Our results suggest numerous points of weakness in the quality of abstracts from systematic reviews published in high impact factor oncology journals. Of the 17 items from the PRISMA extension for abstracts, no systematic review in our sample contained all abstract elements. On average, we found that SRs reported only 8 of the possible 17. Areas that were most lacking included languages included in the search process (12/182), registration number (0/182), funding source (25/182), and strengths and limitations (44/182).
In comparing our results with previous studies, it is evident that abstract reporting from systematic reviews in oncology could use improvement. For example, many abstracts disregarded a description of search details. For systematic reviews, this information represents the thoroughness of the search in locating all relevant evidence, and this information should be easily accessible in a systematic review abstract. Beller et al. found that 90% of abstracts included a search date and 60% included the databases that were searched. Those findings are higher than the 47% of abstracts reporting search date and the 49% listing databases found in oncology abstracts. Another important feature of systematic review methodology is the risk of bias assessment of the included primary studies. Only 19% of oncology review abstracts mentioned a study quality or risk of bias assessment, which was less than half of the systematic reviews that included an assessment in Beller et al.’s article \cite{23714302}. This figure is also much lower than the 43% found by Ochodo 2014 across other clinical areas \cite{Ochodo_2014}.
One way to increase the quality of abstract reporting is to have authors present the information in a structured form. Structured abstracts contain more information, are easier to read and search, and are welcomed by readers \cite{22453695}. Additionally, structured abstracts are favored by publishers as “over 50% of the papers with more detailed abstracts… [are] subsequently published” \cite{23942740}. We propose that oncology journals consider a structured abstract mandate for new submissions that adheres to the PRISMA extension for abstracts. We believe more thorough abstract reporting would occur should journal editors adopt such policies.
Future research should continue examining the quality of abstract reporting in other clinical areas. Aside from this study, only Kiriakou 2014 has examined the quality of abstract reporting among systematic reviews and meta-analyses; therefore, questions remain regarding abstract reporting practices across medical disciplines \cite{24581704}. Furthermore, our analysis indicated no detectable trend in abstract reporting after the PRISMA extension was published. It is possible that due to the fairly recent publication date of this work, it may be too early to note substantial differences over time. Future studies could determine whether more notable trends could be found.