Purpose: The aim of this study was to develop a shortened version of the Oral Health Impact Profile (OHIP) appropriate for use in edentulous patients and to evaluate its measurement properties. Materials and Methods: Data were collected from the Ontario Study of Older Adults and a longitudinal clinical trial of implant-retained prostheses undertaken in Newcastle Dental Hospital, UK. All subjects completed an OHIP at baseline, and UK subjects also completed an OHIP posttreatment. Using an item impact reduction method, a shortened version of the OHIP (called OHIP-EDENT) was derived from both datasets. Discriminant validity and responsiveness properties of this modified version were compared with OHIP-14 and OHIP-49. Results: Using an item impact method of reducing the 49 OHIP items produced very similar subsets in both Canadian and British populations; the modified version had little overlap with the current short version (OHIP-14). Discriminant validity properties of OHIP-EDENT were similar to OHIP-14 and OHIP-49. Using effect sizes to assess sensitivity to change, OHIP-EDENT exhibited less susceptibility to floor effects than OHIP-14 and appeared to measure change as effectively as OHIP-49. Conclusion: The modified shortened version of the OHIP derived in this study has measurement properties comparable with the full 49-item version. This modified shortened version may be more appropriate for use in edentulous patients than the current short version (OHIP-14). Int J Prosthodont 2002;15:446–450.