Scientometrics Approach to Research in Ovine Mastitis from 1970 to 2019 (With a Complete List of Relevant Literature References)

Home , Abstract (summary), Scientometrics, Scientometrics (journal)

pathogens

Communication Scientometrics Approach to Research in Ovine Mastitis from 1970 to 2019 (with a Complete List of Relevant Literature References)

Daphne T. Lianou and George C. Fthenakis * Veterinary Faculty, University of Thessaly, 43100 Karditsa, Greece; [email protected] * Correspondence: [email protected] Received: 24 May 2020; Accepted: 16 July 2020; Published: 17 July 2020

Abstract: The present study is a scientometrics evaluation of refereed publications on bacterial mastitis in sheep; the objectives were the evaluation of the relevant papers and the presentation of quantitative characteristics regarding their scientific content and bibliometric details. The Web of Science platform was used with search terms: [mastitis OR *mammary infection*] AND [sheep OR ewe* OR ovine] for papers from 1970 to 2019; only ‘articles’, ‘reviews’, ‘proceedings papers’, or ‘data papers’ were evaluated, whilst documents related solely to contagious agalactia, mammary aspects of lentiviral infections, or infections of the teats and the udder skin were excluded. Finally, 580 papers were considered in detail. The number of published papers increased from 8 during the 1970s to 273 during the 2010s. These papers originated from 43 countries (most from Greece or Spain, n = 87 from each) and 240 institutions (145 universities and 95 other establishments), of which 35 produced 5 papers each. Most papers present original studies (n = 539) with a few reviews (n = 41). ≥ The original papers refer to dairy (n = 428), meat (n = 113), or wool (n = 1) production systems and present field (n = 329), laboratory (n = 163), or experimental (n = 67) work; the papers report aetiology (n = 146), risk factors (n = 100), pathogenesis (n = 92), diagnosis (n = 88), effects (n = 66), treatment (n = 50), control (n = 36), or descriptive epidemiology (n = 32) of the disease. Papers related to dairy production present more field and fewer experimental work than papers related to meat production; also, in papers describing work performed in dairy sheep, studies about aetiology, risk factors, and diagnosis of the disease predominate, whilst in papers performed in meat sheep, studies about aetiology, pathogenesis, and effects/diagnosis are reported more often. The papers were published in 175 scientific journals (most in Small Ruminant Research, n = 90, or Journal of Dairy Science, n = 54). On average, the papers received 16.8 total citations and 1.6 yearly citations (h-index = 47). Most papers were published in Scimago classification Q1 (n = 240) or Q2 (n = 230) journals and received 23.4 or 15.4 total citations, respectively. Reviews received more citations than original papers; among the latter, papers with work referring to dairy production received more yearly citations than papers referring to meat production; no differences in citations were seen according to type of work or mastitis aspect covered. Most citations were received by papers from France. Papers published in Journal of Dairy Science or Small Ruminant Research received the most citations. In total, there were 1558 individual authors of the papers, with 24 authors having co-authored > 10 papers each (max: 73 papers); on average, there were 5.2 co-authors per paper (min–max: 1–25). Average number of co-authors progressively increased from 2.1 in the 1970s to 6.3 in the 2010s, with original papers having a higher number of co-authors than reviews: 5.3 and 3.7, respectively. Papers from France had highers number of co-authors (7.9). The findings of this first ever scientometrics study into ovine mastitis indicate that the disease has not been studied as other sheep diseases and that future studies in it should be directed to its control.

Keywords: ewe; intramammary infection; mastitis; meta-research; ovine; scientometrics; sheep; subclinical mastitis

Pathogens 2020, 9, 585; doi:10.3390/pathogens9070585 www.mdpi.com/journal/pathogens Pathogens 2020, 9, 585 2 of 13

1. Introduction Scientometrics refers to the study of measuring and analysing scientific literature. Specific topics within this field refer to the measurement of the impact of research papers, the understanding of scientific citations, and the use of the results of such assessments in policy and management contexts [1]. There are overlaps between scientometrics and other scientific fields, e.g., meta-science. Meta-science can be defined as the use of research methods to study and appraise research itself and the areas where improvements can be made; meta-science studies can deal with methods, reporting, reproducibility, evaluation, and incentives, which are related with performance, communication, verification, evaluation, and rewarding research, respectively [2]. Scientometrics and meta-science papers produce new data by using information in previously published research articles. Such articles present features and characteristics of scientific research, with the aim to produce a quantitative assessment of the initial research papers. They differ from review papers, which summarise the state of knowledge on a topic, without reporting new facts or carrying out new analyses. During assessment of the relevant literature, authors of reviews may discuss research papers and provide expert opinions by combining findings and ideas from the primary sources. In sheep, mammary infections can lead to bacterial mastitis (“mastitis”), mycoplasma mastitis (contagious agalactia), and lentiviral mammary infection. Among these, bacterial mastitis (“mastitis”) is the most important. It causes significant adverse financial consequences in sheep flocks and also reduces standards of welfare of sheep [3,4]. In recent years, various review papers on ovine mastitis have been published, presenting an overview or specific aspects of the disease. These reviews do not provide quantitative data of the relevant publications, which in turn could provide further insight into the disease. In fact, a scientometrics study of ovine mastitis has never been published. The present study is a scientometrics evaluation of publications on bacterial mastitis (“mastitis”) in sheep during the last 50 years, from 1970 to 2019. Objectives of the study were the evaluation of the relevant publications and the presentation of quantitative characteristics regarding their scientific content and bibliometric details.

