Article Text
Abstract
Objectives To identify efficient PubMed search filters for the study of outdoor air pollution determinants of diseases.
Methods We listed Medical Subject Headings (MeSH) and non-MeSH terms that seemed pertinent to outdoor air pollutants exposure as determinants of diseases. Proportions of potentially pertinent articles retrieved by each term were estimated. We then formulated two filters: one ‘more specific’, one ‘more sensitive’. Their performances were compared with a gold standard of systematic reviews on associations between diseases and outdoor air pollution. We calculated, for both the filters, the number (of abstract) needed to read (NNR) to identify one potentially pertinent article, exploring three diseases potentially associated with outdoor air pollution.
Results The combination of terms that yielded a threshold of potentially pertinent articles≥40% permitted to formulate the ‘more specific’ filter. On the basis of the combination of all search terms under study we formulated the ‘more sensitive’ filter. In comparison with the gold standard, the ‘more specific’ filter had the highest specificity (67.4%; with a sensitivity of 82.5%) and the ‘more sensitive’ filter had the highest sensitivity (98.5%; with a specificity of 47.9%). For the ‘more specific’ filter and the ‘more sensitive’ one the NNR to find one potentially pertinent article was 1.9 and 3.3, respectively.
Conclusions The proposed search filters help investigating environmental determinants of medical conditions. We published them on: Curti S et al. PubMed search filters for the study of putative outdoor air pollution determinants of disease. BMJ Open. 2016;6 (12):e013092.