Why 196 Cochrane Systematic Reviews were withdrawn in Cochrane Database of systematic review?

Article type
Authors
Ma JC1, Sun X1, Zhao Y1, Tian JH1
1Evidence Based Medicine Center, School of Basic Medical Sciences of Lanzhou University, China
Abstract
Background: Cochrane Systematic Review (CSRs) is the leading resources of the medical care systematic evaluation.The number of CSRs published has increased steadily over the past decade, at the same time, the number of withdrawal CSRs has also increased too. What led to the CSRs withdrawan from Cochrane Database of Systematic Review?

Objectives: To access the reasons of withdrawn of CSRs, so as to search countermeasures to reduce quantities of withdrawn reviews and improve the quality of CSRs.

Methods: An electronic literature search of all CSRs from inception to December 2012 was conducted using the Cochrane Library, with clicking the following search terms ‘By Review Group’. Two reviewers independently determined CSRs eligibility and extracted the withdrawal reviews information and search details. Disagreements were resolved by the third author.

Results: 196 studies were included, which account for 2.74% of the whole CSRs (7143) ;The CSRs are composed of 53 review groups, of which 6 groups have more than 6 withdrawn reviews, 12 groups have 3–5 withdrawn reviews; 15 groups have 1 to 2 withdrawn reviews, and 18 groups have no withdrawn reviews. Table 1 shows that the largest proportion of withdrawn reviews group is: Prostatic Diseases and Urologic Cancers Group (8.89%); followed by Menstrual Disorders and Subfertility Group (8.84%), Pregnancy and Childbirth Group (8.45). There were seven reasons for withdrawn, 43.88% of the withdrawn CSRs were out-of-dated reviews, and 29.08% of the withdrawn CSRs were updated by other CSRs (Table 2).

Conclusions: This survey showed that the whole reasons of withdrawn of CSRs. The main reason for the withdrawn is because the reviews were out-of-dated, and some of CSRs were withdrawn owing to the absence of abstract or detection of errors, which can be avoided by frequent update and strict check. Therefore, efficient measures should be taken to reduce the portion of withdrawn reviews.