STATISTICAL REPORTING IN PHARMACEUTICAL PAPERS FROM
ROMANIAN JOURNALS
DANIEL-CORNELIU LEUCUȚA1, TUDOR DRUGAN1*, ANDREEA FARCAŞ2, ANDREI ACHIMAȘ1
1Department of Medical Informatics and Biostatistics, “Iuliu Haţieganu”, University of Medicine and Pharmacy Cluj-
Napoca, 6 Louis Pasteur, 400349, Cluj-Napoca, Romania
2Drug Information Research Centre Cluj-Napoca, “Iuliu Haţieganu” University of Medicine and Pharmacy Cluj-Napoca,
Pasteur Street No. 6, 400349, Romania
*corresponding author: tdrugan@umfcluj.ro
Download Full Article

Abstract:
The quality of the statistical reporting is poor in various biomedical journals, this being a problem stressed by numerous
studies. Our aim was to evaluate all pharmaceutical papers published in six Romanian journals, in 2013, for statistical
reporting. We used mainly the “Statistical Analyses and Methods in the Published Literature” – SAMPL - guideline to assess
the statistical reporting in the respective original pharmaceutical papers. We found 170 original articles from which 134
(78.82%) reported statistical results. The most frequently used tests were ANOVA (50.9%), and t-test (37.7%). Confirming
that the assumptions of the test were met by the data was not reported in 87.6% of the papers. The most important error found
in the papers using hypothesis testing, was choosing the wrong statistical test (39.6%). The findings were similar to other
published studies. The quality of reporting statistical analyses of pharmaceutical papers published in Romanian journals
should be improved, in order to reduce the errors in studies, improve the methodology and clarity.

Top