A network analysis of the propagation of evidence regarding the effectiveness of fat-controlled diets in the secondary prevention of coronary heart disease (CHD): Selective citation in reviews

A career in numbers: A citation network analysis of the work of RP Millar and his contribution to GnRH research

Here, we reflect on the long career in neuroendocrinology of a single, highly productive scientist ('Bob' Millar), by analysing his oeuvre of published papers through the lens of citation metrics. We use citation network analysis in a novel manner to identify the specific topics to which his papers have made a particular contribution, allowing us to compare the citations of his papers with those of contemporary papers on the same topic, rather than on the same broad field as generally used to normalise citations. It appears that citation rates are highest for topics on which Bob has published a relatively large number of papers that have become core to a tightly-knit community of authors that cite each other. This analysis shows that an author's impact depends on the existence of a receptive community that is alert to the potential utility of papers from that author, and which uses, amplifies, extends and qualifies the contents of their papers-activities that entail reciprocal citation between authors. The obvious conclusion is that a scientist's impact depends on the use that his or her contemporaries make of his or her contributions, rather than on the contributions in themselves.

Oxytocin—a social peptide? Deconstructing the evidence

In this paper, we analyse the claim that oxytocin is a ‘social neuropeptide’. This claim originated from evidence that oxytocin was instrumental in the initiation of maternal behaviour and it was extended to become the claim that oxytocin has a key role in promoting social interactions between individuals. We begin by considering the structure of the scientific literature on this topic, identifying closely interconnected clusters of papers on particular themes. We then analyse this claim by considering evidence of four types as generated by these clusters: (i) mechanistic studies in animal models, designed to understand the pathways involved in the behavioural effects of centrally administered oxytocin; (ii) evidence from observational studies indicating an association between oxytocin signalling pathways and social behaviour; (iii) evidence from intervention studies, mainly involving intranasal oxytocin administration; and (iv) evidence from translational studies of patients with disorders of social behaviour. We then critically analyse the most highly cited papers in each segment of the evidence; we conclude that, if these represent the best evidence, then the evidence for the claim is weak.

This article is part of the theme issue ‘Interplays between oxytocin and other neuromodulators in shaping complex social behaviours’.

Automated citation recommendation tools encourage questionable citations

Citing practices have long been at the heart of scientific reporting, playing both socially and epistemically important functions in science. While such practices have been relatively stable over time, recent attempts to develop automated citation recommendation tools have the potential to drastically impact citing practices. We claim that, even though such tools may come with tempting advantages, their development and implementation should be conducted with caution. Describing the role of citations in science’s current publishing and social reward structures, we argue that automated citation tools encourage questionable citing practices. More specifically, we describe how such tools may lead to an increase in: perfunctory citation and sloppy argumentation; affirmation biases; and Matthew effects. In addition, a lack of transparency of the tools’ underlying algorithmic structure renders their usage problematic. Hence, we urge that the consequences of citation recommendation tools should at least be understood and assessed before any attempts to implementation or broad distribution are undertaken.

Diversity in citations to a single study: A citation context network analysis of how evidence from a prospective cohort study was cited

Between its origin in the 1950s and its endorsement by a consensus conference in 1984, the diet–heart hypothesis was the subject of intense controversy. Paul et al. (1963) is a highly cited prospective cohort study that reported findings inconvenient for this hypothesis, reporting no association between diet and heart disease; however, many other findings were also reported. By citation context and network analysis of 343 citing papers, I show how Paul et al. was cited in the 20 years after its publication. Generally, different findings were cited by different communities focusing on different risk factors; these communities were established by either research foci title terms or via cluster membership as established via modularity maximization. The most frequently cited findings were the significant associations between heart disease and serum cholesterol (n = 85), blood pressure (n = 57), and coffee consumption (n = 54). The lack of association between diet and heart disease was cited in just 41 papers. Yet, no single empirical finding was referred to in more than 25% of the citing papers. This raises questions about the value of inferring impact from citation counts alone and raises problems for studies using such counts to measure citation bias.

Dietary Saturated Fats and Health: Are the U.S. Guidelines Evidence-Based?

