One issue raised was the 14-month time period that we used in order to eliminate the possibility that anniversary-time turbulence might transiently increase self-report of symptoms. Dr. Lister correctly says that normal grief reactions may include some symptoms beyond that time interval. It is, however, important to distinguish between the design of empirical studies and the clinical application of the findings. The clinical application suggested in the article is a diagnostic entity to be added to DSM, as in DSM-V. The diagnosis, as outlined in the end of our article, would be used only when dysfunction was unduly prolonged as judged by a clinician in the individualized context of each patient and the particular cascade of stressor events surrounding the bereavement. For actuarial purposes, in collecting series of cases to establish knowledge about incidence and validity, it is necessary for groups of clinicians to share some definition of "prolonged"; therefore, some generalized times, such as "over 1 year," need to be used.