Table 1.
Characteristics of the included reviews.
| Author (year) | Population | Comparison | Methods | No. studies included | Results | Conclusion |
| Bahji et al. (2021) [20] | Children and adolescents (10–19 years) | Psychosocial interventions for the treatment of self-harm and suicidal behaviour | Search: PubMed, MedLine, PsycINFO and Embase until 2020. Inclusion criteria: RCT, comparison of psychotherapies for suicide and self-harm prevention. | 44 RCTs (5406 participants) | Cognitive behavioural therapy did not show a significant reduction in suicidal ideation (mean deviation (MD) –0.21; 95% confidence interval (CI): –1.63 to 1.21) or self-harm (MD 0.6; 95% CI: 0.31 to 1.19). | Most psychotherapies were reasonably well tolerated and some psychotherapies indicated efficacy for particular measures of self-harm or suicidality. |
| Dialectical behavioural therapies were associated with reductions in self-harm (odds ratio (OR) 0.28; 95% CI: 0.12 to 0.64) and suicidal ideation (weigthed mean difference (WMD) −0.71; 95% CI: –1.19 to –0.23). Mentalization-based therapies were associated with decreases in self-harm (OR 0.38; 95% CI: 0.15 to 0.97) and suicidal ideation (WMD –1.22; 95% CI: –2.18 to –0. 26). | ||||||
| Witt et al. (2021) [21] | Children and adolescents (18 years old) with a history of self-harm | Psychosocial or pharmacological interventions for self-harm | Search: Cochrane Specialized Register of Common Mental Disorders, Cochrane Library, Central, Cochrane Database of Systematic Reviews, MEDLINE Ovid, Embase Ovid and PsycINFO Ovid. | 17 RCTs (2280 participants) | No significant differences were found between CBT and other psychological intervention in the repetition of self-harm (OR 0.93; 95% CI: 0.12 to 7.24), nor between the psychological intervention and pharmacotherapy in suicidal ideation (OR 0.26; 95% CI: 0.07 to 0.98) in minors with depression. | Younger people have cognitive, behavioural and emotional characteristics different from adults, and these should be taken into account in order to make age-specific innovations in the design and content of interventions. |
| Regarding DBT compared to usual treatment, a decrease was found in the repetition of self-harm (OR 0.46; 95% CI: 0.26 to 0.82) and suicidal ideation (standardized mean difference (SMD): –0.43; 95% CI: –0.68 to –0.18). | Studies consider that the most important predictor is a previous event of suicidal behaviour. | |||||
| TBM was associated with a non-significant reduction in repetition of self-harm (OR 0.70; 95% CI: 0.06 to 8.46) compared to usual treatment. | ||||||
| Cox et al. (2014) [22] | Minors (18 years old) diagnosed with depressive disorder | Psychological therapies and antidepressant medication | Search: Cochrane, MedLine, Embase and PsycINFO until 2012. Inclusion criteria: RCT with minors diagnosed with depression. | 11 RCTs (1307 participants) | Cognitive behavioural therapy did not show significant differences between groups in suicidal ideation (MD 0.60; 95% CI: –2.25 to 3.45), at 6–9 months (MD 1.78; 95% CI: –2.29 to 5.85) or at one year (MD 0.90; 95% CI: –1.37 to 3.17). In two trials, psychotherapy decreased suicidal ideation compared to medication (OR 0.26; 95% CI: 0.09 to 0.72) (MD –3.12; 95% CI: –5.91 to –0.33), this effect remained at 6–9 months (OR 0.26; 95% CI: 0.07 to 0.98), but not at one year (p 0.05). | Psychological therapy may be associated with less suicidal ideation. A combination of treatments could protect against suicidality. There is great variability within the data. Additional data is needed. |
| There were no differences between combined therapy and psychotherapy (OR 1.68; 95% CI: 0.53 to 5.34), nor between combined therapy and psychotherapy plus placebo (p 0.05). No results were shown on self-harm. | ||||||
| Mann et al. (2021) [23] | Population with suicidal behaviour or ideation | Scalable, evidence-based suicide prevention strategies | Search: PubMed and Google Scholar between 2005–2019. Inclusion criteria: RCTs and studies on limitation of access to lethal means, educational approaches and antidepressants | 97 RCTs and 30 epidemiological studies | Cognitive behavioural therapy decreased the risk of suicidal behaviour in adolescents with depression. CBT reduced suicide attempts in patients presenting to the emergency department after a suicide attempt and in substance use disorders compared to TAU. No results were reported on self-harm. | CBT is a proven scalable strategy for suicide prevention. |
| Xiang et al. (2022) [24] | Minors (18 years old) diagnosed with depressive disorder | Combined therapy | Search in PubMed, Embase, PsycINFO, WOS, CINAHL, LiLACS and ProQuest until 2020. Inclusion criteria: RCTs on pharmacotherapy and psychotherapy. | 14 RCTs (1325 participants) | Treatment with fluoxetine plus CBT did not have a significant effect for suicidality (suicidal ideation or attempt/behaviour) (OR 1.17; 95% CI: 0.67 to 2.06). Children under 25 years of age treated with antidepressants are more likely to develop suicidal thoughts than adults, especially when treated with venlafaxine. No results were reported on self-harm. In those 16 years of age, the dropout rate was higher. | Despite the limited evidence, therapies combined with CBT may be superior to other active treatment options, although other psychotherapies were not included in the study. Combined therapies have poorly been studied in this age group. |