The last decade has seen nearly 20 papers reviewing the totality of the data on saturated fats and cardiovascular outcomes, which, altogether, have demonstrated a lack of rigorous evidence to support continued recommendations either to limit the consumption of saturated fatty acids or to replace them with polyunsaturated fatty acids. These papers were unfortunately not considered by the process leading to the most recent U.S. Dietary Guidelines for Americans, the country's national nutrition policy, which recently reconfirmed its recommendation to limit saturated fats to 10% or less of total energy intake, based on insufficient and inconsistent evidence. Continuation of a cap on saturated fat intake also fails to consider the important effects of the food matrix and the overall dietary pattern in which saturated fatty acids are consumed.

Low-energy sweeteners and body weight: a citation network analysis

OBJECTIVE

Reviews on the relationship of low-energy sweeteners (LES) with body weight (BW) have reached widely differing conclusions. To assess possible citation bias, citation analysis was used to quantify the relevant characteristics of cited articles, and explore citation patterns in relation to review conclusions.

DESIGN

A systematic search identified reviews published from January 2010 to March 2020. Different characteristics (for example, type of review or research, journal impact factor, conclusions) were extracted from the reviews and cited articles. Logistic regression was used to estimate likelihood of articles with particular characteristics being cited in reviews. A qualitative network analysis linked reviews sub-grouped by conclusions with the types of articles they cited.

MAIN OUTCOME MEASURES

(OR; 95% CI) for likelihood that articles with particular characteristics were cited as evidence in reviews.

RESULTS

From 33 reviews identified, 183 different articles were cited (including other reviews). Narrative reviews were 62% less likely to be cited than systematic reviews with meta-analysis (OR 0.38; 0.16 to 0.86; p=0.03). Likelihood of being cited was higher for evidence on children than adults (OR 2.27; 1.59 to 3.25; p<0.0001), and with increased journal impact factor (OR 1.15; 1.00 to 1.31; p=0.04). No other factors were statistically significant in the main analysis, and few factors were significant in subgroup analyses. Network analysis showed that reviews concluding a beneficial relationship of LES with BW cited mainly randomised controlled trials, whereas reviews concluding an adverse relationship cited mainly observational studies.

CONCLUSIONS

Overall reference to the available evidence across reviews appears largely arbitrary, making citation bias likely. Differences in the conclusions of individual reviews map onto different types of evidence cited. Overall, inconsistent and selective use of the available evidence may account for the diversity of conclusions in reviews on LES and BW.

TRIAL REGISTRATION NUMBER

Prior to data analysis, the protocol was registered with the Open Science Framework (https://osf.io/9ghws).

Evidence for Citation Networks in Studies of Free-Roaming Cats: A Case Study Using Literature on Trap-Neuter-Return (TNR)

Simple Summary

All aspects of an argument need to be evaluated in evidence-based scientific pursuit to prevent the development of self-reinforcing cohorts of studies through the cross-referencing of selected subsets of studies (citation bias). Controling populations of unowned cats using Trap–Neuter–Return and its variants (hereafter TNR) is an example of how a scientific topic develops and is communicated academically and publicly. We found 145 TNR publications between 2002–2019, with the last two years seeing a rapid increase in studies. Publication clusters occur, focusing on: population control, interactions with wildlife, disease transmission (including implications for pet cats, wildlife and humans), free-roaming cats, and feral and domestic cat management. A quarter of all studies were published in Open Access formats, which have the highest potential social impact because they are free to download, share and read. While it is encouraging that diverse views are expressed, the development of clusters of introspective topics risks reduced dialogue between groups of authors with differing views. Journal editors could encourage exchanges between groups by choosing reviewers from different camps to assess manuscripts and by asking authors to acknowledge alternative views.

Abstract

Trap–Neuter–Return and its variants (hereafter TNR) aims to control unowned cat populations. Papers on this topic form a useful case study of how how an area of literature grows, papers become influential, and citation networks form, influencing future study as well as public perceptions of the science. We analysed 145 TNR studies published 2002–2019. Common topics, identified by frequently used language, were population control, interactions with wildlife, disease transmission (including implications for pets, wildlife and humans), free-roaming cats, and feral and domestic cat management. One or more papers on each of these topics was judged influential because of high citations overall, high average citations/year, or frequent mentions in social media. Open Access papers were more influential in social media, raising greater public awareness than studies published in journals that were less accessible. While divergent views exist on a range of topics, the network analysis of the TNR literature indicated potential for forming self-reinforcing groups of authors. While it is encouraging that diverse views are expressed, there is a risk of reduced dialogue interactions between groups, potentially constraining dialogue to refine arguments, share information, or plan research. Journal editors could encourage communication by choosing reviewers from different camps to assess manuscripts and by asking authors to acknowledge alternative views.

