Objective This study aims to explore the association between non-suicidal self-injury(NSSI)and suicide attempts(SA)in adolescents and the mediating effect of cognitive flexibility.Methods A total of 218 depression patients with NSSI who met the Diagnostic and Statistical Manual of Mental Disorders,5th Edition(DSM-5)diagnostic criteria for NSSI were enrolled.Patients were divided into SA group(n=105)and non-SA group(n=113)according to the presence or absence of SA in the last one year.The adolescent non-suicidal self-injury assessment questionnaire(ANSAQ)and the Wisconsin card sorting tests(WCST)was used to assess the frequency of NSSI and cognitive flexibility,respectively.A mediation model was constructed to conduct path analysis,and the product distribution method was utilized to test the mediation effect.Results The difference between SA group and non-SA group in NSSI(20.1±10.7 vs.14.7±9.1)and WCST scores[correct responses percentage(67.3％±14.2％vs.72.9％±12.2％),error responses(39.8±20.3 vs.31.6±17.9),perseverative response(6.7±3.8 vs.5.3±2.9),and non-perseverative errors(37.6±21.0 vs.28.9±18.1)]were significant(P<0.05).Dichotomous logistic regression analysis showed that the frequency of NSSI(OR=1.051,95％CI:1.021-1.082)and the score of perseverative response(OR=1.100,95％CI:1.008-1.199)were significantly associated with suicidal behavior among adolescents with NSSI(P<0.05).Moreover,perseverative response partially mediated the association between NSSI and SA(95％CI of Za×Zb:0.0003-0.0168).Conclusion High NSSI and low cognitive flexibility are risk factors for suicide attempts in NSSI adolescents and NSSI may also affect SA indirectly by lowering cognitive flexibility.

Objective This study aims to explore the association between non-suicidal self-injury(NSSI)and suicide attempts(SA)in adolescents and the mediating effect of cognitive flexibility.Methods A total of 218 depression patients with NSSI who met the Diagnostic and Statistical Manual of Mental Disorders,5th Edition(DSM-5)diagnostic criteria for NSSI were enrolled.Patients were divided into SA group(n=105)and non-SA group(n=113)according to the presence or absence of SA in the last one year.The adolescent non-suicidal self-injury assessment questionnaire(ANSAQ)and the Wisconsin card sorting tests(WCST)was used to assess the frequency of NSSI and cognitive flexibility,respectively.A mediation model was constructed to conduct path analysis,and the product distribution method was utilized to test the mediation effect.Results The difference between SA group and non-SA group in NSSI(20.1±10.7 vs.14.7±9.1)and WCST scores[correct responses percentage(67.3％±14.2％vs.72.9％±12.2％),error responses(39.8±20.3 vs.31.6±17.9),perseverative response(6.7±3.8 vs.5.3±2.9),and non-perseverative errors(37.6±21.0 vs.28.9±18.1)]were significant(P<0.05).Dichotomous logistic regression analysis showed that the frequency of NSSI(OR=1.051,95％CI:1.021-1.082)and the score of perseverative response(OR=1.100,95％CI:1.008-1.199)were significantly associated with suicidal behavior among adolescents with NSSI(P<0.05).Moreover,perseverative response partially mediated the association between NSSI and SA(95％CI of Za×Zb:0.0003-0.0168).Conclusion High NSSI and low cognitive flexibility are risk factors for suicide attempts in NSSI adolescents and NSSI may also affect SA indirectly by lowering cognitive flexibility.