| Ma et al. (2014) [25] | Minors with a diagnosis of depressive disorder | Contemporary interventions for depressive disorder in children and adolescents | Search: Cochrane, CINAHL, EMBASE, LiLACS, MedLine, PSYCINFO and PSYNDEX until 2014. Inclusion criteria: RCT, antidepressants, CBT, fluoxetine combined with CBT and placebo, acute treatment of major depressive disorder. | 21 RCTs (4969 participants) | CBT decreased suicidal ideation compared to fluoxetine (OR 1.88; 95% CI: 1.41 to 2.50). The cumulative odds of the combination therapy of fluoxetine plus CBT was the most effective treatment (95.2%), but it was worse tolerated than placebo and its acceptability was inferior to the use of other antidepressants. No significant differences were found between fluoxetine and the combination therapy (OR 0.77; 95% CI: 0.59 to 1.00). No results of self-harm were reported. | CBT showed lower efficacy, acceptability and safety than the rest and was considered relatively less useful as a first-line treatment. The combination therapy showed the greatest efficacy, although safety remained a major concern. Further research is needed. |
| Zhou et al. (2020) [26] | Minors with a diagnosis of depressive disorder | Interventions and treatments available for depressive disorder in children and adolescents | Search: PubMed, Embase, LiLACS, CINAHL, PsycINFO, WOS, Cochrane and ProQuest until 2019. Inclusion criteria: RCTs, acute treatment, 18 years old with depression. | 71 RCTs (9510 participants) | The combined therapy of fluoxetine plus CBT was more effective for the prevention of suicidality (OR 0.13; 95% CI: 0.00 to 0.59). Venlafaxine was associated with a significantly higher risk of suicidal ideation and behaviour (OR 8.31; 95% CI: 1.92 to 343.17). No results of self-harm were reported. | Psychotherapies were superior to control group, but more research is required as these therapies were adaptations of treatments developed for adults. |
| Most RCTs of psychotherapy were assessed as having a high risk of bias and had lower dropout rates and baseline severity scores than RCTs of pharmacotherapy. | ||||||
| Frías et al. (2015) [27] | Patients (parents or minors) with a diagnosis of bipolar disorder (type I, II, non-specific) | Psychosocial treatments in pediatric bipolar disorder | Search: PsycINFO and PubMed until 2014. Inclusion criteria: Minors 6–19 years old with a diagnosis of bipolar disorder. | 4 case studies, 9 open trials, 7 RCTs, 8 systematic reviews and 5 theoretical trials (606 patients) | At 2-month follow-up, CBT showed a reduction in depressive symptoms, including suicidal behaviour. DBT showed a 3-fold greater reduction in suicidal ideation compared to TAU. No results of self-harm were reported. | Research on CBT for suicide prevention is scarce and evidence insufficient. DBT seems promising in this area and findings need to be replicated in larger samples. |
| Ougrin et al. (2015) [28] | Minors 18 years of age with a history of self-harm | Pharmacological, social and psychological interventions for self-harm in adolescents | Search: Cochrane, MedLine, PubMed, PsycINFO and EMBASE until 2014. Inclusion criteria: RCT, comparison with TAU or placebo, with self-harm | 19 RCTs (2176 participants) | There were no significant differences in suicide attempts (risk difference: –0.03; 95% CI: –0.09 to 0.03) between therapies. | Improving adherence to treatment seems key to reducing the risk of self-harm and suicide. Although the pooled effectiveness of the interventions versus TAU was significant, there is a high heterogeneity across studies. Further research is therefore needed. |
| The pooled risk difference for any self-harm was –0.07 (95% CI: –0.01 to –0.13). The NNT was 21 (95% CI: 11.2 to 98.5) at 10 months. For non-suicidal self-harm there was a non-significant risk reduction (risk difference: –0.1; 95% CI: –0.21 to 0.00) compared to TAU. Furthermore, studies with a strong family component (risk difference: –0.14; 95% CI: –0.27 to –0.02) or with multiple sessions (risk difference: –0.09; 95% CI: –0.017 to 0.00) were associated with a significant reduction in self-harm. DBT showed a significant reduction in the risk of self-harm and suicidal ideation. |
CBT, cognitive behavioural therapy; DBT, dialectical behavioural therapy; TAU, usual treatment; RCT, randomized clinical trial; DM, mean difference; SMD, standardized mean difference; NNT, Number Needed to Treat; CI, confidence interval; OR, odds ratio. Suicidality includes suicidal ideation or attempt/behaviour.