2. Materials and Methods

2.1. Search Procedure The Web of Science platform (www.webofknowledge.com; Clarivate Analytics, Philadelphia, USA) was used for the search of relevant publications. Only publications in this platform were assessed and included in the study. Search terms employed were: [mastitis OR *mammary infection*] AND [sheep OR ewe* OR ovine]. Timespan was set to ‘Custom year range (1970–2019)’. The following Web of Science collections were included in the search: (i) Science citation index expanded 1970–2019 (1970 being the earliest year for which relevant information is provided by the platform) and (ii) Emerging sources citation index 2015–2019. The initial search produced 1379 records. Document analysis of these was performed and the following types of documents were excluded: ‘meeting abstracts’, ‘notes’, ‘letters’, ‘editorial material’, ‘news items’, ‘corrections’, ‘early access papers’, and ‘reprints’. After their exclusion, a total of 1326 records remained, classiﬁed as ‘articles’ (n = 1250), ‘reviews’ (n = 76), ‘proceedings papers’ (n = 43), and ‘data papers’ (n = 1), for further assessment. The abstracts of all these 1326 records were read and evaluated in the platform. Papers that did not deal with mastitis in sheep (e.g., papers not including sheep work, papers not including work on mammary diseases or their causal organisms) were excluded. Further, papers related solely to contagious agalactia or the mammary aspects of lentiviral infections were also excluded. Moreover, papers referring solely to infections of the teats and the udder skin (e.g., Papilloma virus infection of teats, contagious ecthyma, staphylococcal dermatitis, sheep pox) with no reference or association to bacterial mastitis were also excluded. Pathogens 2020, 9, 585 3 of 13

2.2. Paper Evaluation After the above abstract review, 580 papers were considered for detailed evaluation. All these were individually assessed. In each paper, the following details were recorded.

Year of publication. • Journal in which it was published. • Country and scientific establishment of origin (the country(ies) and the scientific establishment(s) • of only the first and the last authors were taken into account). Number of literature references included in the relevant list. • Total number of citations received by the paper until the end of 2019. • Number and names of all co-authors. • Type of paper: (i) original (paper presenting and providing new information, e.g., results or • analyses, including case reports) or (ii) review (paper surveying and summarizing previously published studies, with no presentation and report of new facts or analyses). For original papers, the following details were also recorded.

I Sheep production system to which the work referred: (i) dairy, (ii) meat, or (iii) wool; all systems that applied were assigned per paper.

I Type of study described in the paper: (i) experimental (study performed predominantly with animals and involving experimental work, e.g., bacterial inoculations, allocation of animals to groups), (ii) ﬁeld (study performed predominantly with animals and not involving experimental work, e.g., clinical work, survey work, farm work), or (iii) laboratory (study performed predominantly without animals, although material on which it was based would have originated from animals); only one of these three types was assigned per paper.

I Mastitis aspect described in the paper: (i) aetiology, (ii) control, (iii) descriptive epidemiology, (iv) diagnosis, (v) eﬀects, (vi) pathogenesis, (vii) risk factors, or (viii) treatment; up to two of these were assigned per paper.

2.3. Data Management and Analysis All data were entered into Microsoft Excel. Descriptive analysis was performed initially. The frequency of the various outcomes was evaluated in tables of cross-categorised frequency data by use of the Pearson chi-square test as appropriate. Comparisons between continuous data were performed by use of one-way analysis of variance. Correlations were performed as indicated and significance of the result was evaluated. Statistical significance was defined at p < 0.05.

3. Results All the 580 papers individually assessed are indexed in the Web of Science, fulﬁl the search criteria, and present work on ovine mastitis. A complete list of these papers with their details is in Table S1.

3.1. Year of Publication and Origin of Papers The number of papers published per decade in the topic increased progressively: 8 (1.4%) papers were published in the 1970s, 22 (3.8%) in the 1980s, 109 (18.8%) in the 1990s, 168 (28.9%) in the 2000s, and 273 (47.1%) in the 2010s (Figure1). Pathogens 2020, 9, 585 4 of 13 Pathogens 2020, 9, x FOR PEER REVIEW 4 of 13 Pathogens 2020, 9, x FOR PEER REVIEW 4 of 13

FigureFigure 1. NumberNumber of of papers papers published published on on mastitis mastitis in in sh sheepeep from from 1970 1970 to to 2019, 2019, classified classified according according to to Figure 1. Number of papers published on mastitis in sheep from 1970 to 2019, classified according to thethe decade decade during during which which they they were published. the decade during which they were published. Papers originated from 43 different countries (Table S2). Most papers originated from Greece Papers originated from 43 different countries (Table S2). Most papers originated from Greece or or SpainPapers (n originated= 87 from from each, 43 15.0%), different Italy countries (n = 78, (Tab 13.4%),le S2). UnitedMost papers Kingdom originated (n = 50, from 8.6%), Greece Brazil or Spain (n = 87 from each, 15.0%), Italy (n = 78, 13.4%), United Kingdom (n = 50, 8.6%), Brazil (n = 40, Spain(n = 40, (n 6.9%)= 87 from, France each, (n 15.0%),= 31, 5.3%), Italy United (n = 78, States 13.4%), of AmericaUnited Kingdom (n = 26, 4.5%), (n = 50, Australia 8.6%), Brazil (n = 24, (n 4.1%),= 40, 6.9%), France (n = 31, 5.3%), United States of America (n = 26, 4.5%), Australia (n = 24, 4.1%), Israel (n 6.9%),Israel France (n = 23, (n 4.0%), = 31, and5.3%), Austria United or States Turkey of (nAmeric= 17a from (n = each,26, 4.5%), n = 2.9%). Australia Three (n hundred= 24, 4.1%), two Israel papers (n = 23, 4.0%), and Austria or Turkey (n = 17 from each, n = 2.9%). Three hundred two papers originated =originated 23, 4.0%), and from Austria the first or four Turkey countries (n = 17 above from ea (52.1%ch, n = of 2.9%). all papers), Three whilehundred 480 two papers papers were originated from the from the first four countries above (52.1% of all papers), while 480 papers were from the above 11 fromabove the 11 first countries four countries (82.8%). above When (52.1% geographical of all papers), regions while were 480 considered, papers were it was from found the thatabove most 11 countries (82.8%). When geographical regions were considered, it was found that most papers countriespapers originated (82.8%). fromWhen the geographical countries of Southernregions Europewere considered, (n = 303, 52.2% it was of allfound papers), that the most countries papers of originated from the countries of Southern Europe (n = 303, 52.2% of all papers), the countries of originatedCentral and from Northern the countries Europe (nof =Southern85, 14.7%), Europe and the (n countries= 303, 52.2% of the of Middleall papers), East (nthe= 63,countries 10.9%). of Central and Northern Europe (n = 85, 14.7%), and the countries of the Middle East (n = 63, 10.9%). CentralDi ffanderences Northern in the Europe year of (n publication = 85, 14.7%), are and significant the countries between of the the Middle above 11East countries (n = 63, (10.9%).p < 0.001). Differences in the year of publication are significant between the above 11 countries (p < 0.001). PapersDifferences from Brazil, in the Greece, year andof publication Italy are the are most significant recent ones between (average the above year of 11 publication: countries (p 2013, < 0.001). 2011, Papers from Brazil, Greece, and Italy are the most recent ones (average year of publication: 2013, 2011, Papersand 2011, from respectively), Brazil, Greece, whilst and papersItaly are from the most Australia recent and ones the (average United Statesyear of of publication: America are 2013, the oldest2011, and 2011, respectively), whilst papers from Australia and the United States of America are the oldest andones 2011, (average respectively), year of publication: whilst papers 2000 from and Australia 1997, respectively) and the United (Figure States2). of America are the oldest ones (average year of publication: 2000 and 1997, respectively) (Figure 2). ones (average year of publication: 2000 and 1997, respectively) (Figure 2).

