In addition to the NIHPG, there are several other options [14-17] for prospective authors of preclinical research to report what they did and found, including the Animal Research: Reporting of In Vivo Experiments (ARRIVE) [18]. ARRIVE is a reporting guideline developed in partnership with the National Centre for the Replacement, Refinement and Reduction of Animals in Research [19]. The guidance was rigorously developed and followed a review of published animal studies that provided a rationale for moving forward. The development process also included a consensus meeting with a broad spectrum of stakeholders interested in animal research. At its core, the ARRIVE guidance is a 20-item checklist covering what authors should report in the Introduction, Methods, Results, and Discussion sections of their papers. For example, item 8 asks authors to report on the experimental animals used: “Provide details of the animals used, including species, strain, sex, developmental stage (e.g. mean or median age plus age range) and weight”. There is evidence that authors are not providing this information [20], despite it being essential if others are interested in replication [21].
The ARRIVE guidelines and NIHPG have different scope but similar objectives. The NIHPG is primarily focused on journal policy; for instance, the guidance recommends that journals have either no or generous limits on the length of their methods sections. In contrast, the ARRIVE guidelines are focused on reporting within the manuscript itself with its 20-item checklist, compared to the six ‘core’ reporting items found in the NIHPG. There is little difference between the six proposed items and their corresponding items (sub-items) in the ARRIVE checklist. Thus, the NIHPG items may be considered a minimum set of reporting items that are focused on the internal validity and statistics of a paper, whereas the ARRIVE guidelines are comprehensive and cover the entire paper from title to discussion. Indeed, the NIHPG even recommends following community standards such as the ARRIVE guidelines, suggesting that it is intended to co-exist with these other standards.
While the NIHPG and ARRIVE efforts to improve the reporting of preclinical research, are admirable, it is possible they might be seen as confusing, particularly to prospective authors. For example, in journals that endorse both ARRIVE (also endorsed by many journals – [22]) and NIHPG, authors might not know whether to use the NIHPG and/or ARRIVE and may elect to use neither, thus negating the efforts of both groups (in areas of overlap, such as statistics, should authors choose one guidance over another one and, if so, which one?). Indeed, a similar situation occurred during the developing of the CONSORT Statement for reporting randomized trials. The CONSORT guidance was produced about 6 months before the Asilomar guidance. Dr. Drummond Rennie, then deputy editor of JAMA, recommended the two groups work together to produce a single guidance thus providing more clarity and direction for prospective authors. Similarly, the CONSORT group wanted to incorporate the TIDieR guidance for describing interventions [23] into the CONSORT checklist. Members of both groups discussed how best to do this and provided some guidance on this for prospective authors [23].
While the NIHPG guidance appears to have face validity and support, a more pressing question is whether it will have its intended effect, namely, improving the completeness and transparency when reporting preclinical research, as well as facilitating the likelihood of reproducible studies being done. There is likely little merit in editors asking authors to use the guidance – the intervention – if it has only a minimal effect on relevant outcomes such as better reporting and enabling others to adequately replicate methods. Unfortunately, an assessment of the benefits (and harms) of reporting guidelines has not been a top priority of guideline developers [24], to date. We hope the NIHPG developers will actively plan to conduct an evaluation of their guidance and encourage others to do likewise. Attention to relevant outcomes will be important to consider. Evaluations assessing adherence to reporting guidelines (i.e., in addition to endorsement) may provide a more meaningful insight into its impact on completeness of reporting when used at different stages of the editorial process. One such recently performed evaluation [25] incorporated these concepts by comparing the use and non-use of reporting guidelines during peer review on author-revised manuscript quality.
Having produced the NIHPG – an intervention to help improve the transparency and better reporting of preclinical research – it is important to consider its endorsement and implementation. More generally, are journals using similar explicit language regarding the endorsement of either guidance? Our experiences are that there is a wide variability in the language of endorsement and that this is confusing to prospective authors, thus potentially reducing the intended effectiveness of the guidance. Guideline developers are starting to provide greater guidance to help journal editors achieve a more standard and successful endorsement strategy [9].
