The prediction of the seismic vulnerability of bridges is a challenging task for road managers which need to include structural risk prioritisation in their Bridge Management Systems. The large number of bridges designed without anti-seismic requirements and the uncertainties resulting from the lack of knowledge data (e.g. design documents and blueprints), strongly affect the feasibility of this process. A simplified approach for performing fragility analysis considering the epistemic uncertainty associated with an incomplete knowledge level is described. Uncertainties are considered through the statistical generation of a population of index-bridges, which are analysed utilizing a simplified modelling approach and the capacity method spectrum method applied with real ground-motion spectra. The proposed approach is applied on a case-study simply supported-girder RC bridge of the Basilicata provincial road network. The modelling and the influence of the epistemic uncertainty on the fragility are discussed for two knowledge levels corresponding to different data collection strategies. Finally, a validation of the proposed approach is carried out through a comparison with a refined modelling approach and non-linear dynamic analyses.