Figure 2. Average year of publication of papers on mastitis in sheep from 1970 to 2019 from the 11 Figure 2. Average year of publication of papers on mastitis in sheep from 1970 to 2019 from the 11 Figurecountries 2. Average with the year most of papers publication published of papers (abbreviations on mastitis of countryin sheep names from according1970 to 2019 to Internationalfrom the 11 countries with the most papers published (abbreviations of country names according to International countriesNaming Conventionwith the most ISO papers 3166). published (abbreviations of country names according to International Naming Convention ISO 3166). Naming Convention ISO 3166). Papers originated from 240 diﬀerent scientiﬁc establishments around the world, of which 35 Papers originated from 240 different scientific establishments around the world, of which 35 publishedPapers atoriginated least 5 papers from each240 different (Table S3). scientific Of these establishments 35 establishments, around 31 the (88.6%) world, are of locatedwhich 35 in published at least 5 papers each (Table S3). Of these 35 establishments, 31 (88.6%) are located in the publishedthe 11 countries at least with5 papers highest each numbers(Table S3). of Of published these 35 papers.establishments, Among 31 all (88.6%) the establishments, are located in the 145 11 countries with highest numbers of published papers. Among all the establishments, 145 are 11are countries universities with (26 highest of these numbers produced of publishe5 papers),d papers. from whichAmong originated all the establishments, 540 papers (93.1%); 145 are the universities (26 of these produced ≥ 5 papers),≥ from which originated 540 papers (93.1%); the other universitiesother 95 include (26 of research these produced institutes, ≥ services5 papers), of Ministriesfrom which of originated Agriculture, 540 pharmaceutical papers (93.1%); laboratories, the other 95 include research institutes, services of Ministries of Agriculture, pharmaceutical laboratories, 95veterinary include practices,research institutes, etc. (9 of theseservices produced of Ministries5 papers), of Agriculture, from which pharmaceutical originated 236 papers laboratories, (40.7%) veterinary practices, etc. (9 of these produced ≥≥ 5 papers), from which originated 236 papers (40.7%) veterinary practices, etc. (9 of these produced ≥ 5 papers), from which originated 236 papers (40.7%) (p < 0.0001). The establishments that produced more papers are the University of Thessaly (n = 64; (p < 0.0001). The establishments that produced more papers are the University of Thessaly (n = 64;

Pathogens 2020, 9, 585 5 of 13 Pathogens 2020, 9, x FOR PEER REVIEW 5 of 13

11.0%),(p < 0.0001).the University The establishments of Leon, the University that produced of Lo morendon papers(n = 25 each; are the 4.3%), University the Aristotle of Thessaly University (n = 64; of 11.0%),Thessaloniki, the University the National of Leon, Institute the University of Agronomic of London Research (n = 25 each;France 4.3%), (n the= 21 Aristotle each; 3.6%), University the of ComplutenseThessaloniki, University the National of Madrid, Institute ofthe Agronomic Veterinary Research University France of (nVienna= 21 each; (n = 3.6%),18 each; the Complutense3.1%), the KimronUniversity Veterinary of Madrid, Institute the VeterinaryIsrael (n University= 17; 2.9%), of the Vienna Commonwealth (n = 18 each; 3.1%),Scientific the Kimronand Industrial Veterinary ResearchInstitute Organisation Israel (n = Australia17; 2.9%), (n the = 16; Commonwealth 2.8%), and the ScientificUniversity and of Palermo Industrial (n Research = 15; 2.6%). Organisation Australia (n = 16; 2.8%), and the University of Palermo (n = 15; 2.6%). 3.2. Type and Content of Papers 3.2. Type and Content of Papers Most papers presented original studies (n = 539; 92.9%), and fewer ones presented reviews (n = Most papers presented original studies (n = 539; 92.9%), and fewer ones presented reviews (n = 41; 41; 6.1%). Most original papers referred to dairy production (n = 428; 79.4%), with fewer ones referring 6.1%). Most original papers referred to dairy production (n = 428; 79.4%), with fewer ones referring to to meat (n = 113; 21.0%) or wool (n = 1) production. Most original papers presented field work (n = meat (n = 113; 21.0%) or wool (n = 1) production. Most original papers presented field work (n = 329; 329; 61.0%), and fewer ones presented laboratory (n = 143; 26.5%) or experimental (n = 67; 12.4%) 61.0%), and fewer ones presented laboratory (n = 143; 26.5%) or experimental (n = 67; 12.4%) work. work. Original studies reported mostly aetiology (n = 146; 27.1%), risk factors (n = 100; 18.6%), Original studies reported mostly aetiology (n = 146; 27.1%), risk factors (n = 100; 18.6%), pathogenesis pathogenesis (n = 93; 17.3%), or diagnosis (n = 89; 16.8%) of the disease, whilst fewer papers existed (n = 93; 17.3%), or diagnosis (n = 89; 16.8%) of the disease, whilst fewer papers existed on descriptive on descriptive epidemiology (n = 32; 5.9%), control (n = 36; 6.7%), treatment (n = 50; 9.3%), or effects epidemiology (n = 32; 5.9%), control (n = 36; 6.7%), treatment (n = 50; 9.3%), or effects (n = 66; 12.2%) (n = 66; 12.2%) of the disease (Figures 3 and 4, Table S4). of the disease (Figures3 and4, Table S4).

Figure 3. Frequency of published papers on mastitis in sheep from 1970 to 2019 according to type and Figurecontent 3. Frequency (top left: of types published of papers; paperstop on right mastitis: types in of sheep production from 1970 system to 2019 to which according the original to type and papers contentreferred; (top bottomleft: types left of: types papers; of studies top right in: original types of papers; productionbottom system right :to mastitis which the aspects original reported papers in the referred;original bottom papers—colour left: types legendof studies for eachin original pie diagram papers; to bottom its right). right: mastitis aspects reported in the original papers—colour legend for each pie diagram to its right).

