• Open Access

Quality Matters: Publishing in the Era of CONSORT, REFLECT, and EBM

Authors

  • K.W. Hinchcliff,

  • S.P. DiBartola


Dear Editor

This issue of the journal introduces a number of initiatives intended to improve the quality of both the science (ie, evidence) and reporting of studies described in articles published in the journal. These initiatives, and the underlying rationale, are highlighted by our publication of 3 important articles: the REFLECT statement describing an extension to the CONSORT statement for reporting of clinical trials involving livestock1; the article “Quality of reporting of clinical trials involving dogs and cats and associations with treatment effects” describing the quality of reporting of clinical studies involving dogs and cats2; and “A Survey of Evidence in the Journal of Veterinary Internal Medicine Oncology Manuscripts from 1999 to 2007”3 describing characteristics of highly cited articles in veterinary oncology. Together, these articles highlight both the reasons why and the mechanisms by which the journal intends to increase the quality of the material it publishes.

There is an increasing expectation by veterinary scientists that articles in the veterinary literature support evidence-based medicine. Articles that support evidence-based medicine require not only high quality science but also comprehensive reporting of all relevant aspects of the study. The CONSORT guidelines were developed almost a decade ago to improve the quality of reporting of clinical trials conducted in humans. Adoption of these guidelines by veterinary scientists and clinicians has improved the quality of some articles in veterinary medicine. Despite this, there is still considerable ignorance of the CONSORT guidelines and only sporadic use of these guidelines by authors, reviewers, and editors. Evidence to support the imperative to improve the quality of evidence in articles published in the journal comes from the article by Sahora and Khanna, which documents little change in the quality of clinical trials in veterinary oncology reported in the journal between 1999 and 2007. Additionally, the article by Sargeant et al describes a pattern of less than optimal reporting of methodology and results in clinical trials of dogs and cats, and links poor reporting with positive trial results (ie, manuscripts that fail to report key features of study design are more likely to report positive results of the intervention). These studies, along with others in the human medical and veterinary literature, clearly indicate the need for raising the bar on the quality of clinical trials, and for increasing the completeness of reporting of both methodology and results in these trials.

The CONSORT statement, and the derivative and supplementary REFLECT guidelines1 provide veterinarians with useful information on the design and reporting of clinical trials. Hopefully adherence to these guidelines in the reporting of clinical trials in veterinary medicine will improve the quality of articles published in the journal and enhance the capability of readers to use evidence-based principles in their practices.

We recognize that it is not feasible at this time to introduce rigorous and strict adherence to the CONSORT and REFLECT guidelines for all manuscripts accepted for publication in the journal. To do so currently would result in very slender issues of the journal. What we can do is require that authors report on clinical trials using the CONSORT and REFLECT guidelines, even if doing so results in omissions and errors in study design being much more obvious to readers. We believe accurate reporting is as important as sound study design, and as the veterinary scientific community works toward improving the quality of clinical trial design, the quality of reporting also will be improved.

The move to publishing with Wiley-Blackwell also has allowed us to track the articles that are most widely read, as reflected by hits on the website and downloads, in addition to tracking citation rates of articles. Data about readership obtained from website hits and downloads is more immediate than is information about citation rates, and provides information on readership of an article in the weeks to months after it is published. Importantly, this metric of readership is more important than the citation rate because it reflects what readers of the journal are using, not what scientific authors are citing. The revealing statistics are that consensus statements, reviews, high quality randomized-clinical trials (such as the QUEST study4), and large case series are frequently read, whereas case reports are rarely read. We used this information as we developed the editorial policy for the journal.

Beginning with the first issue of 2010, the journal has introduced a new set of editorial policies that are set forth in our revised guidelines for authors and for reviewers. These guidelines, which are available at http://mc.manuscriptcentral.com/jvim, provide the operational details of our new policy. Notably, there will be a higher bar for quality of studies acceptable for publication in the journal based on an assessment of their utility in forming evidence-based medicine decisions, requirement for more complete reporting of methods, a clearer definition of the types of manuscripts considered for publication, and continued scrutiny of case reports for unique value.

Ancillary