Predictors of Suicidal Behaviors during Hospitalization among Adolescents Admitted Due to Suicidal Behaviors: A 10-Year Retrospective Naturalistic Study.

Objectives: Suicidality during hospitalization is a common phenomenon with potential devastating consequences. We attempted to identify risk factors for in-hospital suicidality in a high risk group of adolescent inpatients hospitalized for suicidal behaviors (SB). Methods: The database of a tertiary adolescent psychiatric ward was screened for patients hospitalized consecutively for SB during 2001-2010. Data on documented demographic, clinical, and behavioral risk factors were collected. Suicidal events during hospitalization were classified according to the Columbia Classification Algorithm of Suicide Assessment. Results: The sample included 122 inpatients (53% female) aged 10-19 (Mean=15.77, Standard Deviation=2.89) years admitted for SB. Thirty-seven youth (30%) exhibited SB during the hospitalization period (the "suicidal group"), ten of which attempted suicide while hospitalized. There were no significant differences in demographic and clinical parameters between the suicidal and the non-suicidal groups. Younger age, history of drug use and a history of non-suicidal self-injury (NSSI) were independent predictors of a SA during hospitalization. A previous SA added significant risk to SA during hospitalization only in the group that had a history of NSSI. Conclusions: A high risk of SB exists among adolescents hospitalized for suicidality. The risk assessment for SA during hospitalization should include age, history of drug use and previous SA combined with a history of NSSI. Future studies should expand the efforts to identify potential risk factors of SB during hospitalization in this unique high-risk group.

Cortisol level and perinatal outcome in pregnant women with posttraumatic stress disorder: a pilot study.

Posttraumatic stress disorder (PTSD) affects 12% of women in the United States and could affect childbearing via behavioral and neuroendocrine mechanisms. This pilot study collected preliminary data about the extent to which the low cortisol profile found in patients with PTSD also occurs in the hormonal context of pregnancy, as well as the association between PTSD and less optimal processes and outcomes of pregnancy. Standardized psychiatric diagnostic telephone interviews, salivary cortisol assays, and medical records review were evaluated in a community sample of 25 women pregnant with their first child. Higher PTSD symptom counts correlated with worse overall perinatal outcomes summarized by an Optimality Index Score (n = 22; r = -.725; P < .001). The women whose symptoms met diagnostic criteria for PTSD or partial PTSD had lower peak basal salivary cortisol concentrations (n = 14; mean = .4584 versus .8123; P = .010). Further research on the effects of PTSD on pregnancy processes and outcomes is warranted. Differences in cortisol levels were consistent with the pattern seen in nonpregnant women with PTSD. This finding suggests that salivary cortisol would be a useful biological measure to include in perinatal research on PTSD and childbearing.