The scientific community assumes that rigorous methodology research is more likely to be published in high impact psychiatry journals (HIJ). We aimed to test which methodological variables could predict publication in HIJ.

We conducted a systematic review of the MEDLINE and EMBASE databases from 2013, January 1st to 2015, June 15th. Inclusion criteria were studies that were RCTs whose at least one arm of the study should be fluoxetine regarding adult patients (>18 years old) with MDD. We performed logistic regression regarding the number of participants, intention-to-treat analysis, blinding, multicenter study, sample losses, positive result, sponsorship of pharmacy's industry, and h-index of the last author. A HIJ was considered if journal impact factor was above the median or 3rd quartile of our sample.

Forty-two studies were considered for the final analysis. The results of the univariate logistic regression found no differences between HIJ and low impact psychiatry journals for all methodological variables, except the h-index of the last author. By considering HIJ when impact factor was above the mean, h-index had an odds ratio=1.09 (1.01–1.17), P=0.02; considering HIJ when impact factor was above the 3rd quartile, h-index had an odds ratio=1.07 (1.01–1.14), P=0.02.

Our results indicate that the author productivity may be a relevant predictor for publication in a HIJ in the psychiatry/psychology field. Our study proposes that journals focus on identifying what are the relevant criteria for publication approval in the peer-review process.