New PubMed Integration Supports Scientific Integrity
It is widely accepted that receiving funding from an organization may come with pressure to publish results that are favorable to the funder. Marion Nestle found that 156 of 168 industry-funded research articles reported results that were favorable to their sponsors’ interests. This calls into question the scientific integrity of industry-funded research. In a bid to promote scientific integrity, seven organizations, 62 scientists and doctors, and five United States Senators signed a letter asking the National Institutes of Health and the National Library of Medicine to publish conflict of interest statements with their associated abstracts. This move should increase the likelihood that any potential conflict of interest would be recognized by the readers of PubMed-indexed journals. What can we expect now that these life science journals’ abstracts will contain these conflict of interest statements?
According to a National Library of Medicine technical bulletin, if a publisher of a PubMed-indexed journal has a conflict of interest statement associated with a given article, this statement will now appear beneath the abstract. If you are particularly interested, a search for “hascois” will return all PubMed citations with conflict of interest statements. Michael Jacobson, who is both the president of the Center for Science in the Public Interest and a signatory to the letter, believes that the addition of conflict of interest statements to the abstracts of these articles will make PubMed search results even more meaningful. He also expressed a hope that journalists would make consistent use of this feature when reporting on the findings of PubMed-indexed journals.
Ease of Use
In order to determine if researchers were being influenced by their funders, Marion Nestle tracked 168 industry-funded studies. She reports that acquiring the conflict of interest information was no easy task. Having the conflict of interest statement be associated with the abstract would simplify the process of conducting a similar study in the future. It would also improve the accuracy of such a study. Critically, it should help to indicate the level of scientific integrity and impartiality resident in each PubMed-indexed piece of research. Senator Richard Blumenthal expressed concern that the industry may be funding labs to acquire research in support of their own internal agendas. He commended this step taken by the National Library of Medicine and the National Institutes of Health to enhance transparency in published work. Senator Blumenthal also suggested that making the conflict of interest statements readily available to everyone with the article abstract would allow readers to make judgments about how trustworthy an article’s results were.
The call to include the conflict of interest statement with the abstract is based on the simple truth that many PubMed users will only ever read the abstract. This may be because they then realize the article is not what they need. It could also be that the article is behind a paywall which prohibits access if the reader has no funds to pay for the full article. This may be the case for some journalists who report on scientific matters. There have been many instances where research has been skewed in favor of the funder. As an example, industry-funded research attempted to obscure the link between soda and obesity, diabetes, and heart disease. Knowing what funders or employers scientists are associated with should aid the readers of their papers in deciding how objective their results might be.
Resolving Search Conflicts
Now that the conflict of interest statements are included with the life science journals indexed by PubMed, the National Library of Medicine is trying to find a way to not have these statements impact search results. This additional data included with the abstract may have words, concepts, or phrases that have nothing to do with the scientific content of the article. This could lead to data mining programs that search for indexed words in abstracts in order to compile a list of studies returning results that have nothing to do with the subject of the search. Betsy Humphreys, Acting Director of the National Library of Medicine, suggests that this could be avoided by supplying the conflict of interest data as a hyperlink. Whatever the solution, supplying this additional information may require changes in how publishers submit information to PubMed. Betsy Humphreys stated that these changes would probably be easier for larger publishing houses than smaller ones.
Although there has been some additional work involved in making the conflict of interest statements appear with the abstracts of PubMed journal articles, this change took effect in March 2017. There can be no doubt that this change undertaken by the National Library of Medicine and the National Institutes of Health will promote scientific integrity, especially in life science journals. Having the conflict of interest statements appear with the abstracts of PubMed-indexed journals is definitely a step in the right direction.