Adaptive cognitive behavioral therapy intervention effects on impulse control related to adolescent digital addiction: a systematic review and meta-analysis

With the acceleration of digitalization, digital addiction among adolescents has become a global public mental health challenge, and deficits in impulse control are regarded as one of its core underlying mechanisms. In this review, the term “CBT-oriented adaptive interventions” is used as an umbrella analytical category. It encompasses interventions that include identifiable cognitive-behavioral or behavioral self-regulation components and incorporate individualized, developmental, modular, contextual, or adjunctive adaptation. However, systematic evidence focusing on impulse-control-related outcomes remains limited. This study aimed to synthesize class-level evidence on the effects of CBT-oriented adaptive interventions on adolescent digital-addiction-related outcomes, with particular focus on impulse-control-related outcomes. This study strictly adhered to the PRISMA guidelines. Relevant literature from five databases, including PubMed, Web of Science, the Cochrane Library, Wanfang, and CNKI, was systematically retrieved from database inception to March 2026. Randomized controlled trials and quasi-experimental studies were screened and included according to the PICOS criteria. For this class-level synthesis, interventions were eligible when they incorporated an identifiable cognitive-behavioral or behavioral self-regulation mechanism together with at least one adaptive feature beyond a fixed standardized protocol, including individualized tailoring, developmental adjustment, modular flexibility, contextual adaptation, or targeted adjunctive components. The Cochrane risk of bias assessment tool was used to evaluate study quality. Meta-analysis, heterogeneity testing, subgroup analyses, and publication bias assessments were conducted using RevMan 5.4 and Stata 18.0. Ultimately, 11 studies involving 1,483 participants were included. Meta-analysis suggested improvements in self-control (SMD = 2.21, 95% CI 0.90 to 3.52, I²=94%), internet addiction (SMD=-0.95, 95% CI -1.58 to -0.32, I²=96%), anxiety (SMD=-0.49, 95% CI -0.87 to -0.11, I²=68%), and internet-related psychological needs (SMD=-1.30, 95% CI -2.56 to -0.04, I²=91%); depression did not reach statistical significance (SMD=-0.69, 95% CI -1.45 to 0.06, I²=86%). Subgroup analyses suggested variability across intervention characteristics. The internet-addiction effect was substantially attenuated after trim-and-fill adjustment (adjusted SMD = 0.071, 95% CI -0.038 to 0.179), with the confidence interval crossing the null effect. Overall, aCBT-oriented interventions showed potential benefits for impulse-control-related capacity and several psychobehavioral outcomes among adolescents with problematic digital use. However, substantial heterogeneity and low certainty of evidence limit confidence in pooled estimates. The apparent benefit for internet addiction severity was not supported after trim-and-fill adjustment, suggesting that publication bias or small-study effects may have influenced the original estimate.