Introduction
What is new?
Key finding- •
The number of reviews assessing the quality of randomized controlled trials (RCTs) has dramatically increased in recent years.
- •
The reporting of the methodology seem to improve over time.
- •
There is an important variability in assessing the quality of RCTs.
- •
Definition of criteria used is often lacking.
What this adds to what was known- •
This is the first overview of all reviews assessing the quality of RCTS.
- •
Such reviews are often well reported.
- •
Nevertheless, how quality is assessed still raises important issues.
What is the implication, what should change now?- •
To be relevant, such reviews should use well-defined and acknowledged criteria for assessing the quality of RCTs.
Randomized controlled trials (RCTs) are recognized as the gold standard to assess the efficacy of health care interventions. However, it has been demonstrated that RCTs are not necessarily unbiased [1], [2], [3]. Some studies have shown that inadequacy of certain important methodological items such as allocation concealment is associated with an exaggeration of the estimated treatment effect [1], [2], [3]. Such bias in the conduct of RCTs may have serious consequences for patients' care and decision making. This is why important concerns about the poor quality of RCTs and the low reporting of important methodological details required to judge quality have been raised, especially from the mid-1990s. These concerns led to the publication of the Consolidated Standards of Reporting Trials (CONSORT) Statement in 1996 [4] and updated in 2001 [1] that aimed to improve the reporting of RCTs by providing guidelines.
Another consequence was the increased publication of reviews that aimed to assess the quality of methods used or the reporting of RCTs. In fact, such reviews of publications have a long history. The earliest we know of was published in 1929 [5] and examined the extent to which “statistical logic” was used in a sample of 200 medical–physiological articles from current American periodicals. However, the number and the range of medical areas covered by these reviews are unknown. Also, there are no recommendations concerning their conduct, covering the search strategy, the selection of the studies to be included in the review, the process of data extraction, and the methodology of quality assessment. We aimed to retrieve and review all reviews assessing the quality or the reporting of RCTs that were published in the last 20 years to evaluate their number, characteristics, and reported methodology over time, considering especially how they assessed the RCTs.