Air pollution and publications: historic and emerging trends in research topics - a bibliometric study

Abstract Background
The scientific literature on air pollution and its health effects has grown steadily over the last half a century. In this study, we apply text analysis, clustering, and natural language analysis to uncover emerging research themes and remaining knowledge gaps.
Methods
We used publication data made public through PubMed to analyse long-term and recent research trends focusing on the health impacts of air pollution. We identified research themes by analysing trends in Medical Subject Headings (MeSH) terms and performed a quantitative text analysis of article titles and abstracts. We used natural language analysis to identify topic clusters in the literature.
Results
Emerging research themes included the use of low-cost exposure sensors and satellite-based exposure assessment, increasing exposure assessment related to wood smoke, while, relatively, interest was decreasing in traffic and occupational airborne exposures. There was increasing interest for mitigation strategies, possibly indicating a shift from regulations to interventions. Emerging health outcomes examined included kidney diseases, neurological and mental disorders, diabetes, pregnancy and neonatal outcomes, and fertility. Knowledge gaps identified included the health effects of airborne microplastics, volatile organic compounds, and hydrocarbons, and the relation between air pollutants and liver, bone and autoinflammatory diseases. We found improved reporting of study design in titles and abstracts and increasing numbers of meta-analyses and systematic reviews.
Conclusions
This bibliometric assessment showed that air pollution health research is a maturing research field, while it also highlighted important knowledge gaps. Our results are valuable for guiding future research as well as providing directions for future policy development.