TY - JOUR
T1 - Risk factors for suicide attempt during outpatient care in adolescents with severe and complex depression
AU - van Velzen, Laura S
AU - Toenders, Yara J
AU - Kottaram, Akhil
AU - Youzchalveen, Belinsha
AU - Pilkington, Vita
AU - Cotton, Sue M
AU - Brooker, Abi
AU - McKechnie, Ben
AU - Rice, Simon
AU - Schmaal, Lianne
N1 - Publisher Copyright:
© 2022 Hogrefe Publishing.
PY - 2022/5/12
Y1 - 2022/5/12
N2 -
Background: Young people receiving tertiary mental health care are at elevated risk for suicidal behavior, and understanding which individuals are at increased risk during care is important for treatment and suicide prevention.
Aim: We aimed to retrospectively identify risk factors for attempted suicide during outpatient care and predict which young people did or did not attempt during care.
Method: Penalized logistic regression analysis was performed in a small high-risk sample of 84 young people receiving care at Orygen's Youth Mood Clinic (age: 14-25 years, 51% female) to predict suicide attempt during care (N = 16).
Results: Prediction of suicide attempt during care was only moderately accurate (Area Under the Receiver Operating Curve range 0.71; sensitivity 0.57) using a combination of sociodemographic, psychosocial, and clinical variables. The features that best discriminated both groups included suicidal ideation during care, history of suicide attempt prior to care, changes in appetite reported on the PHQ-9, history of parental separation, and parental mental illness.
Limitation: Replication of findings in an independent validation sample is needed.
Conclusion: While prediction of suicide attempt during care was only moderately successful, we were able to identify individual risk factors for suicidal behavior during care in a high-risk sample.
AB -
Background: Young people receiving tertiary mental health care are at elevated risk for suicidal behavior, and understanding which individuals are at increased risk during care is important for treatment and suicide prevention.
Aim: We aimed to retrospectively identify risk factors for attempted suicide during outpatient care and predict which young people did or did not attempt during care.
Method: Penalized logistic regression analysis was performed in a small high-risk sample of 84 young people receiving care at Orygen's Youth Mood Clinic (age: 14-25 years, 51% female) to predict suicide attempt during care (N = 16).
Results: Prediction of suicide attempt during care was only moderately accurate (Area Under the Receiver Operating Curve range 0.71; sensitivity 0.57) using a combination of sociodemographic, psychosocial, and clinical variables. The features that best discriminated both groups included suicidal ideation during care, history of suicide attempt prior to care, changes in appetite reported on the PHQ-9, history of parental separation, and parental mental illness.
Limitation: Replication of findings in an independent validation sample is needed.
Conclusion: While prediction of suicide attempt during care was only moderately successful, we were able to identify individual risk factors for suicidal behavior during care in a high-risk sample.
UR - http://www.scopus.com/inward/record.url?scp=85130598926&partnerID=8YFLogxK
U2 - 10.1027/0227-5910/a000860
DO - 10.1027/0227-5910/a000860
M3 - Article
C2 - 35548884
SN - 0227-5910
VL - 44
SP - 232
EP - 239
JO - Crisis
JF - Crisis
IS - 3
ER -