In seismic design practice, hollow concrete columns offer an efficient alternative especially for tall bridge columns. The confinement in critical regions of these columns is not well understood and the confinement models developed for solid concrete sections are used to characterize these columns. This report identifies several shortcomings in current practice associated with hollow column design. Using experimental and analytical means, it is shown that the confined concrete behavior in hollow concrete columns is different from that established for solid sections. Recommendations to improve analysis and design of hollow columns are presented with experimental verification. Modification to alter the confinement models developed for solid sections and a design example are also provided.