In this guest blog, Professor Richard Gray outlines his plans to “shock” nursing science into action over its low clinical trial registration rates.
In nursing science, the need to register trials has got little traction, and compliance appears to be shockingly low.
Our group undertook a review of the registration status of 151 randomized controlled trials published in nursing science journals in 2017. We found that only one in ten studies were prospectively registered, and almost two out of three were not registered at all.
These observations should be extremely concerning to the discipline and raises some important questions:
It is some fifteen years since the International Committee of Medical Journal Editors (ICMJE) declared that their journals require “registration of any clinical trials in a public trials registry at or before the time of first patient enrolment as a condition of consideration for publication.”
Why don’t journal editors enforce trial registration? Why don't nurses register trials?
The Journal of Advanced Nursing (JAN), who published our review, have responded positively to issues we raised. The Editor in Chief published a blog in July 2019 stating that the “JAN insists all trials which it publishes - in fact, all studies of patient interventions as defined by the WHO […] - are registered prospectively and demonstrably. If they are not, they will be rejected."
The blog post has a section for comments, and it is interesting that only one person had posted a comment (and that person was me). This made me wonder how much interest our observation is garnering among research colleagues. Altmetrics are intended as a timely measure of "who is talking about your research." Who is talking about trial registration in nursing science? Our altmetric score suggests that five (again, one of which was me) people had tweeted about the study. (Engagement data correct as of 18 July 2019)
Trial registration, or lack thereof, does not, seemingly, grab the attention of nurse researchers. And this is a problem.
Prospective registration is a potent tool in ensuring unbiased reporting of research studies. Registration of a trial before the first participant is enrolled helps identify and deter selective reporting of outcomes (p-hacking and HARKing). And there is evidence to support this assertion: compared to unregistered studies, prospectively registered trials are more likely to be published without authors altering the study protocol or amending the main outcome. The World Health Organisation, the Cochrane Collaboration, and the AllTrials campaign have all come out in enthusiastic support for trial registration.
In leading medical journals, such as the British Medical Journal (BMJ), author guidelines are explicit: “BMJ will not consider reports of clinical trials unless they were registered prospectively before recruitment of any participants.”
Contrasting the editorial position of the BMJ with that of leading nursing science journals may go some way to explaining nurses seeming indifference towards trial registration. As part of our review, we extracted – from author guidelines – the editorial position of the journal about trial registration.
Of 51 nursing science journals that published at least one trial in 2017, a third offered no guidance about trial registration.
Thirteen journals referred authors to CONSORT when reporting a clinical trial. Whilst we acknowledge editors might argue that CONSORT (item 23) infers trial registration, it can hardly be considered a clear and ambiguous statement of intent a la the BMJ.
It seems to me that nursing science is stuck in a cycle of indifference.
Editors do not care about registration and do not enforce compliance. Consequently, researchers do not give heed to the need to register their trials.
Whilst this cycle persists, it is hard to see nursing making any progress.
What might we do? We are just finishing the second review, covering around 190 nursing science trials published during 2018.
We aim to publish a yearly report, and hope that by holding up a mirror, we can “shock” the nursing science profession into action.
It will be interesting to see quite how much “shocking” the profession requires.
Richard Gray is Professor of Clinical Nursing Practice at La Trobe University, Melbourne, Australia. Professor Gray can be contacted on Twitter: @drrichardgray