blog

James Barker (Publishing Executive, F1000) and Dr. Matthew Brooke (Programme Manager – Reporting of Animal Research, NC3Rs) discuss the road to the ARRIVE 2.0 guidelines and how F1000Research is adapting its policies in support of clear, transparent reporting for research using animals.

Animals are an essential part of research, particularly in pre-clinical studies. However, how this research is conducted has historically been underreported in the literature. Given the importance of the results produced by these studies and the ethical concerns around the use of animals, clear and transparent reporting is vital.

A study from the NC3Rs in 2009 investigated the reporting of research using animals. They reviewed 271 articles reporting the use of animals and found:

• Only 59% of studies stated their hypothesis.
• 87% did not use randomization.
• 86% did not use blinding.
• Only 70% of publications that used statistical methods described their methods and used a measure of error or variability.
• Only 8% of articles provided raw data.

These deficits in the reporting of animal research highlighted a critical problem that needed to be addressed. In the words of the authors:

“For scientific, ethical and economic reasons, experiments involving animals should be appropriately designed, correctly analyzed and transparently reported.”

Arriving at ARRIVE

To address these issues, the NC3Rs published the Animals in Research: Reporting In Vivo Experiments (ARRIVE) Guidelines in 2010. Guided by the CONSORT statement for randomized clinical trials, the NC3Rs developed a 20-item checklist describing the minimum standards for the reporting of animal studies. This checklist addressed the reporting deficits highlighted by the 2009 survey. It included details such as:

• The number and specific characteristics of animals used
• Details of housing and husbandry
• The experimental, statistical, and analytical methods

The guidelines were widely adopted by publishers, including F1000Research. However, studies following the publication of ARRIVE have demonstrated that adherence to the guidelines was low. One study looking at PLOS and Nature titles found little improvement in the issues highlighted by the 2009 survey after the endorsement of ARRIVE. This suggests that authors, editors, and reviewers were largely ignoring them. Additionally, a randomized controlled trial on the use of the ARRIVE guidelines (IICARus) showed that simply mandating completion of a checklist did not improve adherence.

The creation of ARRIVE 2.0

In response to these and other findings, the NC3Rs convened an expert international working group to revise and update the ARRIVE guidelines. The group included researchers from academia and industry, funders, journal editors, methodologists, and statisticians.

This process involved a survey of international in vivo researchers to understand the reasons behind poor reporting and whether knowledge of ARRIVE correlated with better study design. Additionally, there was an expert Delphi exercise involving stakeholders in 19 countries and extensive road-testing of draft guidelines by researchers preparing manuscripts. This resulted in the publication of ARRIVE 2.0 in 2020.

Significant changes to the ARRIVE guidelines

The ARRIVE 2.0 guidelines divide a list of 21 guideline items into two sets: ‘the Essential 10’ and ‘the Recommended Set.’ ‘The Essential 10’ are the bare minimum of information that researchers should include in any animal research manuscript. Without these ten guidelines, reviewers and readers cannot confidently assess the reliability of the research findings. In contrast, ‘the Recommended Set’ complements the ten and adds important context to a study. Prioritizing the guidelines in this way highlights those items most important for assessing reliability. It also allows researchers, journals, and other stakeholders to focus initially on these most crucial pieces of information in manuscripts.

Moreover, an extensive and detailed explanation and elaboration (E&E) document accompanies ARRIVE 2.0. This document explains the background and rationale behind each item and provides examples of good reporting from the published literature that comply with each one. As a result, this helps researchers to understand the meaning and necessity of these reporting recommendations. This information, including resources such as translations, author checklists, and a compliance questionnaire for journals, is available on the ARRIVE guidelines website.

F1000 and ARRIVE 

The IICARus trial demonstrated that an editorial emphasis on adherence to the guidelines is essential. At F1000, this has been in practice since the launch of the ARRIVE guidelines in 2010. We make no judgments based on novelty or impact, so our checks focus on ensuring robust reporting to facilitate replication. We explicitly ask submitting authors to conform to ARRIVE. Our editors leave comments to request expansion in areas where items required by ARRIVE are lacking and will point the author to the specific item of the ARRIVE guidelines.

However, our dedication to enforcing adherence to ARRIVE has not always been clear from our guidelines and policies.

With the publication of ARRIVE 2.0, we have decided to clarify our policies around our use of ARRIVE author checklists. The changes to our policies are:

• All articles reporting in vivo animal studies must conform to the ARRIVE Essential 10, though authors are encouraged to comply with the Recommended Set.
• All articles published in the NC3Rs Gateway must conform to the full ARRIVE 2.0 guidelines.
• Authors must provide a completed author checklist  (for either the Essential 10 or full ARRIVE guidelines, as appropriate) and upload it to a repository as extended data in the ‘Reporting guidelines’ section.

Our editors will continue to use the guidelines as the framework for pre-publication checks. With the addition of the published checklist, reviewers and readers will be able to verify the article’s adherence to each item. ARRIVE now joins our list of mandatory guidelines alongside CONSORT and PRISMA.

Journals and publishers have a crucial role in improving the standard of animal research reporting in the scientific literature. The ARRIVE guidelines 2.0 have been designed with implementation by journals and publishers in mind. With these changes, we renew and strengthen our commitment to publishing animal research that is “appropriately designed, correctly analyzed, and transparently reported.” Find our updated policies here.

---