Review articleExposure to traffic-related air pollution and risk of development of childhood asthma: A systematic review and meta-analysis

- Whether children's exposure to traffic-related air pollution contributes to their development of asthma is an unresolved question.
- We conducted a systematic review and a meta-analysis of observational studies published until 8 September 2016 investigating this question.
- Forty-one studies met our eligibility criteria and there was an epidemic growth in publications after 2014.
- Overall, there was notable variability in asthma definitions, TRAP exposure assessment and confounder adjustment.
- The meta-analysis showed positive and statistically significant associations between asthma onset and the exposure to BC, NO2, PM2.5 and PM10, with the least heterogeneity detected in the BC and PM analyses and the most detected in the NO2 and NOx analyses.

Background and objectiveThe question of whether children's exposure to traffic-related air pollution (TRAP) contributes to their development of asthma is unresolved. We conducted a systematic review and performed meta-analyses to analyze the association between TRAP and asthma development in childhood.Data sourcesWe systematically reviewed epidemiological studies published until 8 September 2016 and available in the Embase, Ovid MEDLINE (R), and Transport databases.Study eligibility criteria, participants, and interventionsWe included studies that examined the association between children's exposure to TRAP metrics and their risk of 'asthma' incidence or lifetime prevalence, from birth to age 18 years old.Study appraisal and synthesis methodsWe extracted key characteristics of each included study using a predefined data items template and these were tabulated. We used the Critical Appraisal Skills Programme checklists to assess the validity of each included study. Where four or more independent risk estimates were available for a continuous pollutant exposure, we conducted overall and age-specific meta-analyses, and four sensitivity analyses for each summary meta-analytic exposure-outcome association.ResultsForty-one studies met our eligibility criteria. There was notable variability in asthma definitions, TRAP exposure assessment methods and confounder adjustment. The overall random-effects risk estimates (95% CI) were 1.08 (1.03, 1.14) per 0.5 × 10− 5 m− 1 black carbon (BC), 1.05 (1.02, 1.07) per 4 μg/m3 nitrogen dioxide (NO2), 1.48 (0.89, 2.45) per 30 μg/m3 nitrogen oxides (NOx), 1.03 (1.01, 1.05) per 1 μg/m3 Particulate Matter < 2.5 μm in diameter (PM2.5), and 1.05 (1.02, 1.08) per 2 μg/m3 Particulate Matter < 10 μm in diameter (PM10). Sensitivity analyses supported these findings. Across the main analysis and age-specific analysis, the least heterogeneity was seen for the BC estimates, some heterogeneity for the PM2.5 and PM10 estimates and the most heterogeneity for the NO2 and NOx estimates.Limitations, conclusions and implication of key findingsThe overall risk estimates from the meta-analyses showed statistically significant associations for BC, NO2, PM2.5, PM10 exposures and risk of asthma development. Our findings support the hypothesis that childhood exposure to TRAP contributes to their development of asthma. Future meta-analyses would benefit from greater standardization of study methods including exposure assessment harmonization, outcome harmonization, confounders' harmonization and the inclusion of all important confounders in individual studies.Systematic review registration numberPROSPERO 2014: CRD42014015448.