Abstract In bridge engineering, the three-dimensional behavior of a bridge system is usually reduced to the analysis of a T-beam section with a reduced width of deck in relation to center-to-center spacing of stringers, over which the longitudinal normal stresses are assumed to be uniformly distributed, which is termed as effective flange width. It is defined in the AASHTO specifications primarily for concrete slabs and has inherent applicable limitations because of its empirical nature. This paper provides an analytical shear lag model for effective flange width for orthotropic bridge decks, applicable to various materials including Fiber-Reinforced Polymer (FRP) and concrete decks. To verify this solution, a Finite Element (FE) parametric study is conducted on 44 simply-supported FRP deck-on-steel girder bridges. The results from the shear lag model correlate well with the FE results. The accuracy of this model is further verified by close correlations with an existing empirical solution. It is also illustrated that the shear lag model, with the introduction of a reduction factor, can be applied to predict effective flange width for FRP deck-on-steel girder bridges with partial composite action, by favorable comparisons between the analytical and testing results for a T-beam section cut from a one-third scaled bridge model, which consists of an FRP sandwich deck attached to steel girders by mechanical connectors.