Pathogens 2020, 9, 585 6 of 13 Pathogens 2020, 9, x FOR PEER REVIEW 6 of 13

Figure 4. AssociationAssociation between between type type of of study, study, production system referred to, and mastitis aspect in papers on ovine mastitis published during the pe periodriod 1970–2019 (Study type: 1 1 experimental, experimental, 2 2 field, field, 3 laboratory; Production system: 1 1 dairy, dairy, 2 meat, 3 wool; Mastitis aspect: 1 1 aetiology, aetiology, 2 control, 3 epidemiology, 4 diagnosis, 5 eeffects,ffects, 6 pathogenesis,pathogenesis, 7 risk factors, 8 treatment.treatment. Size of the spheres reflectsreflects number of papers at each point).

There are clear differences differences in the type of work presented in papers according to the production system to which the study referred. Papers Papers relate relatedd to dairy production pr presentedesented proportionately more fieldfield and and fewer fewer experimental experimental work work than papersthan papers related related to meat production;to meat production; for dairy production,for dairy production,the number the of papersnumber presenting of papers pr field,esenting laboratory, field, laboratory, and experimental and experimental work were work 272, were 113, 272, and 113, 43, andrespectively 43, respectively (63.6%, 26.4%,(63.6%, and 26.4%, 10.0%), and whilst 10.0%), respective whilst respective figures for figures papers for related papers to meat related production to meat productionsheep were sheep 58, 31, were and 2458, (51.3%, 31, and 27.4%, 24 (51.3%, and 21.2%)27.4%, (andp = 0.0036).21.2%) (p = 0.0036). There were also some differences differences in the mastitis aspect reported in the papers according to the production system. In In papers papers performed performed in in dairy dairy sheep, sheep, studies studies about about aetiology, aetiology, risk factors, and diagnosis predominated (n = 111,111, 86, 86, and 74, respectively; 25.9%, 20.1%, 17.3%), whilst in papers performed in meat sheep studies about about aetiology, aetiology, pathogenesis, and effects/diagnosis effects/diagnosis were reported more often (n = 35,35, 26, 26, and and 15, 15, respectively respectively;; 31.0%, 31.0%, 23.0%, 23.0%, 13.3%) of the disease ( p = 0.044).0.044). There were also clear differences differences between the 11 countries with the most papers, in the type of paper published ( p = 0.0065),0.0065), in in the production system being referenced ( p < 0.0001),0.0001), in in the type of studies performed ( p < 0.0001),0.0001), and in the mastitis aspect presented (p << 0.0001). Similar Similar differences differences were also recordedrecorded whenwhen only only the the 6, 6, the the 4, 4, or or the the 3 countries3 countries with with the the most most papers papers were were included included in the in theanalysis analysis (p < (p0.008 < 0.008 inall in cases)all cases) (Table (Table S5). S5).

3.3. Journals Journals In total, total, the the 580 580 papers papers were were published published in 175 scientific scientiﬁc journals. Journals Journals in in which which over over 10 10 papers papers were published are are the the following: following: SmallSmall Ruminant Ruminant Research Research (n(n = =90,90, 15.5%), 15.5%), JournalJournal of Dairy of Dairy Science Science (n =(n 54,= 54, 9.3%), 9.3%), VeterinaryVeterinary Microbiology Microbiology (n(n = =29,29, 5.0%), 5.0%), JournalJournal of of Dairy Dairy Research Research (n(n == 22,22, 3.8%), Pesquisa VeterinVeterináriaária Brasileira and The Veterinary Record (n(n == 14 each, 2.4%), The Veterinary Journal (including British Veterinary Journal), and Journal of Veterinary Medicine B (including Zentralblatt für

Pathogens 2020, 9, 585 7 of 13 Pathogens 2020, 9, x FOR PEER REVIEW 7 of 13

VeterinärmedizinBritish Veterinary Reihe Journal B)), (n and = 11Journal each, 1.9%); of Veterinary cumulatively, Medicine 245 B (including papers (42.2%)Zentralblatt were fürpublished Veterinärmedizin in these 8Reihe journals. B) (n = 11 each, 1.9%); cumulatively, 245 papers (42.2%) were published in these 8 journals. InIn total, total, 225 225 of of the the 245 245 papers papers in in these these journals journals (91.8%) (91.8%) originated originated from from 10 10 of of the the 11 11 countries countries withwith the the most most papers papers (i.e., (i.e., except except Turkey); Turkey); in in contrast, contrast, if if all all articles articles are are taken taken into into account, account, a a smaller smaller proportionproportion (463 (463 of of 580; 580; 79.8%) 79.8%) originated originated from from these these 10 countries ( p < 0.0001s).0.0001s). There There was was no no further further patternpattern ofof statisticalstatistical association association between between these these 8 journals 8 journals and the and 11 countries the 11 countries with the most with publications the most publications(p > 0.95) (Figure (p > 0.95)5, Table (Figure S6). 5, Table S6).

FigureFigure 5. 5. AssociationAssociation between between journals journals and and countries countries of of origin origin of of papers papers on on ovine ovine mastitis mastitis published published duringduring the the period period 1970–2019 1970–2019 (abbreviations (abbreviations of of coun countrytry names names according according to to International International Naming Naming ConventionConvention ISO ISO 3166; 3166; abbreviations abbreviations of of journals journals from from left left to to right: right: JournalJournal of of Dairy Dairy Research Research, ,JournalJournal of of DairyDairy Science Science, ,JournalJournal of of Veterinary Veterinary Medicine Medicine B, B, Pesquisa Pesquisa Veterinária Veterinária Brasileira Brasileira, ,SmallSmall Ruminant Ruminant Research Research, , TheThe Veterinary Veterinary Journal Journal,, TheThe Veterinary Veterinary Record ,, Veterinary Microbiology ).

