Abstract: Since 1995 we have provided a bereavement follow-up service but the benefits of this to clinical care have not previously been reported . We reviewed the 1995 results, documentation methods, system design, discussed with reviewers their data collection, interview process and its effect on them and their practice. Next-of-kin of only 99/151 patients ( M87 age 14-88) were contacted 16-149 median 34 days after the death. Sleep disturbance (55/99) were the most common problems. We now give next-of- kin literature including a pamphlet which introduces our service and discusses these problems. Most next-of-kin (77/99) commented favorably on DCCM care, but 16/99 identified inadequate facilities and communication problems as service deficiencies. We have improved facilities (paint, furniture, telephone, drinks machine). We now notify patients general practitioners of every death and bring families back for meetings with staff when necessary. We changed systems of case identification and recording of next-of-kin information. Nurse interviewers (8) identified their needs for improved telephone skills, education about grief and research, debriefing after calls and their need for acknowledgement of participation in this service. Study days now address these issues. Interviewers difficulties with interview content have lead to changes. They suggested peer review of data forms to improve completeness which now occurs. Some found the interview process personally disturbing, but all reported positive changes in their practice with bereaved families.Within two years a bereavement service has led to improvements in facilities, communication, and staff expertise, it is now part of our holistic care