A comparison of quality of abstracts of systematic reviews including meta-analysis of randomized controlled trials in high-impact general medicine journals before and after the publication of PRISMA extension for abstracts: A systematic review and meta-analysis

Background: Journal abstracts including those reporting systematic reviews (SR) should contain sufficiently clear and accurate information for adequate comprehension and interpretation. The aim was to compare the quality of reporting of abstracts of SRs including meta-analysis published in high-impact general medicine journals before and after publication of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) extension for abstracts (PRISMA-A) released in April 2013. Methods: SRs including meta-analysis of randomized controlled trials published in 2012, 2014, and 2015 in top-tier general medicine journals were searched in PubMed. Data was selected and extracted by two reviewers based on the PRISMA-A guidelines which recommend to include 12 items. The primary outcome was the adjusted mean number of items reported; the secondary outcome was the reporting of each item and factors associated with a better reporting. Adjustment was made for abstract word count and format, number of authors, PRISMA endorsement, and publication on behalf of a group. Results: We included 84 abstracts from 2012, 59 from 2014, and 61 from 2015. The mean number of items reported in 2015 (7.5; standard deviation [SD] 1.6) and in 2014 (6.8; SD 1.6) differed and did not differ from that reported in 2012 (7.2; SD 1.7), respectively; adjusted mean difference: 0.9 (95 % CI 0.4; 1.3) and -0.1 (95 % CI -0.6; 0.4). From 2012 to 2014, the quality of reporting was in regression for "strengths and limitations of evidence" and "funding"; contrariwise, it remained unchanged for the others items. Between 2012 and 2015, the quality of reporting rose up for "description of the effect", "synthesis of results", "interpretation", and "registration"; but decreased for "strengths and limitations of evidence"; it remained unchanged for the other items. The overall better reporting was associated with abstracts structured in the 8-headings format in 2014 and abstracts with a word count <300 in 2014 and 2015. Conclusions: Not surprisingly, the quality of reporting did not improve in 2014 and suboptimally improved in 2015. There is still room for improvement to meet the standards of PRISMA-A guidelines. Stricter adherence to these guidelines by authors, reviewers, and journal editors is highly warranted and will surely contribute to a better reporting.