Background: Electroconvulsive therapy ( ECT ) is the effective treatment in psychiatric patients. However, the controversies about the risk and benefit limit the use of this procedure. Some efficacies of ECT may be interesting to use in hospitalized patients such as rapid onset, risk reduction but it may increase the length of stay (LOS ) in previous study.
Objectives: To study the use of ECT in hospitalized psychiatric patients in the aspect of indication , diagnosis and the length of stay.
Material and Method: From August to September 2004 , all 51 cases of psychiatric inpatient medical records were studied . The top five of diagnosis were schizophrenia (49%) , bipolar disorder (23.5%), acute psychosis (7.8% ),depressive disorder (5.9%) and dementia (5.9%). ECT was performed in 22 cases (43.1%).
Results: The average length of stay was 21.5 ±14.5 days . The mean LOS of the ECT group 25.9±15.8 days was not quite longer compared with 17.8 ±12.7 days of the Non ECT group comparison between the ECT group (64 %) and the non ECT group (36 %) of schizophrenia , the LOS of each group was 27.3±16.7and 16.9±8.8 days which was also not significant . LOS of ECT of the bipolar group (20%) and the non ECT bipolar group (80%) was 10.5±4.9 and 23.0±18.3 days which was not statistically significant (p=0.39). There were 3 major indications for ECT such as a severe violence case, suicidal case, refractory case with the mean LOS of 19.9±10.9, 17.5±12.8, 43.3±21.2 and 31.7±8.3 days subsequently. The patients who had the refractory indication had a longer length of stay than other indications.
Conclusion: Overall average length of stay was not longer in the ECT group . The ECT group had a longer length of stay than the non ECT group in the refractory treatment case only . ECT can be used in a psychiatric ward with the efficacy of its indication without prolonged length of stay .

Keywords: Electroconvulsive therapy (ECT), Length of stay (LOS), Psychiatry, Diagnois, Inpatient, Thai