Major depressive disorder (MDD) is the most severe depression type and one of the leading causes of morbidity worldwide. Animal models are widely used to understand MDD etiology, pathogenesis, and treatment, but the efficacy of this research for patients has barely been systematically evaluated. Such evaluation is important given the resource consumption and ethical concerns incurred by animal use.
We used the citation tracking facilities within Web of Science and Scopus to locate citations of original research papers on rats related to MDD published prior to 2013—to allow adequate time for citations—identified in PubMed and Scopus by relevant search terms. Resulting citations were thematically coded in eight categories, and descriptive statistics were calculated. 178 publications describing relevant rat studies were identified. They were cited 8,712 times. More than half (4,633) of their citations were by other animal studies. 794 (less than 10%) were by human medical papers. Citation analysis indicates that rat model research has contributed very little to the contemporary clinical understanding of MDD. This suggests a misuse of limited funding hence supporting a change in allocation of research and development funds targeting this disorder to maximise benefits for patients.
Depression is the leading cause of disability worldwide (World Health Organization,
2019). Nowadays, it is judged to affect more than 320 million people of all ages and
genders (Vos et al., 2016), even though it is more frequent in women than man
(Ferrari et al., 2013).
MDD etiology is not completely understood yet. Most authors agree that there is a combination of biological and environmental factors that determine the triggering of the disorder (Mandelli and Serretti, 2013). Biological factors to take into account include genes, neurotransmitters, and hormones, while environmental factors include childhood trauma, stressful life events, sexual abuse, low educational attainment, and differences in personality traits
Recent studies also suggest that there is a link between inflammation and MDD, suggesting that MDD has an inflammatory subtype (Beijers et al., 2019), but the claims that inflammation has a role in etiology of MDD are still being disputed (Miller, 2018). The same happens with changes in gut microbiome in MDD patients (Winter et al., 2018). While the link between gut microbiome and depression is well documented, the question of the causality in the connection between the two remains to be robustly answered (Winter et al., 2018). It is important to mention that these patterns may be true for all biological changes found in MDD patients. It is almost impossible to determine if the biological changes caused MDD or if MDD caused the biological changes. Conversely, most environmental factors involved in MDD are definitely a primary cause. In this regard, the big unanswered question that remains is why does the same life event trigger MDD in one person and not in another.
The citation analysis was performed between January and August of 2019. PubMed and SCOPUS were searched for publications using rat models to investigate MDD. We searched PubMed using Medical Subject Heading search terms (MeSH terms): “Depressive Disorder, Major” AND “rat” OR “rodent.” MeSH terms are a comprehensive list of key terms made available by PubMed designed to identify all relevant studies in an area (Uman, 2011). So, searching for “MDD” retrieves other nomenclatures for the same disorder such as melancholia. Similarly, the search term “rat” retrieves papers using all rat species. We used PubMed filters to exclude review articles (“review,” “systematic review,” “meta-analysis,” “bibliography”) as well as opinion articles (“biography,” “auto-biography,” “comment,” “editorial,” “interview”).
Web of Science is a major scientific citation indexing service that encompasses over 50,000 scholarly books, 12,000 journals, and 160,000 conference proceedings. Scopus is the largest citation database; it covers nearly 36,377 titles from approximately 11,678 publishers. For each rat study, we recorded the total number of times it was cited, and allocated each citation to one or more of seven categories, defined prospectively: – Animals. This category included all animal studies from observational ethological studies to invasive procedures as defined by Knight (2011), i.e., interfering with bodily integrity (whether through puncture or incision) or production of genetically modified animals. This category also included severe procedures (as defined by current European Legislation Directive 2010/63/EU) commonly used in mental disorder research such as inescapable electroshock or isolating social animals for long periods. We recorded within this category which animal papers focused on MDD and which focused on other subjects
The 178 original rat studies focused on MDD that were published before the end of 2013 were cited 8,712 times by August 2019. Of these 178, 87 (49%) studies were never cited in subsequent publications describing human studies on MDD, and 53 (30%) were never cited in any publications related to human research, either focused on MDD, or on other disorders such as post-traumatic stress disorder or bipolar disorder. As shown in Figure 1, rat studies were mainly cited by other animal research papers (4,641), followed by review papers (2,909), human studies (794), in vitro papers (211), editorials (58), in silico papers (57), and human social papers (one). 230 citations were unavailable to us due to access or language barriers. These were removed from further analysis.
The proportion of citations by human medical papers was 9.1% while the proportion of citations by animal papers was 53.3%. This corresponds to a mean difference between the proportions of citations by human and by animal papers of −46% (p < 0.001). Beyond the statistical significance, this is certainly a considerable practical difference that reflects almost 100∗(53.3–9.1)/53.3 less citations by human papers than by animal papers.
The majority of the rat papers located in this study were cited by subsequent animal research papers, but about half (49%) of the original papers retrieved were never cited in subsequent papers related to MDD in humans, and in fact about a third (30%) were never cited in any subsequent human studies. Our citation analysis revealed that only around a tenth (9.1%) of the total number of citations were by human medical papers. This result contradicted the hypothesis that the proportion of citations by human medical papers would be substantially higher when compared to other research categories. This hypothesis was based on the assumption that animal models are essential to clinical research, as promulgated by several authors (e.g., Wang et al., 2017; Akil et al., 2018). Hence, our results raise doubts about the justification for these animal studies.
The results of our study are in agreement with previous studies that empirically evaluated the contribution of animal models to human healthcare either through citation analysis (e.g., Shapiro, 1998; Knight, 2007; Long et al., 2014; Carvalho et al., 2016) either through other methods such as systematic reviews (e.g., Perel et al., 2007), social science studies (e.g., Shapiro, 1998; Compton et al., 2019), historical analysis (e.g., Menache, 2012), among others (for an extensive review on this topic see, for example, Knight, 2019). This suggests that biomedical research resorting to animal models is not normally considered significant, or particularly visible to, the human medical research community. Supporters of animal models of human disorders claim that this happens: (a) due to differences in the way basic animal work and human clinical trials are conducted, and propose a change to a translational biomarker-based approach within early steps of pre-clinical research (Garner, 2014); or that this occurs (b) due to failings in study design, conduct, analysis, and reporting (as described by Pound and Ritskes-Hoitinga, 2018), which could be resolved with better reporting and better methodological quality (as proposed by Fabian-Jessing et al., 2018).