61% of total variance. However, scree plot suggested unidimensional structure. In Chinese version, EFA extracted four factors and labeled Favorable to breastfeeding , Favorable to formula-feeding , Convenience and Sociological influences . In Canadian and Singapore version, the most sense model based on EFA was a three –factors model and labeled Favorable to breast feeding , Convenience and Favorable to formula feeding . Conclusion: Four-factor model and three- factor model can be used in clinical practices and research. There is a need to further test single-factor model.