Characteristics of meta-analyses related to acceptance for publication in a medical journal.

Editors of medical journals select manuscripts for publication based, in part, on the perceived quality of the manuscript submitted. The objective of this study was to describe associations between acceptance for publication and quality-related methodologic characteristics of meta-analyses. This was a prospective observational study. The setting was editorial offices of JAMA and offices of external reviewers. The manuscripts reviewed were 112 consecutive meta-analyses submitted to JAMA during 1996 and 1997 whose authors agreed to participate. The main outcome measures were ratings of 16 methodologic characteristics reflecting quality of the meta-analysis and acceptance for publication. A "high" rating for one methodologic characteristic, whether the report of the meta-analysis provided sufficient detail to enable replication, was related significantly to publication (RR = 2.79, 95% CI = 1.13-6.89). This relationship persisted when other variables were controlled for in the model. Generally, rejected manuscripts had fewer factors rated "high," but differences were not significant. We found that inclusion of sufficient detail to allow a reader to replicate meta-analytic methods was the only characteristic related to acceptance for publication.