To investigate the current prevalence of open access (OA) in the field of dentistry, the means used to provide OA, as well as the association between OA and citation counts.
PubMed was searched for dental articles published in 2013. The OA status of each article was determined by manually checking Google, Google Scholar, PubMed and ResearchGate. Citation data were extracted from Google Scholar, Scopus and Web of Science. Chi-square tests were used to compare the OA prevalence by different subjects, study types, and continents of origin. The association between OA and citation count was studied with multivariable logistic regression analyses.
A random sample of 908 articles was deemed eligible and therefore included. Among these, 416 were found freely available online, indicating an overall OA rate of 45.8%. Significant difference in OA rate was detected among articles in different subjects ( P < 0.001) and among those from different continents ( P < 0.001). Of articles that were OA, 74.2% were available via self-archiving (‘Green road’ OA), 53.3% were available from publishers (‘Gold road’ OA). According to multivariable logistic regression analyses, OA status was not significantly associated with either the existence of citation ( P = 0.37) or the level of citation ( P = 0.52).
In the field of dentistry, 54% of recent journal articles are behind the paywall (non-OA) one year after their publication dates. The ‘Green road’ of providing OA was more common than the ‘Gold road’. No evidence suggested that OA articles received significantly more citations than non-OA articles.
Open Access (OA), free online access to journal articles without any restrictions posed by subscriptions, is a recent revolution in scientific publishing made possible by the Internet . By disseminating research findings to the largest possible readership, OA is believed to accelerate research, enrich education, and benefit all stakeholders including researchers and practitioners, especially those in low-income countries and resource-poor institutions . Also, it is a recommended measure to reduce avoidable waste in biomedical research .
Since the release of the Budapest Open Access Initiative (BOAI) in 2002, numerous efforts have been made by governments, funders, institutions and publishers to promote OA . In April 2008, the US National Institutes of Health (NIH) began to require that all grantees submit to PubMed Central (PMC) their final manuscripts upon acceptance for publication, to be made OA no later than 12 months after the publication date . The same year also saw the establishment of the Harvard OA mandate . Thereafter, many funding/government agencies and institutions across the world followed suit . As of December 2015, a total of 750 OA mandates have been registered on the ROARMAP (The Registry of Open Access Repository Mandates and Policies) website (roarmap.eprints.org), representing a five-fold increase since January 2005.
Due to the abovementioned importance of OA and efforts/resources devoted to its development, it is crucial to assess periodically the prevalence of OA and relevant characteristics using a precise and comprehensive approach . Although the OA status of biomedical articles has been investigated in several previous studies , to our knowledge, there has been no such study in the field of dentistry. In addition, the ‘citation advantage’ of OA has been under debate . Whether OA articles in dentistry receive more citations than non-OA articles still remains unstudied.
Therefore, the objectives of this study were:
To determine the prevalence of OA among journal articles in the field of dentistry;
To analyse the current means used to make these articles OA; and
To investigate the association between OA and citation count.
Materials and methods
As in previous similar research , the sample was selected from the PubMed database due to its popularity in the biomedical field. The electronic search was conducted on 13th Feb, 2015 to identify dental journal articles published during the year of 2013 (search strategy see Table 1 ). A follow-up of slightly more than one year was deemed appropriate because 12 months is the maximum embargo that NIH permits ; most publishers allow authors to self-archive their articles after a 12-month delay; and that many journals make their contents OA after an embargo period (‘delayed OA’) of one year .