Discussion

The current global market of scientific publication is a complex one that has evolved over hundreds of years. Searches of UlrichsWeb in 2008 show a total of 219 774 active journals, of which 64 620 are active and academic/scholarly (Boolean search) and 24 059 active and reference journals.4 Working from Ulrich's and the Thompson (SIS) citation index, it is estimated that 23 750 journals published in 2006 yielded a total of 1 350 000 peer-reviewed journals.13 Once published, these journals are searched, read and referenced by readers and researchers worldwide. These scientific journals for analytical purposes can be divided into two broad categories commercial publications and non-commercial publications. The commercial publishers seek to maximise profit through publishing, while non-commercial/academic publishers (universities and academic societies) as in the Sri Lankan context are mainly non-profit institutions whose objectives are usually linked to furthering the interests of their research or professional development. For Sri Lankan journals, the main sources of funding are advertisers, donors and societies/institutions. Their involvement can be susceptible to influence by external factors such as the state of the local or global economy. This therefore could place limits on the ability of Sri Lankan medical journals to maintain their operations.

There were 28 medical journals in Sri Lanka which published issues in 2008/2009. In 2005, the whole of Africa had only 59 medical journals published in 33 countries.10 The average cost of publishing one article in Sri Lanka was LKR 6646 (US$60), while the cost of printing one page was LKR 1848 (US$17). In contrast the estimated cost of scientific publications in the UK2 and USA14 is US$2000 (LKR 220 000) and US$3000 (LKR 330 000) respectively per article. The comparatively low cost for Sri Lankan journals can be attributed to the lack of expenses related to copy-editing, peer-review and dedicated staff, and office maintenance. However, the savings from these may compromise the quality of peer review, copy-editing and punctuality of publications. Though these are relatively lower than the figures from the developed world, publishing an article in a Sri Lankan medical journal incurs a significant cost to the publisher. In addition, in a developing country such as Sri Lanka, where the per capita monthly income is LKR 9100 (US$83)15 and where a majority of the research work is self-funded, the cost of publication per article of LKR 6646 (US$60) is substantially high. Authors, researchers, academic and funding organisations, and the government of Sri Lanka should be aware of this. Despite the availability of international journals, local journals still have a vital role to play in disseminating local knowledge, translating it into policy and practice, and contributing to national development.16 Therefore, all stake holders, especially the government and funding organisations, should recognise the importance of medical journals' potential as a tool for reduction in health inequity and poverty. Publication of medical journals should be supported financially by the state institutes, too, to improve their quality and sustainability. This would help the journals to be published more frequently and punctually, which in turn would foster local research and increase the publishing capacity of Sri Lanka.

In the Sri Lankan model of journal publication, the total fixed costs of publication are minimal, as editorial services, peer review and proofing services are volunteered. Hence, the costs of publishing are mainly dependent on variable costs. This cost distinction is important in comparing publishing practices. Fixed costs of both print and electronic journals are recovered in three main ways: through subscriptions, author-side payments or contributions of time by editors and others, as in the Sri Lankan setting. Journal printing costs elsewhere tend to be in the $30–50 per subscription range, and distribution costs in the $10–20 range. The variable cost per article in a subscription is about $0.40, and the per-article page in a subscription is under $0.04. The variable costs of print journals are recovered as part of subscription prices, or, for society journals, personal subscriptions are recovered through member fees. The high price of some journals is due to a very large fixed cost (when there are many articles) that is recovered by a small number of subscribers.

The Sri Lankan medical journals averaged 16 articles (range 5–55) and 61 pages (range 26–238) per journal or about 3.8 pages per article, which is relatively small in comparison with international life science journals including medicine. In addition, only 5.0% of the contributions were from foreign countries, of which 39% were from neighbouring India. Out of the 28 journals studied in this analysis, only one of the journals (Ceylon Medical Journal) was indexed in PubMed and carried an impact factor. Thus, although the Sri Lankan model may foster research publications and local dissemination of knowledge, international contributions and accessibility to the international audiences remain low. Making journals accessible online is one probable solution this issue. Online publication would also help to reduce the printing and distribution costs.12 In addition, the availability of appropriate software makes it possible to make the submission of articles, peer review process and copy-editing to be done online conveniently for both authors and reviewers. Online publication on a live website would increase the visibility of the journal and its contents. This may help to alleviate the disadvantages of not being included in the indexing systems run by the developed world. The International Network for the Availability of Scientific Publications, a non-profit-making organisation based in the UK, maintains a website for Sri Lankan scientific journals (Sri Lanka Journals Online). At present, there is no extra cost for the journals to be included in this website. Eleven of the Sri Lankan medical journals are available online through this website. The journal website has gained increasing popularity at both national and international levels. Google Analytics shows that from 23 September to 31 December 2008, there were 1364 visits from 82 different countries. In contrast, from 1 January to 30 June 2010, there were 55 756 visits from 186 different countries: the top four sources for visitors were: Sri Lanka (13 843 from seven cities), India (7740 from 130 cities), USA (6424 from 2264 cities) and UK (3796 from 368 cities). The United Nations' repeated calls to provide universal access to the scientific and medical literature have not been heeded by most journals in the world.16 However, all Sri Lankan medical journals which are available online allowed free access to all their articles. Although this may not be a genuine form of open-access publication, this is a healthy trend which should be continued and fostered for wider dissemination of knowledge. The editors and publishers of other medical journals in Sri Lanka, too, should work towards making their journals available online using the Sri Lanka Journals Online website. With online publication, it may be possible to reduce the number of copies printed and thereby to reduce the escalating printing expenses. One drawback of the online publishing system would be the reluctance of the traditional researchers and reviewers to accept the computer-based system.

Our study has several limitations. The lack of a comprehensive list of medical journals in the country resulted in authors having to limit their journal search to the aforementioned sources. Thus, there is a possibility of inadvertently non-identifying certain medical journals in print and previously printed. In addition, the limited number of publications from the fields of allied health science (nursing, physiotherapy and pharmacy) were not included in the study.