Abstract
Most of us are overloaded with information. Even when we have time to read some of the information that arrives on our computers daily, it is difficult to identify which information will be most useful and to recall the most up-to-date findings when we need them.
This is where systematic reviews come in. A systematic review focuses on a specific research question and tries to identify, appraise, select and synthesize all high-quality research evidence relevant to the question. These reviews are standard practice in clinical studies, but are not yet widely conducted in the field of laboratory animal science. Given that many studies using laboratory animals are aimed at improving human health (and health care), it seems logical that research using animals is reviewed in a similar way and adheres to similar high-quality standards. Apparently, this is not the case yet.
In order to highlight the importance of systematic reviews for the field of laboratory animal science, we launched the First International Symposium on Systematic Reviews in Laboratory Animal Science in Nijmegen (The Netherlands).
Conference highlights
During the meeting several important reasons for performing systematic reviews in laboratory animal science were discussed: (1) increasing the precision in estimating treatment effect estimates for trials in humans, (2) reducing the risk of false-negative results, (3) improving the applicability of preclinical animal research to clinical research, which will subsequently lead to improved methodological quality, (4) assessing the strength of evidence to determine whether an effect exists in a particular direction, (5) investigating heterogeneity among various studies, and last but not least, (6) obtaining novel results and generating new hypotheses without the need to perform new or additional animal studies.
However, although the need for systematic reviews in laboratory animal science was recognized, it was pointed out that conducting a systematic review in animal science involves several methodological challenges. First, in laboratory animal science, a variety of animal species and strains are used in a variety of experimental designs. This large variation/heterogeneity between the animal studies impairs a reliable comparison between the study results of these individual studies. Second, the methodological quality of individual animal studies is usually poor and is often inadequately reported. 1 This poor methodological quality increases the risk of bias which subsequently hampers the interpretation of overall effect sizes when performing a meta-analysis. In order to change this, editors and scientists need to demand the highest quality when performing and reporting animal studies. Malcolm Macleod, the keynote speaker of the symposium pointed out that there is only a weak association between the impact factor of journals and methodological quality in laboratory animal science (R 2 = 0.06). This point is in line with our finding that the journal with a relatively low impact factor scored highest on the Gold Standard Publication Checklist. 2 Third, interpreting the results of a meta-analysis can be hampered by publication bias, as negative and neutral results are less likely to be published than positive results. 3,4
The speakers of the symposium agreed that in order to implement systematic reviews in laboratory animal science, the aforementioned methodological and reporting issues need to be addressed and solved. Earlier clinical initiatives to further improve the conduct and reporting of systematic reviews in animal studies could be followed as good examples. 5,6 Furthermore, they all felt that by increasing the number of systematic reviews in animal studies, more awareness will be generated regarding the critical assessment of the strengths and weaknesses in study design, conduct and analysis of animal experiments. One way to reach these goals is to convince politicians and funding agencies of the need for systematic reviews in laboratory animal science.
The penultimate speaker, MP Esther Ouwehand pointed out that the Dutch parliament passed two motions regarding systematic reviews. In the first motion, the Dutch parliament agreed to promote systematic reviews in experimental medical science. In the second motion, they adopted the requirement for a general database/registry of animal studies that will help to prevent unnecessary duplication of animal studies and reduce publication bias.
Furthermore, the Netherlands Organisation for Health Research and Development (a funding agency called ZonMw) already require a systematic review as part of clinical grant applications to identify the added value of the proposed research. Due to the success of this programme, they recently developed a new programme for animal research entitled ‘more knowledge with fewer animals’; this programme includes a module on the synthesis of evidence in which a systematic review is one of the methods. This module also encourages funding for publishing neutral results and the organization of national workshops on systematic reviews of animal studies, intended for scientists considering conducting animal research that will be funded by ZonMw.
To the best of our knowledge, this approach is unique in the world and should be mirrored by other funding agencies, as it encourages researchers to change their scientific practices and meet public expectations. This approach will also help reduce excessive waste.
In conclusion, the First International Symposium on Systematic Reviews in Laboratory Animal Science (