FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2419302359> ?p ?o ?g. }

• W2419302359 endingPage "16" @default.
• W2419302359 startingPage "14" @default.
• W2419302359 abstract "Our interest in improving the quality of ecotoxicology research arose primarily as a consequence of reading many research papers that report what seemed to us to be surprising results. Although there are many reasons why reported effects of a chemical could be surprising, our attention was drawn particularly to reports of ubiquitous pollutants of the aquatic environment causing adverse effects to various aquatic species at concentrations at, or even below, those present in the aquatic environment (i.e., in a typical river); a very small selection of these papers is presented in Table 1. For example, the plasticizer di-(2-ethylhexyl)-phthalate, a chemical produced in very high amounts and present in the aquatic environment worldwide, has been reported to impair reproduction of fish at a concentration of 20 ng/L, which was the lowest concentration tested out of many 1, yet it is often reported to be present in the aquatic environment at micrograms per liter concentrations. This poses the question; How are fish managing to reproduce if environmental concentrations of di-(2-ethylhexyl)-phthalate are much higher than those reported to adversely affect the ability of fish to reproduce? The same general question can be asked of the effects of many chemicals stated to occur at environmentally relevant concentrations. The lowest concentration of a chemical having a significant biological effect that we are aware of is 30 pg/L; Franzellitti et al. recently reported that this concentration of an antidepressant, fluoxetine, had a wide range of effects on marine mussels 2. It is important to emphasize that we are not claiming that these reported effects are wrong, that is, are not repeatable. However, we do not know whether or not they are repeatable, and we think that it is important to establish repeatability, particularly when extremely low effect concentrations are observed, before using such data in shaping regulatory guidelines. This recent phenomenon of chemicals apparently causing adverse effects at extremely low concentrations has been noted by others. For example, Ford has correctly pointed out, when summarizing the results of a series of papers on the effects of antidepressant pharmaceuticals on aquatic organisms, that “recently, an increasing number of studies has raised concerns over the impact of antidepressants in the aquatic environment due to effects on aquatic organisms (both invertebrates and vertebrates) being observed at very low and environmentally relevant concentrations … even down into the pg/L range” 3. Further, before concluding that these reported effects at very low concentrations cannot be correct, we should remember that some chemicals definitely can adversely affect aquatic organisms when present at extremely low, even sub–nanogram per liter, concentrations; ethinyl estradiol and tributyl tin are examples. The central tenet of science is its repeatability. Before publishing results, scientists should ensure that their results are repeatable. Doing so would usually entail repeating an experiment. However, our preliminary, and currently unpublished, analysis of recently published ecotoxicology papers shows that the majority of papers present results from only 1 experiment. Not only do the original researchers rarely repeat an experiment to assess the robustness of the results, but other researchers rarely repeat them as well. In many cases, it would be very difficult, if not impossible, to repeat an experiment conducted by another group of scientists because insufficient details are provided to ensure that all elements of the initial experiment are faithfully repeated. It is also a well-recognized phenomenon that studies repeating previously reported work are less likely to be accepted for publication in scientific journals than are those reporting “novel” data. We are aware of only 1 explicit attempt to independently repeat a surprising ecotoxicological finding, the results of which were worrying. Three research groups, working independently, reported that the antiinflammatory pharmaceutical diclofenac had adverse effects on fish at low microgram per liter concentrations. Particularly prominent were effects on the liver and kidneys. However, when comprehensive attempts were made by another group of scientists to repeat these findings, they were not repeated (see that paper for citations to the original articles) 4. When the histological slides (of kidney and liver) from 2 of the 3 original studies, plus those from the repeat study, were reanalyzed, blind, by an independent group of fish pathologists, the original finding of adverse effects was not upheld 5. The panel of experts who reanalyzed the slides concluded that it “found little evidence to support other reported effects of diclofenac in trout, and thus the overall NOEC was determined to be >320 µg/L” (no observed effect concentration). Although further research is required to unequivocally establish what concentrations of diclofenac cause what effects, this admittedly single example provides an insight into the potential repeatability of at least some ecotoxicology research. In contrast to the situation with ecotoxicology, numerous efforts have been made recently to assess the reproducibility of the biomedical literature. The results have not been encouraging; a significant proportion of published biomedical research seems to not be repeatable 6 for a wide variety of reasons including poor experimental design, inadequate and incorrect analysis of results, biased analysis, and incomplete reporting of results. Greater scrutiny of results and conclusions published in high-profile journals has led to increasing numbers of significant corrections and outright retractions of papers. Such close scrutiny of papers does not occur yet in the ecotoxicology field, which probably helps explain why, as far as we are aware, there have not been any retractions from ecotoxicology journals. We consider that the quality of published ecotoxicology research is not as good as it could and should be. Although there are many possible explanations for this (e.g., an increasingly highly competitive employment market in this area as well as the reluctance of scientific journals to publish negative or repeat data combined with the pressure to publish high volumes of research to obtain funding), there are no excuses. Poor-quality research will lead to poor-quality decision making by regulators and, hence, poor protection of the environment. If our job as academic researchers is to conduct research that is robust enough such that it could be used to help protect the environment, we should be doing it as well as it can be done. It is clear that this is not always the case. In an attempt to help researchers improve the quality of their research, we recently proposed 12 basic principles that, if followed, should lead to better research 7, which would be of more help and use to regulators. In particular, we hope that following those principles will help young researchers as they become independent scientists. We were fortunate enough—or lucky enough—to have first class scientists as supervisors and mentors when we were learning how to do research at the beginning of our careers. Better training of young researchers, including an emphasis on scientific integrity, and reduced pressure on academic scientists to publish would help greatly in improving the quality of ecotoxicology research. John P. Sumpter Catherine A. Harris Institute of Environment, Health and Societies Brunel University London, UK" @default.
