What useful information does a search summary table provide: a case study from a mixed methods systematic review

Session: 

Oral session: Searching and information retrieval (3)

Date: 

Thursday 24 October 2019 - 11:00 to 12:30

Location: 

All authors in correct order:

Bethel A1, Rogers M1, Abbott R1, Hunt H1, Boddy K1, Whear R1, Thompson-Coon J1
1 University of Exeter, UK
Presenting author and contact person

Presenting author:

Jo Thompson-Coon

Contact person:

Abstract text
Background: search summary tables (SST) provide the results of the search methods at the end of a systematic review. Information collected in the SST includes the databases searched, references included in the review, databases in which they were found, which supplementary search methods were used,whether the researchers discovered any further relevant references, sensitivity and precision calculations, and all the numbers in the PRSIMA flow chart.

Objectives:
1) To complete a SST for a specific systematic review: 'They’ve walked the walk: a systematic review of quantitative and qualitative evidence for parent-to-parent support for parents of babies in neonatal care'.
2) To discover what useful information can be obtained from this systematic review-specific SST.

Methods: we completed a SST using the information from the systematic review along with the searching data held by the information specialist in the team. It was then published as a supplementary file of the systematic review and deposited separately in the University of Exeter’s online repository, ORE.

Results: the SST results show that:
1) of the 13 databases searched, 4 found no included references;
2) of the 3 references found by supplementary searching, one was not in any of the three main databases but two were; the search did not pick them up;
3) the term search term counsellor (truncated and adjacent to the peer terms) was missing from the search strategy, this would have picked up the two papers in MEDLINE;
4) two theses were included in the synthesis, both of which were picked up from database searching;
5) no database found unique references;
6) searching two databases, MEDLINE and PsycINFO, would have picked up all the included references;
7) website searching, handsearching and searching on specific organizations' websites found no further includable references or studies;
8) overall sensitivity was 80 and overall precision was 0.26;
9) the number needed to screen to find one reference to screen at full text was 39;
10) the number needed to read at full text to find one included reference was 10.

Conclusions: SSTs provide evidence at both individual and collective levels for informing systematic review searches. Peer reviewer comments suggest this is a welcome addition to the reporting of systematic review searches; SSTs provide the information in an accessible and concise format which can be readily understood by non-specialists. Useful information has been obtained from the SST in this case study and it will add to the evidence-base on information retrieval methods for systematic reviews. Other information specialists can use the evidence and data in it when developing and running searches for systematic reviews.

SSTs are a further tool to help inform database choices and search methods, it is hoped that they will continue to be completed as part of published systematic reviews to aid transparency and reproducibility of the search methods.

Patient or healthcare consumer involvement: throughout the systematic review