Enhancing Communication and Understanding with the New ARRIVE Guidelines for PublicationBy Marianne Evola
The National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs) has released a series of publishing guidelines for research conducted on animals. The new Animal Research: Reporting of In Vivo Experiments (ARRIVE) Guidelines address ways to improve the reporting of experimental protocols, but these guidelines also should be considered while designing experiments so that researchers can avoid problems with peer review and publishing once an experiment is complete.
The goal of these guidelines is not to interfere with creativity in research but rather to better educate scientists on design considerations when conducting research with animals as well as raise the quality of reporting to enhance the ability to replicate research. These guidelines are a good reminder that research is not about “trade secrets.” Rather, the ability to replicate results is central to the validity of research, and these publication guidelines are designed to address the reported rise in replication failure. The link to the ARRIVE Guidelines can be found at the end of this article.
The ARRIVE Guidelines are most interesting because they are unremarkable. At first read, they seem to be a list of common practices of which most labs already adhere. However, after thinking about the guidelines a bit and how I was going to discuss them in this month's article, I started pondering how my training may have differed from the training of others and how the guidelines address some of the interesting transformations of science in the past couple of decades. My graduate training was in animal behavior. We used measurable behavior as a dependent variable in our research. As such, a great deal of our training was on how to control the many environmental and training conditions that could alter a behavioral outcome. Stress, environmental conditions, sensory input, circadian rhythms, novelty, animal handling and transport, and even loud talking/laughing in limited lab space are all known to impact animal behavior. Admittedly, the laughing and talking is often a difficult factor to control since the lab supervisor comes across as a bit of a scrooge to personnel that are enjoying healthy discourse and sharing ideas. Regardless, we spent a great deal of time assessing the environment and experimental considerations to minimize the impact of extraneous factors on our research. As such, our research addressed the many procedural questions raised in the ARRIVE Guidelines. However, as I thought more about it, I wondered how much of that was actually included in our research reports.
Again, these are not secrets kept from publishing but rather daily lab struggles to control the research environment. For example, when I was in graduate school a fellow student was conducting an observational assessment of her rats with various drug manipulations while they were free to engage in a normal repertoire of behavior (i.e., behavior that was not manipulated or constrained). Due to space issues, she was required to conduct this work in a small room contained in a very high-traffic area of the lab. She quickly learned that that any extraneous noise disrupted rat behavior and thus would markedly impact her data if she did not work to control it. Therefore, she actively chased people out of the research area when she started her daily experiments, posted signs for quiet, and then throughout the day would continue to hush and chase out extraneous personnel from that research area. Notably my laugh, which has been known to carry, was of particular concern to her, and thus, I was exiled during her experimental procedures. Luckily, at that time, I was doing most of my work in another building, so I was not often around, but I was quickly escorted out if she became aware of my presence. No one actively tried to disturb her experiments, nor were we overtly disruptive, but rather we were engaging in our daily work in a manner that was very normal to us. However, due to the sensitivity of her experimental measures, our behavior needed to change. Eventually, because of her assertive and sometimes unpopular control of that research space, it became recognized as a quiet zone. I don't raise this point to criticize her for controlling the environment, nor to the personnel who struggled to remember that the work environment needed to change. I raise this point to question to what extent the need for quiet was raised in publication and whether others, not accustomed to assertively addressing environmental factors in in vivo work, would have gone to the same effort to control the research environment. If they did not, their research results would markedly differ from hers.
Actually, it is rather intriguing that many research labs can be obsessed with standardizing practices that may not be reported in the literature. Often new student assistants are overwhelmed with how picky and precise labs can be about maintaining research protocols and controlling procedure. We worked to control the time of day (i.e., the controlled light/dark cycle) for experiments to address the impact of circadian rhythms on experiments. We controlled how animals were transported so that transport was routine and stress free. Animals were acclimated to handling, environments, treatments, equipment and procedures before any critical treatment or experiments were conducted to address the impact of novelty. The list of controls can go on and on. As students were introduced to research procedures, the new students often perceived mentors or lab supervisors as obsessive control freaks. This was something that I would warn new students about when they started working in our lab. Experiments needed to be conducted in a very particular way so that procedures were consistent among the student assistants to minimize error as well as variability. Every procedural control had been introduced for a reason, and I encouraged them to ask for an explanation when they felt that I was too controlling. As students began to understand why we wanted such precise control of experimental procedures, they would readily adhere to the protocol. We take great effort to teach new students these tightly controlled practices, and it is these practices that should probably be better addressed in the literature.
As I pondered this, I realized that every discipline has picky and precise practices that are stressed to new students, yet they are so standard that they are probably not mentioned in publication because to do so is to waste words and publishing space. To members of a particular academic branch, these fixed practices are so standard that reporting them in publication is much like telling someone to “turn on the water” when giving instruction on how to wash dishes. There has never been any intention to keep research practices secret, but rather they were so historically standard in an associated academic lineage that everyone stopped wasting the space in publication. It is the failure to publish these practices that may have contributed to the recently identified crisis of replication failure in biomedical research. It is also these practices that the new ARRIVE Guidelines for publication are seeking to address.
Since I was in graduate school, many branches of research have undergone notable changes so that previously diverging areas of science have been bridged back together. One of these growing bridges is between in vivo and in vitro biomedical research. I remember a “journal club” meeting when I was in graduate school where a couple of the faculty members in our department asserted that researchers conducting behavioral and/or in vivo research needed to develop in vitro research skills to maintain future marketability. They presented job postings from several issues of Science to demonstrate that most research positions listed were in biochemistry and genetics work. Since that time, I witnessed researchers trained only in in vitro research struggle to address recommendations from grant reviews that requested in vivo assessment of promising in vitro treatments and/or techniques. As such, the literature has revealed a progression toward whole animal work by labs that had not historically worked in whole animal models.
However, when many of these labs went to the literature to figure out how to address these research questions, many of the historical controls and procedural considerations that in vivo labs have always addressed were are not readily present in publication. As labs attempted to bridge in vitro to in vivo work, they failed to address many standard procedural factors due to the “holes” in the literature. As a result, experimental procedures in biomedical research markedly differ across disciplines and between labs. It is not surprising, therefore, that difficulties with research replication have resulted. The ARRIVE Guidelines were constructed to raise our awareness to many of these practices that are often not shared in contemporary literature and ensure that researchers consider them during experimental design as well remember to publish these practices. In addition, it provides peer reviewers and editors a template of factors to consider and/or request when reviewing manuscripts that may include in vivo procedures of which they are not familiar.
Failure to replicate research results is a crisis that is being addressed in many areas of research, including biomedical research. Unfortunately, failure to replicate is often being interpreted as poor ethical practices by researchers, when in fact it may be poor communication of standard practices. The ARRIVE Guidelines may be an effective tool for addressing communication oversight. Therefore, these guidelines should be reviewed by both new and seasoned researchers to improve experimental design, training and communication of research. In addition, these guidelines may also provide communication structure between in vivo and in vitro collaborators regarding the many controls that are needed for in vivo research that are not reported or described in the literature. Use and discussion of the ARRIVE guidelines may create understanding between collaborators when the in vivo researcher starts to reprimand disruptive talking and laughter in the research setting and basically begins to assert their many “control freak” ways.
Marianne Evola is senior administrator in the Responsible Research area of the Office of Research & Innovation. She is a monthly contributor to Scholarly Messenger. Alice Young, associate vice president for research/research integrity, is a contributing author/editor.