WhenWhen the the type type and and content content of of papers papers were were considered considered between between these these eight eight journals, journals, there there were were significantsignificant differences differences in in the the type type of of papers papers published published ( (pp == 0.045)0.045) as as well well as as in in the the production production system system beingbeing referenced, referenced, the the type type of of studies studies perfor performed,med, and and the the mastitis mastitis aspect aspect presented presented ( (pp << 0.0010.001 for for all all comparisons).comparisons). Similar Similar significant significant differences differences were were recorded recorded also also when when the the comparisons comparisons were were made made withwith the the findings findings of only the 4 or the 3 journals withwith thethe mostmost paperspapersincluded included in in the the analysis analysis (Figure (Figure6, 6,Table Table S7). S7). On average, the papers include 33.3 1.0 literature references (mean standard error of the On average, the papers include 33.3 ±± 1.0 literature references (mean ±± standard error of the mean).mean). Papers publishedpublished sincesince 2000 2000 included included significantly significantly more more references references than than those those published published earlier earlier(p < 0.0001) (p < 0.0001).. Review Review papers papers included included significantly significantly more references more references than did originalthan did papers, original 81.6 papers, versus 81.629.6, versus respectively 29.6, respectively (p < 0.0001); moreover,(p < 0.0001); papers moreover, describing papers laboratory describing work laboratory included morework referencesincluded morethan onesreferences describing than experimental ones describing work experime and fieldntal work: work 32.2, and 30.7, field and 28.2,work: respectively 32.2, 30.7, ( pand= 0.0310), 28.2, respectivelybut no differences (p = 0.0310), existed but between no differences papers according existed tobetw mastitiseen papers aspect thereinaccording (p = to0.48) mastitis (Table aspect S8). therein (p = 0.48) (Table S8).

Pathogens 2020, 9, 585 8 of 13 Pathogens 2020, 9, x FOR PEER REVIEW 8 of 13

Figure 6. Association between journals and aspect of mastitis covered in papers on ovine mastitis Figure 6. Association between journals and aspect of mastitis covered in papers on ovine mastitis published during the period 1970–2019 (abbreviations of journals from left to right: Journal of Dairy published during the period 1970–2019 (abbreviations of journals from left to right: Journal of Dairy Research, Journal of Dairy Science, Journal of Veterinary Medicine B, Pesquisa Veterinária Brasileira, Small Research, Journal of Dairy Science, Journal of Veterinary Medicine B, Pesquisa Veterinária Brasileira, Small Ruminant Research, The Veterinary Journal, The Veterinary Record, Veterinary Microbiology). Ruminant Research, The Veterinary Journal, The Veterinary Record, Veterinary Microbiology). 3.4. Impact of Papers 3.4. Impact of Papers The papers received (until the end of 2019) on average 16.8 1.0 total citations and 1.6 0.1 The papers received (until the end of 2019) on average 16.8 ± 1.0 total citations and 1.6 ±± 0.1 citations per year in the Web of Science platform. For all the papers, h-index was 47 and i10-index citationswas 280. per There year was in the a clear Web correlation of Science betweenplatform. the For number all the papers, of literature h-index references was 47 inand a paperi10-index and was the 280.number There of was citations a clear received correlation (r = 0.224 between for correlation the numb wither of theliterature total number references of citations, in a paperr = 0.456and the for numbercorrelation of citations with the received number of(r = citations 0.224 for per correlation year, p < 0.0001).with the Whentotal number original of and citations, review r papers = 0.456 were for correlationconsidered with separately, the number a correlation of citations was alsoper year, evident p < between0.0001). When references original and yearlyand review citations papers (r = were0.400 consideredfor original separately, papers and ar =correlation0.369 for reviews,was alsop ev< 0.011),ident between but not betweenreferences references and yearly and citations total citations (r = 0.400(r = 0.058 for original and r = papers0.242, respectively, and r = 0.369p >for0.055). reviews, p < 0.011), but not between references and total citationsWith (r regard= 0.058 to and Scimago r = 0.242, classification respectively, (year: p > 2019),0.055). most papers were published in Q1 (n = 240) or Q2 (nWith= 230) regard journals, to Scimago with fewer classification ones in Q3 (year: (n = 63)2019) or, Q4most (n =papers22) journals were publ or inished journals in Q1 outside (n = 240) this orclassification Q2 (n = 230) (n journals,= 25). Papers with fewer in Q1 ones journals in Q3 received (n = 63) more or Q4 citations (n = 22) andjournals more or yearly in journals citations outside than thispapers classification in Q2 or in (n Q3 =/ Q425). or Papers outside in theQ1 classificationjournals received journals: more 23.4 citations2.0 and and 2.3 more0.2, yearly 15.4 citations1.5 and ± ± ± than1.4 papers0.1, and in5.3 Q2 or0.5 in andQ3/Q4 0.5 or0.1, outside respectively the classification (p < 0.0001). journals: 23.4 ± 2.0 and 2.3 ± 0.2, 15.4 ± 1.5 ± ± ± and 1.4More ± 0.1, citations and 5.3 were ± 0.5 received and 0.5 ± by 0.1, papers respectively published (p < in 0.0001). the last 20 years (p < 0.0002). Most citations (totalMore citations citations per were paper, received total yearly by pape citationsrs published per paper, in the total last cites) 20 years were (p received< 0.0002). by Most papers citations with (totalan origin citations in France; per paper, these total were year publishedly citations mostly per paper, (74.0%) total in Q1cites) journals were received (p < 0.0005). by papers Also, with papers an originwith origins in France; from these establishments were published in France mostly (National (74.0%) Institute in Q1 journals of Agronomic (p < 0.0005). Research, Also, Universitypapers with of originsToulouse) from were establishments the ones with in most France total (National citations and Institute total yearlyof Agronomic citations; theseResearch, establishments University also of Toulouse)published were papers the mostly ones with in Q1 most journals total (citations80%). However,and total yearly establishments citations; outside these establishments France (University also ≥ publishedof Leon and papers University mostly of in Thessaly) Q1 journals were (≥80%). the ones However, with most establishments total cites (Table outside S9). France (University of LeonOn and average, University reviews of Thessaly) received more were total the ones citations with and most total total yearly cites citations(Table S9). per paper (35.4 9.9 ± andOn 3.6 average,0.7, respectively) reviews received than did more original total paperscitation (15.4s and total0.8 andyearly 1.4 citations0.1) (p per< 0.0001), paper (35.4 but fewer± 9.9 ± ± ± andtotal 3.6 cites ± 0.7, (1450 respectively) and 8296, than respectively). did original Papers papers with (15.4 work ± 0.8 referring and 1.4 to± 0.1) dairy (p production< 0.0001), but received fewer totalsignificantly cites (1450 more and yearly 8296, citations respecti thanvely). did Papers papers with referring work to meatreferring production, to dairy although production no di receivedfferences significantlywere evident more in the yearly distribution citations according than did to journalpapers classification referring to (Figuremeat production,7). With regard although to type no of differencesstudy, although were evident significantly in the more distribution experimental accordin org laboratory to journal (55.2%classification and 51.7%, (Figure respectively) 7). With regard than tofield type (35.3%) of study, works werealthough published significantly in Q1 journals more (pexperimental= 0.0006), no significantor laboratory differences (55.2% in and the citations51.7%, respectively)received between than papersfield (35.3%) of the works three typeswere werepublished found in (p Q1> 0.59).journals With (p regard= 0.0006), to mastitisno significant aspect, differencesdifferences in were the notcitations significant received in the between classification papers ofof thethe journals three types in which were papersfound ( werep > 0.59). published With regard(p = 0.067) to mastitis; also, di aspect,fferences differences in the average were not yearly significant citations in received the classification by papers of did the not journals differ accordingin which papersto the mastitiswere published aspect (p (p= =0.407). 0.067); Detailsalso, differences are in Table in S9.the average yearly citations received by papers did not differ according to the mastitis aspect (p = 0.407). Details are in Table S9.

