Repositório BVS

OBJECTIVE:

LILACS is a triple language database edited by BIREME and PAHO in Sao Paulo, Brazil, with interfaces in Spanish, Portuguese and English. This database indexes more than 670 medical journals in Latin America and Caribbean. Only 40 of these are indexed also in MEDLINE. So, LILACS is a unique source of scientific information [www.epm.br/cochrane/lilacs.htm]. If we believe that a comprehensive, unbiased search is one of the key differences between a systematic review and a traditional review, (like stated by the Cochrane Reviewers' Handbook), and if the LILACS database came to proof to be an useful source of studies that fit the inclusion criteria of systematic reviews, this database must be included among those searched in Cochrane systematic reviews.

METHODS:

We searched the Cochrane Database of Systematic Reviews using the term "Neoplasms" in order to identify complete reviews. For each review, we sought if the authors had already searched LILACS. If not, we performed a LILACS search in order to try to find articles that could fit the inclusion criteria of the review. For randomized studies (RCT), we used the optimal search strategy for RCT for use in LILACS described by Castro et al (1999). For others methodological designs, we used free search.

RESULTS:

We retrieved 45 completed systematic reviews. None described a search in LILACS. In our search, we found articles that may fit the inclusion criteria for 23 (51%) of these reviews. In four reviews (9%) we found articles that could fit the inclusion criteria, but we were not able to make a clear assertive. In the others 18 reviews (40%) we found no articles that could fit the inclusion criteria. So, in 60% (27/45) of this sample of Cochrane reviews a search in LILACS could be useful. If Cochrane reviews are really intending to include as many articles as possible, a LILACS search must be mandatory. Editors of review groups and peer-reviewers must take attention to the inclusion of this database among those searched. The access to and the search in LILACS is easy and improving LILACS is a internet free access database at www.bireme.br. There is an optimal search strategy for clinical trials already described (Castro et al 1999), and the Brazilian Cochrane Center is conducting a process of re-tagging] RCT in the appropriate publication type.

CONCLUSION:

Cochrane systematic reviews must use LILACS like routine searched database