Limits...
A Critical Look at Biomedical Journals' Policies on Animal Research by Use of a Novel Tool: The EXEMPLAR Scale.

Martins AR, Franco NH - Animals (Basel) (2015)

Bottom Line: For editorial policies to meaningfully impact attitudes and practice, they must not only be put into effect by editors and reviewers, but also be set to high standards.Results show a much greater focus of editorial policies on regulatory compliance than on other domains, suggesting a transfer of journals' responsibilities to scientists, institutions and regulators.Scores were not found to vary with journals' impact factor, country of origin or antiquity, but were, however, significantly higher for open access journals, which may be a result of their greater exposure and consequent higher public scrutiny.

View Article: PubMed Central - PubMed

Affiliation: Faculty of Sciences, University of Porto, Rua do Campo Alegre S/N, 4169-007 Porto, Portugal. up201101588@fc.up.pt.

ABSTRACT
Animal research is not only regulated by legislation but also by self-regulatory mechanisms within the scientific community, which include biomedical journals' policies on animal use. For editorial policies to meaningfully impact attitudes and practice, they must not only be put into effect by editors and reviewers, but also be set to high standards. We present a novel tool to classify journals' policies on animal use-the EXEMPLAR scale-as well as an analysis by this scale of 170 journals publishing studies on animal models of three human diseases: Amyotrophic Lateral Sclerosis, Type-1 Diabetes and Tuberculosis. Results show a much greater focus of editorial policies on regulatory compliance than on other domains, suggesting a transfer of journals' responsibilities to scientists, institutions and regulators. Scores were not found to vary with journals' impact factor, country of origin or antiquity, but were, however, significantly higher for open access journals, which may be a result of their greater exposure and consequent higher public scrutiny.

No MeSH data available.


Related in: MedlinePlus

The mean “Osborne score” for journals with the same EXEMPLAR classification, with an overlaid best-fit regression line (R2 = 0.39). Error bars represent ± 1 standard deviation of the mean for “Osborne score”, for a confidence interval of 95%. No journal was classified with an EXEMPLAR score of 17, 18 or 19 points. Correlation was stronger (Pearson’s r = 0.736), for lower scoring journals (under 10 points in the EXEMPLAR scale, corresponding to 91% of the sample).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4494415&req=5

animals-05-00315-f005: The mean “Osborne score” for journals with the same EXEMPLAR classification, with an overlaid best-fit regression line (R2 = 0.39). Error bars represent ± 1 standard deviation of the mean for “Osborne score”, for a confidence interval of 95%. No journal was classified with an EXEMPLAR score of 17, 18 or 19 points. Correlation was stronger (Pearson’s r = 0.736), for lower scoring journals (under 10 points in the EXEMPLAR scale, corresponding to 91% of the sample).

Mentions: EXEMPLAR score and scores by the Osborne et al. scale (Figure 5) showed a positive correlation (Pearson’s r = 0.623, p < 0.001). The sample showed a mean “Osborne score” of 2.98 points with a standard deviation of 1.28. Less than 25% of journals’ scored more than three points (out of 12), and 8% of journals had a nil score by the Osborne et al. scale. Impact factor, country of origin, scientific category and first year of edition of scientific journals had no effect on the overall EXEMPLAR score. The model of publication (subscription vs. open access journals) was however shown to influence the overall EXEMPLAR score, as significantly more open access journals scored eight points or above (χ2(1) = 17.87, p < 0.001) than subscription based journals (Figure 6). Also, scores varied significantly between publishers (χ2(53) = 96.968, p = 0.001), as well as between journals from the same publisher.


A Critical Look at Biomedical Journals' Policies on Animal Research by Use of a Novel Tool: The EXEMPLAR Scale.

Martins AR, Franco NH - Animals (Basel) (2015)

The mean “Osborne score” for journals with the same EXEMPLAR classification, with an overlaid best-fit regression line (R2 = 0.39). Error bars represent ± 1 standard deviation of the mean for “Osborne score”, for a confidence interval of 95%. No journal was classified with an EXEMPLAR score of 17, 18 or 19 points. Correlation was stronger (Pearson’s r = 0.736), for lower scoring journals (under 10 points in the EXEMPLAR scale, corresponding to 91% of the sample).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4494415&req=5

animals-05-00315-f005: The mean “Osborne score” for journals with the same EXEMPLAR classification, with an overlaid best-fit regression line (R2 = 0.39). Error bars represent ± 1 standard deviation of the mean for “Osborne score”, for a confidence interval of 95%. No journal was classified with an EXEMPLAR score of 17, 18 or 19 points. Correlation was stronger (Pearson’s r = 0.736), for lower scoring journals (under 10 points in the EXEMPLAR scale, corresponding to 91% of the sample).
Mentions: EXEMPLAR score and scores by the Osborne et al. scale (Figure 5) showed a positive correlation (Pearson’s r = 0.623, p < 0.001). The sample showed a mean “Osborne score” of 2.98 points with a standard deviation of 1.28. Less than 25% of journals’ scored more than three points (out of 12), and 8% of journals had a nil score by the Osborne et al. scale. Impact factor, country of origin, scientific category and first year of edition of scientific journals had no effect on the overall EXEMPLAR score. The model of publication (subscription vs. open access journals) was however shown to influence the overall EXEMPLAR score, as significantly more open access journals scored eight points or above (χ2(1) = 17.87, p < 0.001) than subscription based journals (Figure 6). Also, scores varied significantly between publishers (χ2(53) = 96.968, p = 0.001), as well as between journals from the same publisher.

Bottom Line: For editorial policies to meaningfully impact attitudes and practice, they must not only be put into effect by editors and reviewers, but also be set to high standards.Results show a much greater focus of editorial policies on regulatory compliance than on other domains, suggesting a transfer of journals' responsibilities to scientists, institutions and regulators.Scores were not found to vary with journals' impact factor, country of origin or antiquity, but were, however, significantly higher for open access journals, which may be a result of their greater exposure and consequent higher public scrutiny.

View Article: PubMed Central - PubMed

Affiliation: Faculty of Sciences, University of Porto, Rua do Campo Alegre S/N, 4169-007 Porto, Portugal. up201101588@fc.up.pt.

ABSTRACT
Animal research is not only regulated by legislation but also by self-regulatory mechanisms within the scientific community, which include biomedical journals' policies on animal use. For editorial policies to meaningfully impact attitudes and practice, they must not only be put into effect by editors and reviewers, but also be set to high standards. We present a novel tool to classify journals' policies on animal use-the EXEMPLAR scale-as well as an analysis by this scale of 170 journals publishing studies on animal models of three human diseases: Amyotrophic Lateral Sclerosis, Type-1 Diabetes and Tuberculosis. Results show a much greater focus of editorial policies on regulatory compliance than on other domains, suggesting a transfer of journals' responsibilities to scientists, institutions and regulators. Scores were not found to vary with journals' impact factor, country of origin or antiquity, but were, however, significantly higher for open access journals, which may be a result of their greater exposure and consequent higher public scrutiny.

No MeSH data available.


Related in: MedlinePlus