Pathogens 2020, 9, 585 9 of 13 Pathogens 2020, 9, x FOR PEER REVIEW 9 of 13

Figure 7. Association between production system referred to in papers,papers, journal classiﬁcation,classification, and citations in papers on ovine mastitismastitis publishedpublished during the periodperiod 1970–20191970–2019 (Production(Production system: 1 dairy, 22 meat,meat, 33 wool;wool; ScimagoScimago classiﬁcation:classification: 11 Q1,Q1, 22 Q2,Q2, 33 Q3,Q3, 44 Q4,Q4, 55 notnot includedincluded inin thethe database).database).

Papers published in Journal of Dairy Science or Small Ruminant Ruminant Research received the most citations (p << 0.02 for comparisonscomparisons betweenbetween journals).journals). There There were were no correlations between the Scimago classificationsclassifications of the journals and the citationscitations received by paperspapers publishedpublished therein ((p >> 0.13 for all comparisons) (Table S10).S10). 3.5. Authors of Papers 3.5. Authors of Papers In total, there were 1559 individual authors of the papers; among these, 24 authors co-authored In total, there were 1559 individual authors of the papers; among these, 24 authors co-authored over 10 papers each (max: 73). Cumulatively, in the 580 papers, there were 2993 co-authors, i.e., on over 10 papers each (max: 73). Cumulatively, in the 580 papers, there were 2993 co-authors, i.e., on average 5.2 0.1 co-authors per paper (median: 5, min-max: 1–25). Furthermore, there were 640 average 5.2 ±± 0.1 co-authors per paper (median: 5, min-max: 1–25). Furthermore, there were 640 individual authors who were first or last authors in the papers; among these, 6 authors were first or individual authors who were first or last authors in the papers; among these, 6 authors were first or last in over 10 papers each (max: 62). On average, the 24 authors with >10 papers each were first or last in over 10 papers each (max: 62). On average, the 24 authors with >10 papers each were first or last authors in 53.0 6.1% of the papers in which they were co-authors (min: 0.0%, max: 90.9%). last authors in 53.0 ± 6.1% of the papers in which they were co-authors (min: 0.0%, max: 90.9%). The average number of co-authors per paper progressively increased from 2.1 0.4 and 2.6 0.2 The average number of co-authors per paper progressively increased from 2.1± ± 0.4 and 2.6 ±± 0.2 in the 1970s and 1980s, respectively, to 3.7 0.2 and 4.8 0.2 in the 1990s and 2000s, respectively, to in the 1970s and 1980s, respectively, to 3.7 ±± 0.2 and 4.8 ± 0.2 in the 1990s and 2000s, respectively, to 6.3 0.2 in the 2010s (p < 0.0001). In 30 papers (5.2%), there was only one author. 6.3 ± 0.2 in the 2010s (p < 0.0001). In 30 papers (5.2%), there was only one author. Moreover, there was a clear difference in the number of authors in the papers from the various Moreover, there was a clear difference in the number of authors in the papers from the various countries (p < 0.0001); Papers from France, Italy, and Greece had the highest number of co-authors countries (p < 0.0001); Papers from France, Italy, and Greece had the highest number of co-authors (7.9 0.9, 6.4 0.4, and 6.1 0.1, respectively), whilst papers from Australia and the United Kingdom (7.9 ± 0.9, 6.4 ±± 0.4, and 6.1 ±± 0.1, respectively), whilst papers from Australia and the United Kingdom had the lowest (3.3 0.3 and 3.2 0.3). All the above 24 authors with the most papers worked in one had the lowest (3.3 ±± 0.3 and 3.2 ± 0.3). All the above 24 authors with the most papers worked in one of the 11 countries (most in Greece, n = 10) and in one of the 35 establishments (most in University of of the 11 countries (most in Greece, n = 10) and in one of the 35 establishments (most in University of Thessaly, n = 8) from which most papers originated. Among them, 2 authors published papers with Thessaly, n = 8) from which most papers originated. Among them, 2 authors published papers with origins from two different countries and 4 authors published papers from two different establishments. origins from two different countries and 4 authors published papers from two different establishments.

