Meta-analyses indexed in PsycINFO had a better completeness of reporting when they mention PRISMA

Abstract

Objectives: To investigate the effect of the explicit mention of PRISMA, a statement designed to help authors report meta-analyses (MAs), on the reporting completeness of MAs.

Study design and setting: Two investigators evaluated a random sample of 206 MAs indexed in PsycINFO in 2016; 100 explicitly mentioned PRISMA and 106 did not. Two authors independently evaluated the 27 PRISMA items and extracted factors that could potentially be associated with reporting completeness. The data were analyzed descriptively.

Results: Among our 206 MAs, perfect adherence to PRISMA was found in less than 4%, of which 87% explicitly followed PRISMA. The following items were encountered significantly more frequently in MAs that explicitly mentioned PRISMA than in those that did not: summary, protocol, information sources, search strategy, study characteristics, results of individual studies, funding, study selection, risk of bias in individual studies, and bias across studies. The journal's impact factor, endorsement of PRISMA by the journal, number of authors, country of author, open access, and design of the included studies were significantly and positively associated with the explicit mention of PRISMA.

Conclusions: Even if far from optimal, the explicit mention of PRISMA has a positive influence on the reporting completeness of MAs from PsycINFO.

Keywords: Completeness; Meta-analysis; Meta-research; PRISMA statement; Reporting.