Background:

The importance of searching trials registers to identify clinical studies has increased in recent years. This particularly applies to studies starting in 2005 or later, as since then, several initiatives for study registration have been implemented, including legal regulations. The German Institute for Quality and Efficiency in Health Care (IQWiG) conducts early benefit assessments of newly approved drugs on the basis of dossiers submitted by pharmaceutical companies. This includes an evaluation of whether all relevant studies were reported in the dossiers. In this context the question arises of the relevance of searching trials registers.

Objectives:

To determine whether a search in trials registers is sufficient to identify all relevant randomized controlled trials (RCTs) of a new drug.

Methods:

We considered all dossiers on new drugs submitted for early benefit assessment in Germany between January 2011 and June 2014, except dossiers on orphan drugs.
We extracted the RCTs presented in the dossiers’ tables and analysed the related information sources (bibliographic databases, trials registers) reported by the companies.

Results:

We extracted 362 studies from 68 dossiers: 203 studies investigated a new drug, 192 (95%) of which were registered in ClinicalTrials.gov. No entry was reported for any trials register for 11 (5%) studies; after evaluation, two of them turned out to be non-RCTs; nine were RCTs but had been completed before 2005.

Conclusions:

RCTs of new drugs starting in 2005 or later can usually be found in ClinicalTrials.gov. In most cases, a search in trials registers seems sufficient to identify these studies. However, this conclusion refers only to the identification of relevant studies, not to the identification of the related study data. This still requires searches in other sources, such as bibliographic databases and unpublished company documents.