Pathogens 2020, 9, 585 10 of 13 Pathogens 2020, 9, x FOR PEER REVIEW 10 of 13

Original papers had a significantly higher number of co-authors than did reviews: 5.3 0.1 versus Original papers had a significantly higher number of co-authors than did reviews: 5.3±0.1± versus 3.7 0.3 (p = 0.0008). In contrast, there was no difference in the number of co-authors among the 3.7±0.3± (p = 0.0008). In contrast, there was no difference in the number of co-authors among the papers papersaccording according to study to type study (p type= 0.15) (p or= 0.15)mastitis or mastitis aspect ( aspectp = 0.54) (p therein.= 0.54) therein. There was was a a clear clear correlation correlation between between number number of ofco-authors co-authors and and yearly yearly citations citations per perpaper paper (r = (r = 0.228, p < 0.0001) but not with total citations per paper (r = 0.025, p = 0.27). When only papers 0.228, p < 0.0001) but not with total citations per paper (r = −0.025,− p = 0.27). When only papers with > with5.2 authors> 5.2 authors (i.e., the (i.e., average the average co-authors co-authors per paper) per were paper) taken were into taken account, into account,correlation correlation was seen was for bothseen parameters for both parameters (r = 0.226, ( rp == 0.00030.226, pand= 0.0003r = 0.208, and p r= =0.0009,0.208, respectively)p = 0.0009, respectively)(Figure 8). Papers (Figure with8). onePapers of withthe above one of 24 the authors above 24as authorsfirst or aslast first author or last received author received significantly significantly more citations more citations than other than other papers: 20.6 2.0 total citations and 2.0 0.2 yearly citations versus 15.1 1.2 and 1.5 0.1, papers: 20.6 ± 2.0 total± citations and 2.0 ± 0.2 yearly± citations versus 15.1 ± 1.2 and 1.5± ± 0.1, respectively± (respectivelyp = 0.0098 and (p =0.0033,0.0098 respectively). and 0.0033, respectively).

Figure 8. AssociationAssociation between between number number of co-authors co-authors in pa paperspers with total citations and yearly citations received by these papers.

4. Discussion Discussion Ovine mastitis is an importantimportant diseasedisease ofof sheep.sheep. The The disease leads to significantlysignificantly reduced production in aaffectedffected animalsanimals andand adverselyadversely aaffectsffects theirtheir welfarewelfare standards.standards. Probably thethe earliestearliest reference reference to to ovine ovine mastitis mastitis was was in a Frenchin a French language language dictionary dictionary by the Frenchby the veterinarianFrench veterinarian Hurtrel Hurtrel d’ Arboval d’ Arboval in 1823 in [5 1823] in the [5] lemmain the lemma “araign “araignée”ée” (spider). (spider). He described He described an acute an acutefatal inflammation fatal inflammation of the mammaryof the mammary gland of gland ewes andof ewes mentioned and mentioned that the disease that the was disease erroneously was erroneouslybelieved to bebelieved due to to the be bite due of to an the “insect“. bite of an He “i statednsect“. that He stated the disease that the was disease more oftenwas more present often in presentunsheared in unsheared than in sheared than in ewes. sheared He mentioned ewes. He mentioned that the disease that the was disease called was among called French among shepherds French “malshepherds de pis” “mal (evil de of pis” the (evil udder) of the or “l’udder) araign oré “l’e” (thearaignée” spider). (the Nevertheless, spider). Nevertheless, the first known the first scientific known scientificapproach approach to investigating to investigating ovine mastitis ovine was mastitis undertaken was byundertaken the French by veterinarian the French and veterinarian microbiologist and microbiologistEdmund Nocard Edmund in 1886. Nocard Nocard in 1886. [5] quoted Nocard the [5] experience quoted the ofexperience a practicing of a veterinarianpracticing veterinarian about the aboutdisease, the described disease, indescribed detail the in sequence detail the of sequence signs of theof signs disease, of the and disease, performed and detailed performed pathological detailed pathologicaland microbiological and microbiological examinations. examinations. He isolated from He theisolated milk offrom an atheffected milk ewe of aan micrococcus affected ewe and a micrococcusdescribed its and morphological, described its cultural, morpho andlogical, biochemical cultural, characteristics. and biochemical He characteristics. described how He experimental described how experimental intramammary inoculation of secretion of diseased mammary gland or of a broth culture via the teat canal always successfully reproduced a condition very similar to the natural

Pathogens 2020, 9, 585 11 of 13 intramammary inoculation of secretion of diseased mammary gland or of a broth culture via the teat canal always successfully reproduced a condition very similar to the natural disease. Bridré [6] undertook a field investigation among hand-milked ewes and found that the incidence of the disease was approximately 5% and the case fatality 20%. He was the first who attempted to protect sheep against experimentally induced mastitis by means of immunization. Leyshon [7] was the first to investigate the disease in lactating ewes in England, in the region of East Anglia; he isolated from the mammary secretion of the diseased ewes a microorganism, whose morphological characteristics were different from the micrococcus isolated by Nocard (later identified as Mannheimia haemolytica). Minett [8] gave an account of the first reports on ovine mastitis and reviewed references on the occurrence and the aetiology of the disease. Later, Pegreffi [9] gave an account of the subsequent literature and Landau and Tamarin [10] reviewed the literature of the period 1963–1974. Despite its importance, mastitis in sheep has not been adequately studied, and there are facets of the disease that still have not been clarified. There are no scientometrics studies available in the international literature that would allow comparison of the mastitis literature to that about other diseases of sheep. An initial stage search revealed 2099 records (‘articles’, ‘reviews’, ‘proceedings papers’, and ‘data papers’) for [respiratory infection* OR pneumonia] and 1670 records for [abortion] during the same year range as in the current study (for comparison: 1326 records found in the present study at the respective stage of the search). Also, there are no relevant studies to compare the present findings to those for mastitis in other species. Again, an initial stage search as above revealed 11,196 records for [cattle OR cow* OR bovine] and 1504 documents for [goat* OR doe* OR caprine]. These findings lend weight to the idea that mastitis in ewes has not been studied extensively. Nevertheless, the progressive increase of published papers indicates that this lack of scientific work has been identified by researchers, and relevant studies have been designed and performed. Possibly, the recognition by the European Food Safety Authority [3] of the significance of mastitis as a very important disease of sheep might have contributed to the increase of relevant publications in recent years. Most published papers originated from countries in the para-Mediterranean area. This was expected given that a significant number of sheep exist in these countries (approximately 15% of the world’s sheep population) [11]. Sheep production systems in these countries are directed towards milk production, which also explains the increased number of papers referring to dairy production and indicates research oriented towards specific needs of agriculture in these countries. Consequently, most papers also originated from scientific establishments in these countries. As sheep production in the para-Mediterranean area is oriented towards dairy production, one may indicate that the adverse effects of mastitis (e.g., reduced milk production, changes in milk quality) [4] become evident immediately. In contrast, possible effects in flocks with meat production as the primary objective (e.g., reduced growth rate of lambs) [12] cannot be seen as easily, which undermines the recognition of the importance of the disease and hence the development of relevant studies. The findings indicate that some aspects of the disease have been covered extensively, e.g., aetiology, risk factors, and diagnosis. This is reflected in the presence in the international literature of papers that reviewed specific facets of the disease, e.g., [13–15], or methodologies employed for its study, e.g., [16]. However, other aspects of the disease, especially its control, have not been studied adequately, as evidenced by the small number of papers covering these topics. Hence, findings of scientometrics studies are useful and can be taken into account by governments, grant-giving bodies, and researchers to direct resources of research and perform relevant studies accordingly. Control studies into ovine mastitis can thus be a research priority for the near future. The findings also indicate the accumulation of research on ovine mastitis in some establishments. This is a trend that is observed with many research topics. Hosting institutions and researchers accumulate expertise on a topic and capitalize upon it to further expand their work. The presence of large groups also seems to favour scientific output; among the authors with many papers on the subject, most originated from the same establishment, which indicates not only large groups but also Pathogens 2020, 9, 585 12 of 13 long-established ones play a role in supporting the increased number of publications from those groups and of citations to those publications. The journals in which the papers were mostly published are ones with a specific approach to sheep (Small Ruminant Research) or a thematic approach to dairy studies (Journal of Dairy Research, Journal of Dairy Science) rather than journals of general interest. This indicates the increased importance of topical scientific journals and the preference of researchers to publish in them; in that way, researchers address themselves to a more specialised audience with greater interest in their work and the ability to understand it. In conclusion, this is the first ever scientometrics evaluation of mastitis in sheep. The findings indicate that the disease has not been studied as much as other diseases of sheep. Future studies should be directed to the control of the disease by using knowledge obtained from work into its other facets.

