First published in 2000: Robinson J. (2000) Editorial. Journal of Advanced Nursing 31(3), 497–499
On becoming an editor one is immediately aware of the continuous search for quality indicators by which a journal can be judged. As in many other areas of nursing practice, these can prove to be elusive. Perhaps the most widely used ‘hard’ indicator for nursing journals is the Social Science Citation Index (SSCI) Journal Reports that rank journals by their impact factor within the subject category. (SSCI ‘impact’ being derived from the number of papers from a specific journal cited in publications over the previous 2 years.) In 1998 the Journal of Advanced Nursing was ranked ninth amongst nursing journals in the world, first for British nursing journals, but superseded by eight journals published in the United States of America (USA).
There are, however, other ways of looking at journal quality. Anne-Marie Rafferty and Michael Traynor who together staff the Centre for Policy in Nursing Research at the London School of Hygiene and Tropical Medicine, together with Grant Lewison of the Wellcome Trust, have recently published Mapping Published Nursing Research: a third working paper (Rafferty et al. 2000). As part of their wider evaluation exercise to benchmark British nursing research, Rafferty et al.’s findings raise some fascinating issues that should provoke discussion in the United Kingdom (UK), and might also encourage nurses from other countries to replicate their work in their own geographical settings. Using the technique of bibliometric analysis with papers dating from 1988 to 1995 in the Wellcome Trust's Research Outputs Database (ROD), Rafferty et al. found that British nursing research features as one of six of the most rapidly expanding subfields within biomedicine and, within that, as one of the fastest growing areas for development. Starting from a low base of 055% in 1988, the proportion of papers grew to 1·29% of the ROD in 1995 — a rate of growth that outstrips even genetics research. Nursing publication performance, and its overall share of the international market, has also increased within the time period covered. The UK, compared with Australia and the USA, is the most rapidly expanding producer of nursing research publications. Rafferty et al. observe that part of this productivity must be attributed to the influence of the Research Assessment Exercise, now institutionalized as a periodic review of research quality across the UK university sector, also reflected in the Royal College of Nursing Research Society's burgeoning membership (11 000 in mid-1999). Rafferty et al. also found that members of two expert panels, one composed of nurse researchers, and one of National Health Service Trust nurses, agreed well on which journals were important. The Journal of Advanced Nursing, in which 45% (854) of the total output (1845) of cited nursing research papers between 1988 and 1995 appeared, was unanimously agreed to be in the top category. The International Journal of Nursing Studies, with 119 cited papers was also rated highly. Citation analysis was carried out on papers from these two journals that met specific criteria in relation to licensing from the Institute for Scientific Information, one or more funding acknowledgements, and two or more authors. Citation numbers were counted for the year of publication and for 4 subsequent years, except 1995.
It was found that the citation of nursing research papers differs considerably from other subfields in the ROD. To be in the top 10% (38) of the 381 nursing research papers from 1988 to 1994 in the two selected journals, it was necessary to have received seven citations during the 5-year period. By contrast, multiple sclerosis (the highest impact subfield in ROD) would need 45 citations to be in the top 10% of cited papers. Given UK nursing's apparent recent exponential growth in research publishing, it may be expected that this figure will be superseded in the future. Indeed, Rafferty et al. acknowledge that the original sources of their papers (Science Citation Index (SCI) and SSCI) underestimate activity between 1993 and 1995.
Another marked difference identified was that only 31% (581) of the 1845 nursing papers acknowledged any source of funding. Of these, the largest group acknowledged funding from the UK government. Rafferty et al. found that this was the most striking feature that differentiates nursing from biomedicine as a whole. Differences were also found between nursing and the rest of biomedicine on the indicator of the levels of coauthorship and collaboration. 18·5% of nursing papers were coauthored compared with 23% of biomedical outputs from the ROD as a whole. Rafferty et al. conclude that it is hard to draw any firm conclusion from this finding, especially as lower numbers of nursing authors appeared to be associated with publication in higher esteem journals.
In terms of content, Rafferty et al. identify a sharp distinction between what they term ‘endogenous’ and ‘exogenous’ research. The former characterized by nurse education, and the latter by mental health nursing, are concerned, respectively, with research to do with nursing as a profession, and with the nursing of patients. The two groups of papers were found to have quite different characteristics. Endogenous research has a rapid growth in output, fewer authors per paper, published in highly esteemed journals, but with a low chance of external funding. By contrast, exogenous research has a lower rate of growth in output; more authors per paper, published in lower rated journals but with a higher rate of being externally funded. Rafferty et al. suggest that endogenous papers concerned with problems and issues to do with nursing as a profession, differ markedly from the generality of biomedical papers for which more authors per paper and more external funding are also associated with publication in highly rated journals. Exogenous panels, on the other hand, seem to conform to the generality of papers in the ROD and seem fairly typical of biomedical research as a whole.
This brief review of Rafferty et al.’s meticulous and detailed study raises many issues for the future judgement of quality in nursing (and other biomedical) journals. They point out that, in the UK at least, nursing journals have proliferated since 1995, the last year considered in their analysis. A similar future study of papers published in the quinquennium 1996–2000 would undoubtedly serve to clarify some of the issues raised in this important report. One may predict that UK nursing research publications will occupy a growing proportion of the biomedical field in general. However, if this prediction is confirmed, will it also be accompanied by a demonstrated increase in acknowledged funding support? Will there also be an increase in the coauthorship of papers, reflecting the collaboration deemed to be essential if UK nurses wish to be successful contenders for British government funding sources?
One issue not covered in their study (which by its nature would have been impossible) is the contribution of the peer review process to journal quality. It is hoped that this issue will be addressed in a later editorial. In the meantime, Rafferty et al. have done a huge service to nursing and biomedical research by bringing these complex quality issues into the open. As with all good research they leave us even more curious for answers to the questions that they raise. Let us hope that others will take up the marker that they have put down in the search for excellence in journal quality.