Publication redundancy: a case of déjà vu or where have I read this before?
Article first published online: 10 FEB 2009
© 2009 The Authors. Journal Compilation © 2009 Public Health Association of Australia
Australian and New Zealand Journal of Public Health
Volume 33, Issue 1, pages 3–4, February 2009
How to Cite
Lowe, J. (2009), Publication redundancy: a case of déjà vu or where have I read this before?. Australian and New Zealand Journal of Public Health, 33: 3–4. doi: 10.1111/j.1753-6405.2009.00329.x
- Issue published online: 10 FEB 2009
- Article first published online: 10 FEB 2009
Sometimes a journal is unable to publish a good piece of work because it has appeared in print elsewhere. This is painful for the authors and the Editors, and a problem for the journal. Usually we reject papers because reviewers’ assessments are that a paper does not contribute to our knowledge and understanding of an area but, increasingly, we are rejecting academically sound papers because of ‘previous publication’. Sometimes this happens because authors are unaware of the rules governing redundant publication, but ever-changing publication formats certainly add to the problem.
A publication is a piece of work that appears in electronic or paper version and is available to the public. If it can be cited, then it has been published. In common with most top-rate journals, we do not publish material that has previously been published either through peer-review journals or in the popular press. However, we are increasingly aware of the difficulties of authors who are funded from government and non-government organisations that wish to disseminate interesting new findings. It may be through the organisation's website, by direct mail or in a printed report to constituents. In fact, the more successful investigators are with their work, the more difficult it is for authors to keep an embargo on the information prior to the peer-reviewed publication. In previous issues of this journal, Jeanne Daly and Judith Lumley1 as well as Konrad Jamrozik2 have in 2004 and 2005 begun to address this issue of redundancy. As they noted, there are primarily two reasons why redundancy is an important issue.
One of the substantial reasons why redundancy is an issue for all journals is that it can inflate the findings of one study. As the sophistication and ease of electronic searches for the conducting of meta analysis has improved, the same study under similar titles can and will bias the results of the meta analysis. Secondly, when manuscripts and data are published in multiple sources, the ability of that information to be credited to a proper peer-review article is diluted. Citations are then distributed across multiple publications and dilute the impact for each journal. As the index factor is influenced by the number of citations of an article in a journal, the dilution of citations among different journals or sources reduce both the impact of the manuscript and the journal it was published in.
We are often asked: “What is the definition of ‘being published’?” The clearest rule of thumb is that if the information is readily available either through print media, electronic or even through media releases, it is for all practical purposes, published. The dilemma for authors is how do you fulfil your contractual agreement with the agency that has funded you, yet maintain the unpublished status of that information? There are no hard and fast rules, however, there are a number of guidelines we can use to begin to negotiate with agencies.
Most agencies do not require the investigators to report findings that investigate predictors or determinants or conduct a sophisticated analysis. Most reports only require information that focuses on frequencies and the implications of those frequencies. This type of valid analyis can be provided to funding agencies without completely putting the information at risk. Most of the audiences for the reports are used to working with this type of the presentation of data. It is then the authors’ challenge to develop from the report a publication manuscript that moves beyond the frequencies. The new ‘information’ manuscript needs not just to report the frequencies or even use that as the bases for the manuscript, but also to provide a complete re-analysis of the same data. The manuscript may be more complex, more intricate, or it may use different variables to look at predictors or determinants. What it must do is present the data from a different angle. Referring to the earlier publication of demographics or frequencies is not inappropriate (as long as you cite it). If the new manuscript focuses on and provides new knowledge we are interested in it. Our recommendation is to save the determinant/predictor paper for a journal article and provide the agency with survey frequency results.
Qualitative research methods do not have the capacity to calculate frequencies but an analogous approach is to give a descriptive overview of the data, perhaps with a few illustrative quotations. It is then necessary to add a statement to the effect that the quotations are not necessarily representative of the full data set which requires further analysis to establish whether there are differences among study participants and to develop an explanation of why this is so.
If a study produces startling new findings, an agency may be sorely tempted to report these findings to the media. Once this is done, peer review is compromised. In negotiating contracts with your funding agency it is therefore reasonable to point out to the agency that a peer-reviewed publication gives not only credibility, but also status to the information. The knowledge that was discovered due to their funding is more credible when it is published in a top-tier journal. We find that the Australian Government and other major funding organisations typically do not have a concern about holding off the use of information until the publication has come out in print in a peer-reviewed journal. The delay may frustrate but they have the assurance that the findings have withstood critical assessments by informed colleagues.
