Background: This paper aims to ascertain whether media coverage of cancer research articles may increase their citations in the scientific literature by observing whether journal articles covered by media would be cited more frequently than similar articles not covered by the media.
Methods: We selected 4 journals with a wide range of journal impact factors(JIF), JAMA (30.4), Lancet Oncology (24.7), BMJ (16.4) and Medical Journal of Australia (3.8), searched the LexisNexis database to identify cancer related articles reported by the major world media in 2010. We used PubMed to find the journal articles corresponding to the news, grouped them by study type (randomized controlled trial or observational study), and compared the mean citations of articles covered by the media and with those of articles not covered by the media for a 4-year citing period using ISI Web Science.
Results: We identified a total of 393 cancer related article in our 4 journals in 2010, and followed them for a 4-year citing period from 2010-2014. By comparison of the mean of 55.8 citations in 161 articles covered by the media with a mean of 35.7 citations in 232 articles not covered by the media, the citation ratio 1.55 suggested that article citations may be influenced by media coverage, and the JIF did not correlate with the mean citation index. Also, media were more likely to cover observational studies and less likely to cover randomized controlled trials (35% vs. 17%).
Conclusions: this study suggests a possible association between media coverage of research and scientists’ subsequent citations of the work. However, media coverage alone is unlikely to explain entirely our findings, as selection bias on the part of the health reporters is likely to play a significant part as well. Furthermore, this study represents only one snapshot of citation patterns. It is possible that the association between news coverage and citation rates could change with time and geogrphy, which will need further discussion.