Han Wang, Qiuling Hu, Jian Li, Kaining Chen, Xiaoxuan Qi, Rongxin Zhang, Xi Wang, Yongkun Wang, Yiwen Li
Only a small proportion of reported exposure-outcome associations in pediatric allergic diseases are supported by strong evidence. Early-life and environmental exposures, including perinatal factors and pesticide exposure, may represent important risk domains; however, most associations lack sufficient evidence to support causal inference. These findings highlight the need for higher-quality evidence and more rigorous evaluation of environmental risk factors in pediatric allergy research.
BACKGROUND: Pediatric allergic diseases are influenced by a wide range of environmental, perinatal, and developmental exposures; however, the strength and credibility of reported associations remain unclear.
OBJECTIVE: We conducted an umbrella review to systematically evaluate and grade the strength of evidence for exposure-outcome associations in pediatric allergic diseases.
METHODS: We systematically searched for systematic reviews and meta-analyses investigating exposures associated with pediatric allergic diseases. A total of 176 exposure-outcome associations were extracted and re-evaluated using standardized evidence grading criteria, incorporating statistical significance, heterogeneity, sample size, and potential bias. Associations were classified into strong, highly suggestive, suggestive, and weak or non-suggestive evidence.
RESULTS: Sixty systematic reviews and meta-analyses comprising 176 exposure-outcome associations were included. Only three associations were supported by strong evidence: neonatal jaundice, pesticide exposure, and postnatal maternal depression. Six associations were classified as highly suggestive (three from binary outcomes and three from continuous outcomes after applying a pragmatic sample size criterion), and eight as suggestive, while the remaining 159 associations were rated as weak or non-suggestive evidence. Overall, most associations showed limited robustness, with substantial heterogeneity and evidence of potential bias across studies.
CONCLUSION: Only a small proportion of reported exposure-outcome associations in pediatric allergic diseases are supported by strong evidence. Early-life and environmental exposures, including perinatal factors and pesticide exposure, may represent important risk domains; however, most associations lack sufficient evidence to support causal inference. These findings highlight the need for higher-quality evidence and more rigorous evaluation of environmental risk factors in pediatric allergy research.
IMPACT: This umbrella review systematically evaluates the strength of evidence for associations between multiple exposures and pediatric allergic diseases. Only a small proportion of associations are supported by strong or highly suggestive evidence, while most associations show weak or non-suggestive evidence. These findings highlight substantial uncertainty in the current literature and emphasize the need for more robust evidence before drawing causal or clinical conclusions.