Main content area

Assessment of systematic reviews and meta-analyses available for bovine and equine veterinarians and quality of abstract reporting: A scoping review

Buczinski, S., Ferraro, S., Vandeweerd, J.M.
Preventive veterinary medicine 2018 v.161 pp. 50-59
cattle, confidence interval, horses, medicine, meta-analysis, metabolism, nutrition, observational studies, systematic review, veterinarians, veterinary medicine
Evidence-based medicine (EBM) is used in various areas including veterinary medicine. The assessment of the quality of systematic review and meta-analyses (SR-MA) despite their importance in the EBM process is uncommonly performed in veterinary medicine due to the absence of specific dedicated tools. The main objective of this observational study was to examine the extent and nature of SR-MA that can be available online to an equine or bovine veterinarian. Secondary objectives included: (1) to determine if A MeaSurement Tool to Assess systematic Reviews (AMSTAR) tool can be used for equine and bovine SR-MA methodological assessment and if it has a satisfactory interrater reliability in a subsample of these SR-MA and (2) to appraise the completeness of abstract reporting of this sample.A scoping review using equine and bovine medical science SR-MA retrieved from PubMed was performed. A sub-sample of these reviews (n = 30) were independently assessed by 3 different raters using the AMSTAR tool validated for medical reviews. The completeness of abstract reporting was performed using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist.Ten and 168 SR-MA were retrieved for equine and bovine species respectively. For bovine SR-MA, 93 reviews were about nutrition and metabolism topic. On the 30 SR-MA subsamples, with 10 equine and 20 bovine SR-MA randomly chosen for AMSTAR assessment, the median interrater agreement (Kappa) was 0.60 (interquartile range: 0.36–0.71) depending on AMSTAR item and pairs of raters. When focusing on the total score of AMSTAR, the inter-observer intra-class correlation coefficient was very good (0.85, 95% confidence interval (CI): 0.74–0.92). The AMSTAR items that were unfrequently reported (33% or less of reviews) were “a priori” protocol of SR-MA specification, complete report of the list of studies (included and excluded studies), quality assessment of the included studies, publication bias assessment and conflict of interest (reported either for included studies and for SR-MA authors). Abstracts reporting quality was low with a median percentage of complete reported items of 33% (range: 8–58%)In large animal veterinary medicine, SR-MA are uncommonly performed in equine and bovine medicine. The SR-MA can be assessed using AMSTAR with acceptable inter-rater reliability, which is helpful to assess SR-MA methodological quality.