The objective was to devise and apply a novel modelling approach to combine segmental in situ rat perfusion data and in vitro cell culture data, in order to elucidate the contribution of efflux in drug absorption kinetics. The fluoroquinolone CNV97100 was used as a model P-gp substrate. In situ intestinal perfusion was performed in rat duodenum, jejunum, ileum and colon to measure the influence of P-gp expression on efflux. Inhibition studies of CNV97100 were performed in the presence of verapamil, quinidine, cyclosporin A and p-aminohippuric acid. Absorption/efflux parameters were modelled simultaneously, using data from both in situ studies as well as in vitro studies. The maximal efflux velocity was modelled as a baseline value, corrected for each segment based on the expression level. CNV97100 passive diffusional permeability (P(diff)) and its affinity for the efflux carrier (K(m)) were assumed to be the same in all segments. The results indicate the new approach to combine in situ data and in vitro data succeed in yielding a unified, quantitative model for absorption/efflux. The model incorporated a quantitative relationship between P-gp expression level and the efflux functionality, both across in situ and in vitro systems, as well across different intestinal segments in the in situ studies. Permeability values decreased from duodenum to ileum in accordance with the increasing P-gp expression levels in rat intestine. The developed model reflects a strong correlation between in vitro and in situ results, including intrinsic differences in surface area. The successful application of a model approach to combine absorption data from two different experimental systems holds promise for future efforts to predict absorption results from one system to a second system.