Objective: Childhood traumatic experiences increase the likelihood of psychopathological findings in adults. It is important to inquire childhood traumatic experiences in all adult psychiatric patients. However, the best way of this inquiry is not clear. The purpose of this study is to investigate whether a self-rating instrument or a semi-structured interview is better for screening childhood traumatic experiences in psychiatric inpatients. We also aim to investigate the relation between self-mutilative behavior, suicide attempt and childhood trauma accessed with these two instruments so that we can compare the external validity of these methods. Methods: We administered Childhood Trauma Questionnaire (CTQ), a self-rating scale, to 91 psychiatric inpatients and then conducted a semi structured interview (Childhood Abuse and Neglect Inventory, CANI) with these patients. Sociodemograhic data, history of suicide attempt and self-mutilation were also gathered. Findings: Mean sexual abuse, physical abuse and emotional abuse and neglect subscale scores and total score and standard means on CTQ were 7.5 +/- 3.9, 30.6 +/- 12.9, 43.9 +/- 16.8 and 81.9 +/- 29.3, respectively. The rates of childhood trauma on CANI were 36.3% for physical abuse, 18.7% for emotional abuse, 24.2% for physical neglect, 37.4% for emotional neglect, 22.0% for sexual abuse, 6.6% for ensest. Forty-four percent of the participants reported a history of serious suicide and 52.7% of them reported self-mutilative behavior. When we grouped the subjects according to their abuse histories on CANI and compared CTQ scores of these groups, the subjects who reported histories of physical and abuse, emotional abuse or emotional neglect on CANI had higher total and subscale scores on CTQ. However, CTQ total and subscale scores did not differentiate the subjects who reported sexual abuse on CANI from those who did not have such a history. As CTQ did not have subscales for physical neglect and ensest, we did not make comparison with CANI. Histories of suicide attempt and self-mutilation were correlated with physical abuse, emotional abuse and emotional neglect on CANI; however, they were not correlated with any kind of childhood abuse or neglect on CTQ. When logistic regression analysis was applied to CANI data, suicide attempt was found to be correlated with emotional abuse and self-mutilation was found to be correlated with physical abuse. Discussion: When childhood trauma histories of psychiatric inpatients are assessed with a semistructured interview (CANI), we can get more valid results compared to a self-rating instrument results. The key childhood trauma related to suicide attempt is emotional abuse and the key trauma related to self-mutilation is physiccal abuse.