BACKGROUND: Violence against children affects over one billion children globally. International organisations promote parenting interventions as a main strategy to reduce violence against children. Parenting interventions have therefore been implemented rapidly across the globe. Yet, evidence for their longer-term effects remains unclear. We integrated global evidence to estimate effects over time of parenting interventions to reduce physical and emotional violence against children.

METHODS: In this systematic review and meta-analysis, we searched 26 databases and trial registries (14 non-English: Spanish, Chinese, Farsi, Russian, Thai) and conducted an extensive grey literature search up to August 01, 2022. We included randomised controlled trials (RCTs) of parenting interventions based on social learning theory for parents of children aged 2-10 years, without time or context restrictions. We critically appraised studies using Cochrane's Risk of Bias Tool. Data were synthesised using robust variance estimation meta-analyses. This study is registered with PROSPERO, CRD42019141844.

FINDINGS: We screened 44,411 records and included 346 RCTs. Sixty RCTs reported outcomes on physical or emotional violence. Trials were distributed across 22 countries (22% LMICs). Risk of bias was high for various domains. Outcome data ranged from 0 weeks to 2 years after the intervention, and was largely based on parent self-report. Parenting interventions reduced physical and emotional violent parenting behaviours immediately after the intervention, at 1-6 months follow-up and at 7-24 months follow-up, but effects were smaller over time.

INTERPRETATION: Our findings suggest that parenting interventions can reduce physical and emotional violence against children. Effects are maintained up to 24 months follow-up, but with diminished effect sizes. With global policy interest and imminent importance, research beyond 2 years and how effects can be better sustained over time is urgently needed.