As shear requirements for prestressed concrete girders have changed, some structures designed using older specifications do not rate well with current methods. However, signs of shear distress have not been observed in these bridge girders and they are often deemed to be in good condition. The primary objective of this research program was to investigate the accuracy of existing shear distribution factors, which are used to estimate bridge system live load effects on individual girders, and provide recommendations on shear distribution to be used in Minnesota with three components: a full-scale laboratory bridge subjected to elastic and inelastic behavior, field testing of bridges, and a numerical parametric study. Results from the components were integrated to develop recommendations for rating prestressed concrete girder bridges. Laboratory bridge inelastic testing indicated shear force redistribution after cracking and before ultimate failure. Use of elastic distribution factors is conservative for shear distribution at ultimate capacity. Elastic laboratory testing was used to validate the finite element modeling technique and study the behavior of a traffic barrier and end diaphragm, which affected shear distribution. Ignoring the effects of the barrier was conservative and ignoring the effects of the end diaphragm was slightly inaccurate by 4 to 6 percent but was warranted for simplicity. Parametric study results indicated that a ratio of longitudinal stiffness to transverse stiffness could be used as a screening tool. If the stiffness ratio was less than 1.5, shear demand from a simple, conservative grillage analysis may be more accurate than shear demand from AASHTO distribution factor methods. Grillage analysis shear demand results due to permit trucks may also be more accurate, regardless of the screening tool ratio.