Dietary saturated fat and heart disease: a narrative review

The American Heart Association (AHA) recently published a meta-analysis that confirmed their 60-year-old recommendation to limit saturated fat (SFA, saturated fatty acid) and replace it with polyunsaturated fat to reduce the risk of heart disease based on the strength of 4 Core Trials. To assess the evidence for this recommendation, meta-analyses on the effect of SFA consumption on heart disease outcomes were reviewed. Nineteen meta-analyses addressing this topic were identified: 9 observational studies and 10 randomized controlled trials. Meta-analyses of observational studies found no association between SFA intake and heart disease, while meta-analyses of randomized controlled trials were inconsistent but tended to show a lack of an association. The inconsistency seems to have been mediated by the differing clinical trials included. For example, the AHA meta-analysis only included 4 trials (the Core Trials), and those trials contained design and methodological flaws and did not meet all the predefined inclusion criteria. The AHA stance regarding the strength of the evidence for the recommendation to limit SFAs for heart disease prevention may be overstated and in need of reevaluation.

Co-Authorship Network Analysis of Prevention Research Centers: An Exploratory Study

OBJECTIVE

The Centers for Disease Control and Prevention's Prevention Research Centers (PRCs) collaborate on public health activities with community agencies and organizations. We evaluated these collaborations by studying the relationships between co-authors from the PRCs and community agencies that published at least 1 article together in the first year of the program.

METHODS

We identified all the authors of articles published by PRCs and collaborating members in peer-reviewed journals between September 2014 and September 2015 and constructed a network showing the links between and among all the authors. We characterized the network with 4 measures of social structure (network components, network density, average clustering coefficient, average distance) and 3 measures of individual author performances (degree-, betweenness-, and closeness-centrality).

RESULTS

The 413 articles had 1804 individual authors and 7995 co-authorship relationships (links) in 212 peer-reviewed journals. These authors and co-authors formed 44 separate, nonoverlapping groups (components). The largest "giant" component containing most of the links involved 66.3% (n = 1196) of the authors and 73.7% (n = 5889) of the links. We identified 136 "information brokers" (authors with high closeness centrality: those who have the shortest links to the most authors). Two authors with high betweenness centrality (who had the highest number of co-authors; 104 and 107) had the greatest ability to mediate co-authorships. Network density was low; only 0.5% of all potential co-authorships were realized (7995 actual co-authorship/1 628 110 potential co-authorships).

CONCLUSION

Information brokers and co-authorship mediators should be encouraged to communicate more with each other to increase the number of collaborations between network members and, hence, the number of co-authorships.

Selective citation in the literature on the hygiene hypothesis: a citation analysis on the association between infections and rhinitis

OBJECTIVE

Our objective was to assess the occurrence and determinants of selective citation in scientific publications on Strachan's original hygiene hypothesis. His hypothesis states that lack of exposure to infections in early childhood increases the risk of rhinitis.

SETTING

Web of Science Core Collection.

PARTICIPANTS

We identified 110 publications in this network, consisting of 5551 potential citations.

PRIMARY AND SECONDARY OUTCOME MEASURES

Whether a citation occurs or not, measured and analysed according to the preregistered protocol.

RESULTS

We found evidence for citation bias in this field: publications supportive of the hypothesis were cited more often than non-supportive publications (OR adjusted for study design [adjOR] 2.2, 95% CI 1.6 to 3.1), and the same was the case for publications with mixed findings (adjOR 3.1, 95% CI 2.2 to 4.5). Other relevant determinants for citation were type of exposure, specificity, journal impact factor, authority and self-citation. Surprisingly, prospective cohort studies were cited less often than other empirical studies.

CONCLUSIONS

There is clear evidence for selective citation in this research field, and particularly for citation bias.