Objective:

To elucidate the epidemiological evolution of pediatric acute poisoning and identify independent clinical and demographic risk factors for clinically significant poisoning (PSS ≥ 2).

Approach:

• Study Design: Retrospective cohort study of 2,345 children (<18 years) with acute poisoning from 2021 to 2024.
• Data Handling: Multiple imputation was used for missing data.
• Statistical Analysis: Multivariable logistic regression identified independent risk factors; model discrimination and calibration were assessed using AUC-ROC and Hosmer-Lemeshow test.

Key Findings:

• The poisoning spectrum shifted from agricultural chemicals to psychotropic drugs (18.5%–31.2%).
• Clinically significant poisoning occurred in 422 patients (18.0%).
• Independent risk factors for moderate-to-severe outcomes included: age >12 years (aOR 2.34), intentional self-harm (aOR 3.12), pre-hospital delay >6 h (aOR 2.78), multiple-agent exposure (aOR 2.15), exposure to agricultural herbicides (aOR 5.45), and exposure to psychotropic drugs (aOR 3.86).
• The model achieved an AUC-ROC of 0.892 (95% CI: 0.864–0.920).
• Clinically significant poisoning was associated with acute organ dysfunctions and required PICU admission (71.1%) and blood purification (44.1%).
• Mortality was 1.2% (28/2,345), occurring exclusively in the PSS ≥ 2 group.

Interpretation:

Limitations:

• The study is retrospective and conducted at a single center, which may limit generalizability.
• Potential biases in data collection and reporting may exist.

Conclusion: