A H Henderson1, T Upile2, Y Pilavakis2, N N Patel2. 1. Department of ENT, The Great Western Hospital, Swindon, UK. Arthur.henderson@doctors.org.uk. 2. Department of ENT, University Hospitals Southampton NHS Trust, Southampton, UK.
Abstract
OBJECTIVES: Journals increasingly use reporting guidelines to standardise research papers, partly to improve quality. Although defining journal quality is difficult, various calculated metrics are used. This study investigates guideline adoption by otolaryngology journals and whether a relationship exists between this and journal quality. DESIGN, SETTING, PARTICIPANTS: Retrospective MEDLINE database review for English language, Index Medicus, journals of interest to otolaryngologists (October 2013). MAIN OUTCOME MEASURES: The resulting journals were examined for the number of guidelines endorsed and then tabulated against surrogate measures of journal quality (Impact factor, Eigenfactor, SCImago, Source-Normalised rank). The primary outcome measure was the number of recognised reporting guidelines endorsed per journal. This was then correlated against journal quality scores. For comparison, a further small sample correlation was performed with 6 randomly selected and 6 high-profile clinical non-otolaryngology journals. RESULTS: 37 otolaryngology journals were identified. Number of guidelines used and quality scores were not normally distributed. Mean guideline usage was 1.0 for otolaryngology journals, 1.5 for randomly selected, and 5.5 for the high-profile journals. Only 18/37 (49%) otolaryngology journals endorsed any guidelines, compared with 11/12 non-otolaryngology journals. Within otolaryngology, Eigenfactor positively correlated with guideline use (r = 0.4, n = 44, p < 0.01) otherwise no correlation was found between guideline endorsement and journal quality. CONCLUSIONS: Reporting guideline endorsement within otolaryngology journals is low. Although it might be expected that use of reporting guidelines improved quality, this is not reflected in the derived quality scores in otolaryngology. This may reflect low levels of use/enforcement, that quality indicators are inherently flawed, or that generalised guidelines are not always appropriate or valued by editors.
OBJECTIVES: Journals increasingly use reporting guidelines to standardise research papers, partly to improve quality. Although defining journal quality is difficult, various calculated metrics are used. This study investigates guideline adoption by otolaryngology journals and whether a relationship exists between this and journal quality. DESIGN, SETTING, PARTICIPANTS: Retrospective MEDLINE database review for English language, Index Medicus, journals of interest to otolaryngologists (October 2013). MAIN OUTCOME MEASURES: The resulting journals were examined for the number of guidelines endorsed and then tabulated against surrogate measures of journal quality (Impact factor, Eigenfactor, SCImago, Source-Normalised rank). The primary outcome measure was the number of recognised reporting guidelines endorsed per journal. This was then correlated against journal quality scores. For comparison, a further small sample correlation was performed with 6 randomly selected and 6 high-profile clinical non-otolaryngology journals. RESULTS: 37 otolaryngology journals were identified. Number of guidelines used and quality scores were not normally distributed. Mean guideline usage was 1.0 for otolaryngology journals, 1.5 for randomly selected, and 5.5 for the high-profile journals. Only 18/37 (49%) otolaryngology journals endorsed any guidelines, compared with 11/12 non-otolaryngology journals. Within otolaryngology, Eigenfactor positively correlated with guideline use (r = 0.4, n = 44, p < 0.01) otherwise no correlation was found between guideline endorsement and journal quality. CONCLUSIONS: Reporting guideline endorsement within otolaryngology journals is low. Although it might be expected that use of reporting guidelines improved quality, this is not reflected in the derived quality scores in otolaryngology. This may reflect low levels of use/enforcement, that quality indicators are inherently flawed, or that generalised guidelines are not always appropriate or valued by editors.