One of the areas where we frequently encounter problems with earlier publication is in Indigenous health. It could be argued that research which can make a difference to public health practice should be published as soon as possible and peer-review processes are slow in comparison with publication in a newsletter. It is true that peer-review processes are particularly slow with papers addressing Indigenous health because of the difficulty in finding suitable reviewers. On the other hand, we have had papers where reviewers have recommended a different form of analysis which has significantly altered the findings of a study. What happens then if the earlier results have already been implemented? A compromise is for a limited circulation of results to participants and concerned organisations in a manuscript clearly marked ‘Draft for comment. Do not cite’. In addition a reasoned appeal to journals for a quick review process might help to speed to review process.
It saddens us as Editors to have to reject a good piece of work because it has been published elsewhere. The end result is a lose/lose for all of us. Our hope is that we can work with authors during the negotiations. If they have questions please do not hesitate to contact us. As writers ourselves we see it as our individual and collective responsibility to make sure that redundancy of publications does not occur.
In this issue
This month we have some ‘wicked problems’ for you to ponder. Such ‘wicked problems’ are often bound by complexity, and complexity is the subject of the first set of papers. There are a number of problems highlighted here, the overall message being that the data you have might be a long way from perfect, but might be the only data you have and need to be loved (respectfully treated) for what they are, warts and all.
Anna Ziersch and colleagues used a South Australian telephone survey to investigate social capital as a health benefit and show that policy should be responsive to the different strengths and weaknesses found in urban and rural settings. Peter Feldman, Deb Warr and colleagues sought to make sense of data from the Neighbourhood Renewal projects in two papers. In the first, they have investigated the somewhat depressing if unexpected relationship between our neighbourhoods and self-reported health status. They show that there seems to be a link but they highlight something of a chicken-and-egg situation – does inadequate housing cause ill health or do we have ill health and then end up in poor housing – and does all this stem from upstream problems such as lack of a job, which originated in a lack of education? In the second paper they analyse responses to open-ended questions in the survey and identify the way in which ‘social incivilities’ result in feelings of insecurity and anxiety. Another set of complex problems is addressed by Gabrielle Bammer and colleagues who, 10 years ago and way ahead of their time, began thinking about how randomised controlled trials, in this case of a buprenorphine drug substation trial, are translated into policy and practice. They identify seven key clinical and policy implementation criteria which most affect program uptake.
We have two papers addressing asylum seekers in Australia, a ‘wicked problem’ never far from the front pages of our newspapers. Vanessa Johnston has conducted an insightful analysis of Australia's asylum policies in the light of the Prime Minister Kevin Rudd's Labor government's reforms and summarising what is known about the health of refugees seeking shelter and protection in our wide brown and sparsely populated land. In her conclusions she reminds us about the ethical and legal dangers of applying legal sanctions and thereby sacrificing people subsequently shown to be bona fide exiles, simply because a few people might possibly be a problem. Then, Glenda Koutroulis shows how traditional public health interventions, especially quarantine and surveillance, have been used as a way of controlling the influx of political refugees. Her extraordinary analysis shows how the use of language and gradual dehumanising slides into ‘othering’ refugees, making it possible for them to be treated in dehumanised ways — an important lesson about a very wicked problem.
The next set of wicked problems is about Indigenous health. Priscilla Pyett and colleagues make some practical suggestions for making and maintaining working relationships with partners in Indigenous health. Naomi Priest and colleagues have conducted a review of studies about Australian Indigenous child health in selected databases and websites. They show that studies focus on physical rather than other kinds of health. David Thomas and colleagues have considered two methods of measuring tobacco consumption in remote communities, an important problem when considering cessation and reduction programs.
The next wicked problem we encounter is about measuring health trends. Amr Abou Alnour and James Harrison have conducted a record linkage study in which they consider the curious case of declining suicide rates. They show that incomplete routine data result in underestimation of suicide deaths. The issue considered by Martin Tobias and Li-Chia Yeh is about the mortality trends in an analysis of linked datasets through the New Zealand Census-Mortality Study. Better access to quality health care has the potential to reduce health inequalities.
How best to measure risk is the last of our wicked problems. Justin Newton Scanlan and Anita Bundy have compared the health of young Australians without jobs in cross-sectional studies during times of low and high unemployment, showing that not having a job is bad for your social and psychological health. Lin Mao and colleagues show that psycho-social ill health is experienced by a group of 509 gay men attending high HIV caseload general practices in Sydney. The authors call for structured community health promotion activities. Knowledge and risk of papillomavirus is the topic of Ruth McNair and colleagues community-based self-selected and self-completed questionnaire study of 349 Australian lesbian and bisexual women.
The Letters section has more wicked problems including smoking bans in pubs, sexually transmitted infections, and problems with interpreting cancer registries.