TY - JOUR
T1 - Risk of first general hospital admissions for self-harm among people with personality disorder
T2 - a retrospective cohort study
AU - Rush, Kelsey
AU - Fok, Marcella Lei Yee
AU - Moran, Paul A
AU - Dorrington, Sarah
AU - Shetty, Hitesh
AU - Stewart, Robert
AU - Chang, Chin Kuo
PY - 2021/10/1
Y1 - 2021/10/1
N2 - For personality disorder (PD), it is unknown whether additional psychiatric conditions increases their risk of self-harm. We thus performed a retrospective cohort study, using data from secondary mental healthcare in South London, linked to Hospital Episodes Statistics in England to identify self-harm admissions. Adults with diagnosed PD, followed up in a 6-year period, were involved (N=7,475). Cox regression was used to model the risk of self-harm admissions, with comorbid depression, substance use disorders (SUD), severe mental illness (SMI), anorexia nervosa (AN) and bulimia nervosa (BN) as primary exposures, with sociodemographics considered as confounders. In multivariable analyses, comorbid SUD (adjusted HR=1.66; 95%CI: 1.45, 1.90), depressive disorder (1.25; 1.09, 1.44), AN (1.63; 1.10, 2.39) and BN (1.65; 1.09, 2.51) were positively associated with increased risks of hospitalisation due to self-harm. However, negative association was found for comorbid SMI. The findings highlight the importance of assessing and treating comorbid psychiatric disorders in PD.
AB - For personality disorder (PD), it is unknown whether additional psychiatric conditions increases their risk of self-harm. We thus performed a retrospective cohort study, using data from secondary mental healthcare in South London, linked to Hospital Episodes Statistics in England to identify self-harm admissions. Adults with diagnosed PD, followed up in a 6-year period, were involved (N=7,475). Cox regression was used to model the risk of self-harm admissions, with comorbid depression, substance use disorders (SUD), severe mental illness (SMI), anorexia nervosa (AN) and bulimia nervosa (BN) as primary exposures, with sociodemographics considered as confounders. In multivariable analyses, comorbid SUD (adjusted HR=1.66; 95%CI: 1.45, 1.90), depressive disorder (1.25; 1.09, 1.44), AN (1.63; 1.10, 2.39) and BN (1.65; 1.09, 2.51) were positively associated with increased risks of hospitalisation due to self-harm. However, negative association was found for comorbid SMI. The findings highlight the importance of assessing and treating comorbid psychiatric disorders in PD.
U2 - 10.1521/pedi_2020_34_489
DO - 10.1521/pedi_2020_34_489
M3 - Article (Academic Journal)
C2 - 33107805
SN - 0885-579X
JO - Journal of Personality Disorders
JF - Journal of Personality Disorders
ER -