Despite a significant increase in the volume of biomedical research over the past decade, there has been limited translational success in clinical medicine [1, 2]. Reproducibility specifically for animal research is low [3–5]. In attempts to address this problem, the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines as well as the revised National Institutes of Health grant application process have proposed standards for research involving animals to enhance the quality of experimental design, study conduct, and analysis of results [6–8]. These steps are intended to reduce bias and ultimately improve reproducibility and facilitate the translation of biomedical research to novel clinical applications that improve patient outcomes. Additionally, there is an ethical dilemma regarding animal welfare as well as financial waste related to permitting investment into research without tangible returns [9]. Specifically for the field of critical care medicine, small studies have shown that animal research methodology, study design, and reporting tends to lack rigor in several important areas [10–13].

Improvements in reporting of key experimental design features could enable readers to better judge sources of bias and eventually enhance validity and likelihood of translation. The objective of our study was to evaluate all critical care journals and compare reported animal experimental research in 2005 vs. 2015 regarding power analysis, randomization, and blinding procedures. Our hypothesis was that there had been increased implementation of these methods in 2015 compared to 2005. Also, we sought to provide information on the status quo of reported experimental design features to promote rigor.

We performed an observational bibliometric analysis of animal research published in critical care medicine journals using PRISMA and STROBE guidelines [14, 15]. Journals were selected based on their inclusion on the Thomson Reuters™ Journal Citation Reports® subject category “Critical Care Medicine” [16]. A PubMed search included animal experimental studies published in 2005 and 2015. Our primary search criterion was that the article was reporting on an animal study based on an experiment. Animals were further defined as: “any of a kingdom of living things composed of many cells typically differing from plants in capacity for active movement, in rapid response to stimulation, in being unable to carry photosynthesis, and lack of cellulose cell walls” [17]. We excluded meta-analyses, case reports, historical articles, letters, review articles, and editorials. One investigator manually assessed the PubMed search results for animal experimental studies. Then, the PubMed filter “other animals” was applied to the initial search results to detect any animal experimental studies not found in the manual search. Journals that did not publish at least ten animal studies in both 2005 and 2015 were excluded from the analysis (Fig. 1). To assess consistency in the identification of manuscripts reporting on animal experimental research, a second investigator blinded to the results of the first investigator independently searched two journals that were randomly selected from the seven journals included in this study.

Next, we rated all animal studies selected. A computer-generated randomization scheme was used to randomize articles by both year and journal before the analysis (Excel, Microsoft Co., Redmond, WA). Studies were analyzed using their full-text Portable Document Format (PDF). Reporting of power analysis, randomization, and blinding was then graded using a 0–3 point scale (0-not mentioned, 1-mentioned but specified as not performed, 2- performed but no details given, 3-performed and details given) [18]. To assess inter-rater agreement for criterion ratings, we randomly selected 10 % of the total articles for re-rating by a second investigator blinded to the results of the first investigator.

The quality of research and reporting of animal studies in critical care medicine journals is an area of increased interest, especially as reproducibility and successful translation of basic science results to clinical application has been low [22–24]. In addition to impeding progress in the development of novel therapies, these issues also present ethical concerns [9, 25–27]. In attempts to improve animal research quality, initiatives such as the ARRIVE guidelines have been created to improve the methodological rigor and to enhance translation [8]. To date, there are few studies examining the reporting of recommended experimental design feature to increase scientific rigor and reduce bias in animal experimental critical care research.

In our study, we evaluated the methodological quality of animal research in critical care journals in 2005 and 2015 and found a significant increase in the reporting of power analyses, randomization, and sample size calculations. Our hypothesis that these metrics are more commonly reported in 2015 compared to 2005 was confirmed. Introduced in 2010, the ARRIVE guidelines [8] may have been one of several factors that led to the improved reporting of recommended study design features in 2015. Our analysis using an ordinal scoring system still found the lowest rating category to be the most common one for every criterion assessed, even in 2015. Contemporary research in the field of critical care reports on recommended procedures to improve experimental design rigor only in a minority of manuscripts. This is in line with the limited published literature on this topic. Bara et al. [13], reviewed 77 animal research articles published in critical care journals over a six-month period in 2012. They found that 61% reported randomization and 6% of these reported some type of allocation concealment and only 2% reported a method of randomization.

Huet et al. [12] highlighted the importance on enhancing animal research quality including improving the use of the 3Rs (replacement, reduction, refinement), which are the guiding principles for ethical animal testing [28–30]. They emphasized, however, that there continues to be poor adherence to these recommendations. Festing et al. [3], emphasized the historical significance of animal research and the major contributions resulting from it: animal research has led to the advancement of immunization medicine, use of vitamins in almost eliminating diseases such as scurvy and rickets, and the discovery of insulin and its effect on metabolic diseases. Yet, they also identified a lack of adherence to good practices of research design as a major impediment to progress in medicine.

Although enhanced translation is the ultimate goal of measures to improve experimental design rigor, it remains to be determined if there has been an improvement in reproducibility or successful translation of animal experimental research results. Given the significant time lag between the description of basic science results and publication of clinical trial results, proof of a direct relationship between reported experimental design rigor and translation to novel therapies for critically ill patients will be challenging. It is also possible that some articles may not have described quality metrics that were in fact utilized in the research protocol. In addition, editors and reviewers may have recommended reporting according to the more recent ARRIVE [8] guidelines during the review process. The observed difference between 2005 and 2015 may, therefore, reflect more a change in reporting as opposed to a change in experimental practices. Of note, an innovative online tool, the “Experimental Design Assistant” was introduced in October 2015 as a guide for researchers to assist in the rigorous design of experiments [31]. However, none of the articles included in our study mentioned utilizing this resource. Further, our search strategy may not have detected all animal research articles in critical care journals in the two time periods examined. However, almost perfect agreement existed between two independent investigators in this regard. Critical care relevant research is published in other (non-critical care medicine specific) journals, and we did not include non-critical care journals in this study. Indeed, when comparing 2005 to 2015, the annual number of animal experimental manuscripts published in critical care journals decreased by 139 articles. This contrasts with findings that overall, publications in the medical literature have been increasing in the last decade [32, 33]. Finally, publication bias was not assessed in this study. Publication bias likely has a significant impact on the quality of animal research and its ability to be translated into successful clinical trials [34, 35].

The application and reporting of recommended quality metrics in animal experimental research published in critical care medicine journals continue to be modest. However, the increase in reported measures aimed to improve experimental design quality and reduce sources of bias in 2015 compared to 2005 is promising. Reporting of blinding, randomization, and sample size estimates should be encouraged in future animal experimental publications in critical care medicine. The routine justification for the presence or absence of these study design features should be considered in reports on animal experimental research.