In this letter to the editor, the authors comment on a paper by Powell et al (2000) on suicide among psychiatric inpatients. They discuss a 10-year review they did of the same phenomenon in Irish psychiatric hospitals. The validity of calculating inpatient suicide rates of the basis of an admissions dominator is questioned. They describe the approach, the person-year method, they used in their study & refer to a paper in which they discuss the imperfections of this approach. (2 refs.)