Supplementary Materials: The following are available online at http://www.mdpi.com/2076-0817/9/7/585/s1, Table S1: Details of 580 papers published from 1970 until 2019 on ovine mastitis and listed in the Web of Science platform; Table S2: Alphabetical list of countries from which papers on ovine mastitis have been published during the period –2019; Table S3: Alphabetical list of scientific establishments from which at least 5 papers on ovine mastitis have been published during the period 1970–2019; Table S4: Classification of papers on ovine mastitis published during the period 1970–2019, according to paper type, sheep production system, study type and mastitis aspect covered therein. Table S5: Number of papers on ovine mastitis published during the period 1970–2019 originating from the 11 countries with highest number of papers published, classified according to paper type, sheep production system, study type and mastitis aspect covered. Table S6: Associations between journals and countries of origin of papers on ovine mastitis published during the period 1970–2019; Table S7: Number of papers on ovine mastitis published during the period 1970–2019 in the 8 journals with the majority of relevant papers, classified according to paper type, sheep production system, study type and mastitis aspect covered. Table S8: Literature references in papers on ovine mastitis published during the period 1970–2019, according to year of publication, country of origin, establishment of origin, journal in which published and characteristic of paper. Table S9: Citations received by papers on ovine mastitis published during the period 1970–2019, and impact of the respective journal, according to year of publication, country of origin, establishment of origin, journal in which published and type of paper, production system, type of study and aspect of mastitis covered. Table S10: Citations received by papers on ovine mastitis published during the period 1970–2019, according to the journal in which they have been published. Author Contributions: Conceptualization, Methodology, Investigation, Data curation, Formal analysis, Validation, Visualization, Writing—original draft, Writing—review and editing, D.T.L. and G.C.F. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Leydesdorff, L.; Milojevic, S. Scientometrics. arXiv 2013, 1208, 4566. 2. Ioannidis, J.P.A.; Fanelli, D.; Dunne, D.D.; Goodman, S. Meta-research: Evaluation and improvement of research methods and practices. PLoS Biol. 2015, 3, e100226. [CrossRef][PubMed] 3. European Food Safety Authority. Scientific opinion on the welfare risks related to the farming of sheep for wool, meat and milk production. EFSA J. 2014, 12, 3933–4060. 4. Gelasakis, A.I.; Mavrogianni, V.S.; Petridis, I.G.; Vasileiou, N.G.C.; Fthenakis, G.C. Mastitis in sheep—The last 10 years and the future of research. Vet. Microbiol. 2015, 185, 136–146. [CrossRef][PubMed] 5. Nocard, M.E. A note on the gangrenous mastitis of dairy ewes. Annls Inst. Pasteur Paris 1887, 1, 417–428. 6. Bridré, J. Gangrenous mastitis of dairy ewes: Pathogenesis and vaccination. Bull. Soc. Cent. Med. Vet. 1907, 61, 500–506. 7. Leyshon, W.J. An examination of a number of cases of ovine mastitis. Vet. J. 1929, 85, 286–300, 331–344. [CrossRef] 8. Minett, F.C. Prevention of ovine mastitis by the use of staphylococcus toxoid. J. Comp. Path. Ther. 1939, 52, 167–182. [CrossRef] 9. Pegreffi, G. Mastitis in sheep and in goats. Bull. Off. Int. Epizoot. 1963, 60, 1009–1040. 10. Landau, M.; Tamarin, R. Ovine mastitis research, 1963–1973—A review. Refuah Vet. 1974, 31, 134–147. Pathogens 2020, 9, 585 13 of 13

11. ChartsBin. Sheep Livestock Data by Country. 2020. Available online: http://chartsbin.com/view/39461 (accessed on 3 May 2020). 12. Fthenakis, G.C.; Jones, J.E.T. The eﬀect of experimentally induced subclinical mastitis on milk yield of ewes and on the growth of lambs. Br. Vet. J. 1990, 146, 43–49. [CrossRef] 13. Leitner, G.; Rovai, M.; Merin, U. Clinical and subclinical intrammamay infection caused by coagulase negative staphylococci negatively aﬀect milk yield and its quality in dairy sheep. Small Rumin. Res. 2019, 180, 74–78. [CrossRef] 14. Vasileiou, N.G.C.; Mavrogianni, V.S.; Petinaki, E.; Fthenakis, G.C. Predisposing factors for bacterial mastitis in ewes. Reprod. Dom. Anim. 2019, 54, 1424–1431. [CrossRef][PubMed] 15. Fragkou, I.A.; Boscos, C.M.; Fthenakis, G.C. Diagnosis of clinical or subclinical mastitis in ewes. Small Rumin. Res. 2014, 118, 86–92. [CrossRef] 16. Katsafadou, A.I.; Vasileiou, N.G.C.; Fthenakis, G.C. Use of proteomics in the study of mastitis in ewes. Pathogens 2019, 8, 134. [CrossRef][PubMed]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).