To assess the relationship between protocol registration of systematic reviews published between August 2009 and June 2015 in MEDLINE via PubMed, and PRISMA adherence, as well as outcome reporting bias.
Level of Investigation
Analytical
Summary of Findings
21% of the 284 included systematic review registered or published their protocol. The association between PRISMA adherence and protocol registration was not statistically significant. The association between outcome reporting bias and protocol registration was not statistically significant.
Number of systematic reviews included
284
Number of eligible systematic reviews assessed
1584
Treatment impacted
No
Treatment impacted description
Interpretation impacted
No
Interpretation impacted description
Does the article have a publicly available protocol?
Yes
Did the review team include both methodological and clinical experts?
Not Applicable
Sampling strategy for obtaining the systematic reviews under assessment
Comprehensive searches on a particular topic (clinical or methodological)
Was the sample size of systematic reviews subject to any formal sample size or power calculation?
No
Were any or all of the study selection and data extraction performed in duplicate?
Yes
Were any statistical adjustments made to correct for confounding?
No
Are the methods of the article described in enough detail to replicate the study?
