BACKGROUND: Psychotic disorders produce important morbidity and disability in children and adolescents. There have been few relevant treatment trials, encouraging assessment of research aimed at testing efficacy and safety of antipsychotics for juveniles. We aimed to compare the short- and long-term efficacy and safety of antipsychotics to treat psychotic disorders among children and adolescents.
METHODS: Four major bibliographic databases (PubMed, MEDLINE, PsycINFO, and EMBASE) were searched for clinical trials of antipsychotics in children or adolescents, from database inception to May 2021. We searched for clinical trials comparing antipsychotics with control conditions for juvenile psychosis based on blinded review by 2 independent investigators (C.S.Y. and M.L.). We adhered to the Preferred Reporting Items for Systematic Reviews and Meta-analyses and applied the Cochrane risk-of-bias tool to appraise study quality. One reviewer (A.B.) performed data abstraction which was confirmed by 2 independent, blinded reviewers (C.S.Y. and M.L.). Primary outcomes were scores rating psychosis symptoms and dichotomized retention in treatment protocols versus dropouts because of adverse events. Effect sizes were pooled using frequentist random-effects network meta-analysis modeling to generate summary rate ratios (RRs) and Cohen d standardized mean differences.
RESULTS: Systematic searching generated 1330 unique records. Of these, short-term (n = 15, for 6 [3-12] weeks) and long-term (n = 10, for 12 [6-60] months) treatment trials involved 2208 (39.2% females; median age, 15.3 years), and 1366 subjects (35.0% females; median age, 15.6 years), respectively. Short-term reduction of psychosis scores ranked clozapine (d = -1.35; 95% confidence interval [CI], -1.97 to -0.73]), molindone (-1.22; 95% CI, -1.68 to -0.75), olanzapine (-1.12; 95% CI, -1.44 to -0.81), and risperidone (-0.93; 95% CI, -1.22 to -0.63) as the most effective agents. In longer-term treatment, only lurasidone was effective. Clozapine (RR, 12.8) and haloperidol (RR, 5.15) led to more all-cause and adverse event-related dropouts. There were few trials/drug (1 each for aripiprazole, asenapine, lurasidone, molindone, paliperidone, and ziprasidone, short term; aripiprazole, clozapine, haloperidol, lurasidone, and molindone, long-term). Heterogeneity and inconsistency were high, especially in long-term trials, without evidence of publication bias.
CONCLUSIONS: Some antipsychotics were effective and tolerated short term, but longer-term evidence was very limited. The overall paucity of trials and of adequate controls indicates that more well-designed randomized controlled trials are required for adequate assessment of antipsychotic drug treatment for juveniles.
SYSTEMATIC REVIEW REGISTRATION: PROSPERO CRD42021232937.