• W2419302359 created "2016-06-24" @default.
• W2419302359 creator A5015001678 @default.
• W2419302359 creator A5060826690 @default.
• W2419302359 date "2015-12-21" @default.
• W2419302359 modified "2023-09-26" @default.
• W2419302359 title "<i>In Response</i>: An academic perspective" @default.
• W2419302359 cites W1972311809 @default.
• W2419302359 cites W2003679016 @default.
• W2419302359 cites W2019456977 @default.
• W2419302359 cites W2029458910 @default.
• W2419302359 cites W2034159505 @default.
• W2419302359 cites W2084594774 @default.
• W2419302359 cites W2089757970 @default.
• W2419302359 cites W2093225920 @default.
• W2419302359 cites W2144981148 @default.
• W2419302359 cites W2148568928 @default.
• W2419302359 cites W2157997539 @default.
• W2419302359 cites W282628544 @default.
• W2419302359 doi "https://doi.org/10.1002/etc.3195" @default.
• W2419302359 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/26771347" @default.
• W2419302359 hasPublicationYear "2015" @default.
• W2419302359 type Work @default.
• W2419302359 sameAs 2419302359 @default.
• W2419302359 citedByCount "2" @default.
• W2419302359 countsByYear W24193023592022 @default.
• W2419302359 countsByYear W24193023592023 @default.
• W2419302359 crossrefType "journal-article" @default.
• W2419302359 hasAuthorship W2419302359A5015001678 @default.
• W2419302359 hasAuthorship W2419302359A5060826690 @default.
• W2419302359 hasBestOaLocation W24193023591 @default.
• W2419302359 hasConcept C12713177 @default.
• W2419302359 hasConcept C154945302 @default.
• W2419302359 hasConcept C15744967 @default.
• W2419302359 hasConcept C41008148 @default.
• W2419302359 hasConcept C86803240 @default.
• W2419302359 hasConceptScore W2419302359C12713177 @default.
• W2419302359 hasConceptScore W2419302359C154945302 @default.
• W2419302359 hasConceptScore W2419302359C15744967 @default.
• W2419302359 hasConceptScore W2419302359C41008148 @default.
• W2419302359 hasConceptScore W2419302359C86803240 @default.
• W2419302359 hasIssue "1" @default.
• W2419302359 hasLocation W24193023591 @default.
• W2419302359 hasLocation W24193023592 @default.
• W2419302359 hasOpenAccess W2419302359 @default.
• W2419302359 hasPrimaryLocation W24193023591 @default.
• W2419302359 hasRelatedWork W1990804418 @default.
• W2419302359 hasRelatedWork W2028917246 @default.
• W2419302359 hasRelatedWork W2082860237 @default.
• W2419302359 hasRelatedWork W2095322754 @default.
• W2419302359 hasRelatedWork W2098170879 @default.
• W2419302359 hasRelatedWork W2130076355 @default.
• W2419302359 hasRelatedWork W2605852498 @default.
• W2419302359 hasRelatedWork W2748952813 @default.
• W2419302359 hasRelatedWork W2899084033 @default.
• W2419302359 hasRelatedWork W4250032403 @default.
• W2419302359 hasVolume "35" @default.
• W2419302359 isParatext "false" @default.
• W2419302359 isRetracted "false" @default.
• W2419302359 magId "2419302359" @default.
• W2419302359 workType "article" @default.