Science appears healthy on paper, but at what cost? Although the volume of published research doubles every 9 years,1 retraction rates are increasing faster.2 Recent studies find that up to 40% of researchers admit engaging in questionable research practices—involving shortening studies to give more favorable results, and “spinning” negative results into positive findings.3 Pressures are partly to blame: almost 60% of respondents to a national survey of scientific researchers in the United Kingdom reported being tempted or pressured to compromise their integrity and standards around scientific practices and reporting.4
Factors around today’s science create a petri dish for misconduct to thrive.
Incentives to create and maintain success narratives via institutional metrics and rankings combine with individual hunger to “stand out” from the academic crowd, progress a career, or—more pragmatically—secure a scarce academic job or peer esteem. Allegations of misconduct can now attract national media attention5 and are hugely damaging to public perceptions and trust in science, and raise justifiable doubts about the results of both future and past research. Even the normally conservative Economist magazine labeled the current ubiquitous system of scientific peer-review a “failure”, with journals seeking papers for publicity and “headlines” while regulations provide few checking mechanisms for fraud.6 This situation is compounded by editorial conflicts: more than half of journal editors in a recent study accept payments (a mean of $28,000) from pharmaceutical or device companies.7
What can be done to address this malaise? Authors, journals, and institutions have to work better together. Scientific journals form vital axes between scientists, industry, and the public because journals not only store and disseminate knowledge but also create the main scientific currency of commercial and career progression. Yet, the progress of science is too important to be left to a journal’s inclination around trial regulation, especially when such vested interests occur and impartiality, transparency, and study quality are so important to health care decision-making.
Reporting standards should be widened and extended to ensure that what really matters is the scientific content of those involved in a study, not the direction of its results. There is a pressing need for more standard definitions and protocols across our major journals that prioritize transparency in reporting and methods of clinical trials, including the publication of negative trial and meta-analysis findings.
It is concerning that in many instances study data are “owned”, not by the scientists associated with a study, but the pharmaceutical company overseeing it—an imbalance with serious consequences for public safety and scientific integrity when findings (particularly when negative) are withheld from the public domain. Mandatory availability of open data provides the ultimate means to assess the validity of analyses and interpretations and should be considered the norm for publicly funded research. Trials carried out by industry should be led via a model based on direct partnerships with health care organizations, with input and oversight from committees involving independent scientists, professional associations, ethicists, and patients.
To read this article in its entirety please visit our website.
-David R. Thompson, PhDa,, Alexander M. Clark, PhDb
This article originally appeared in the May 2019 issue of The American Journal of Medicine.
