David Wright, Professor John Powell
The need for full publication of funded research
Full publication of research findings is an important factor in maximising the value of funded research. Failing to publish research findings can lead to clinical or public health practice not being informed by current evidence; a failure to invest in new interventions proven to be effective; and a failure to disinvest in less effective or harmful treatments. Failure to publish research can also lead to the unnecessary duplication of studies, perhaps causing patients to be recruited to research where the results should already be known. Publishing research findings is therefore fundamental to securing returns on research investment, informing and improving clinical practice and protecting patients from unnecessary risks.
The health research publication industry has developed significantly in recent years, particularly with the growth in freely accessible online journals. It is surprising therefore that the publication rate for health research has remained at about 50%.
One study reported that only half of European Union funded health research funded in the period 1998-2006 resulted in an identifiable report. In addition, research studies with positive or significant results have been shown to be more likely to have their findings published than those with negative or nonsignificant results, despite the latter findings often being just as important if not more so. Clinical trials with positive results are published in journals about one year earlier than trials with non-positive results.
The NIHR is unusual in being a health research funder that publishes transparent, full accounts of every piece of commissioned research in its own free, open-access, permanently archived peer-reviewed journal. The first journal in the NIHR Journals Library was Health Technology Assessment, established in 1997, with an Impact Factor of 5.027 in March 2016 and rated first out of 88 in the Health Care Sciences and Services category of Thomas Reuter’s 2014 Journal Citation Reports.
The journal was used as a model for other journals aligned with NIHR programmes: Efficacy and Mechanism Evaluation, Health Services and Delivery Research, Programme Grants for Applied Research, and Public Health Research. All researchers funded through these programmes publish full accounts of their research, regardless of whether findings are positive, neutral or negative. The NIHR Journals Library also requests that authors provide a detailed account, including challenges and lessons learned, to inform other researchers, capturing this knowledge and know-how to help others avoid pitfalls. In line with other leading journals, the NIHR Journals Library asks researchers to provide a statement on how to access their research data. To date, the NIHR Journals Library has published over 1000 reports.
NIHR Journals Library achievements in publishing research
A series of studies conducted by the NIHR Research on Research programme has assessed the extent to which the Health Technology Assessment Journal has responded to the principles of the Adding Value in Research. These have shown that the journal has been successful in achieving a near full publication rate, with 98% of studies funded by the NIHR HTA Programme commissioned in the period 2002–2011 publishing. Trials published in Health Technology Assessment were shown to publish faster on average than those funded by the US National Institutes of Health and evidence syntheses published on average faster than Cochrane reviews. A further study reported that studies publishing in the Health Technology Assessment journal in the period 2007–2012 were rated as being of statistically higher clinical relevance than other top rated papers in leading international journals. Assessments of relevance were independently rated by physicians, located around the world.
The NIHR Journals Library has also demonstrated its success in addressing publication bias, publishing studies regardless of study outcome. In the period 1997–2015, 42% of studies published in the five NIHR Journals have published nowhere else. A study recently reported that of 131 NIHR HTA funded trials publishing in the Health Technology Assessment journal, 66 showed no difference in cost or effectiveness and four showed an unambiguous preference for either cost or effectiveness.7 This finding is to be expected given the unbiased commissioning of research and illustrates that the journal publishes all results, whether positive, negative or neutral.
Whilst the NIHR Journals serve as a full repository of completed studies, the NIHR recognises the value of supporting other publications that focus on study outcomes and are targeted to specific evidence user audiences. NIHR-funded researchers published over 1000 non-NIHR Journals Library publications in the period April 2012–March 2015. The NIHR Journals Library remain widely accessed, however, receiving over 200,000 visits to its website in April 2015–March 2016 alone.
How did the NIHR Journals Library meet the requirements of adding value?
The NIHR Journals Library promotes excellence and innovation in editorial practice, ensuring that full accounts of research are published and study findings are disseminated in a timely and accessible manner, while retaining a rigorous approach underpinned by peer-review by at least four experts on each report. The NIHR Journals Library continually adapts its editorial activities, responding to the needs of evidence users and the researcher community. For example, the NIHR Journals Library now requests that all authors include Patient and Public Involvement statements in their reports.
The NIHR Journals Library supports NIHR’s policy on open access and its commitment to transparency through the publication of freely downloadable ‘gold’ open access publications on the Journals Library website. The NIHR Journals Library continues to seek to innovate editorial practice and is currently developing the NIHR Journals Library website to publish research findings within the context of full study documentation, including an open access protocol and links to post-reporting dissemination (such as post study evidence syntheses).