Preprints in Clinical Research?


On September 11, 2015, the National Heart, Lung, and Blood Institute (NHLBI) announced that it was stopping its Systolic Pressure Intervention Trial (“SPRINT”). The Institute’s Data Safety and Monitoring Board (DSMB) had reviewed interim data and concluded that the results demonstrated clear benefit from aggressive blood pressure lowering. The trial enrolled over 9300 adults with systolic hypertension and increased cardiovascular risk and randomized them to standard control (aiming for a target systolic blood pressure of 140 mm Hg) or to aggressive control (aiming for a target blood pressure of 120 mm Hg). The DSMB members found that patients randomized to aggressive control were less likely to suffer from major cardiovascular events and less to likely to die for any reason. The NHBLI reviewed the DSMB’s recommendations and agreed to stop the study.

However, when NHLBI announced its decision on September 11 (disclosure – I was at the NHLBI at the time and was involved with these events), it did not release a number of details, details like how many patients experienced events in each study arm. For some in the scientific community, NHLBI’s unwillingness to release those details were concerning and emblematic of delays in transmitting clinical research results to the public. Eric Topol (Scripps Translational Science Institute) and Harlan Krumholz (Yale University) published an Op-Ed piece in the New York Times arguing that the time has come for clinical researchers to post their unpublished results, or “preprints” on the Internet. Preprints are manuscripts that are not yet officially published and that have not yet undergone formal peer review and editing.

In the December 19 issue of the Lancet, Drs. Topol, Krumholz, and I published a follow-up Commentary asking the question about whether it is “Time for prepublication culture in clinical research?” In many fields of science, like physics and mathematics, prepublication is the norm. There are a number of prepublication venues including arXiv and PeerJ. Biology has been late for this party, which some argue is a serious problem as it is taking longer and longer for biomedical scientists to get their findings published. The discussion of preprints has become increasingly active with the launch of a relatively new site, bioRxiv, which posts biology preprints, and an upcoming meeting on “Accelerating Science and Publication in Biology”, which I and several members of NIH leadership will be attending. Many publishers of journals in the broader life sciences research space are “preprint-friendly”. Clinical researchers, however, have almost never posted preprints because of a longstanding culture that dates backs to 1969, when the New England Journal of Medicine posted its well-known “Ingelfinger Rule,” stating “Articles are accepted for consideration with the understanding that they are contributed for publication solely in this journal.” Over time this has come to mean no prepublication – if a full manuscript is posted on the Internet that means it cannot be considered for eventual publication in a clinical research venue like the New England Journal.

We encourage you to read the commentary and join a dialogue on scientific communications, whether in clinical research in particular or biomedical research in general. In a subsequent edition of Open Mike we’ll take a look at one bioRxiv preprint on “Relative Citation Ratio” as an example of how the process might work.

Before submitting your comment, please review our blog comment policies.

Leave a Reply

Your email address will not be published. Required fields are marked *