[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
Purchase Options:
[Skip to Content Landing]
Citations 0
Letters
December 20, 2000

Non-English Reports of Medical Research—Reply

Author Affiliations
 

Stephen J.LurieMD, PhD, Senior EditorIndividualAuthorPhil B.FontanarosaMD, Executive Deputy EditorIndividualAuthor

JAMA. 2000;284(23):2996-2997. doi:10.1001/jama.284.23.2993

In Reply: We agree with Dr Bassler and colleagues that the exclusion of non–English-language studies may lead to bias in meta-analysis results. Our MEDLINE search included all languages and revealed 25 publications of potential interest that were not in English. However, we felt that excluding these publications would not lead to significant bias in our meta-analysis. Of the 15 non-English studies that we were able to retrieve, none met our inclusion criteria. The remaining studies were either published prior to 1985, the earliest date that an English-language study was identified as meeting our inclusion criteria, or had an English abstract available online indicating that the study was a case series and therefore could not provide relative risks (RRs).

In their search of the non-English literature, Bassler et al found a single study1 that met our inclusion criteria. However, this study was published in 2000, and thus could not have been found in our search (1966-1999). Two other pertinent studies published in 2000 also have come to our attention. These found no significant association between clinical chorioamnionitis and cPVL2 or cerebral palsy.3 When all 3 new studies1-3 are added to the meta-analysis, the updated summary RR for clinical chorioamnionitis and cPVL is 1.7 (95% confidence interval [CI], 1.1-2.6) and for cerebral palsy, 1.8 (95% CI, 1.4-2.4).

Bassler et al mention an important source of "language bias" found in clinical trial research published by investigators from German-speaking European countries. Whether such a bias exists for observational studies originating in other non–English-speaking countries is not clear. Of note, our meta-analysis included 7 studies that were performed in countries where English is not the primary language. Five of these studies reported results that were not statistically significant.

We believe that the exclusion of non–English-language studies did not introduce significant bias to our meta-analysis. However, we agree that unless otherwise justified, meta-analyses should include publications in all languages and that literature searches for meta-analyses should include EMBASE, a database that is more inclusive of studies performed in European countries.

References
1.
Bauder  FHvon Siebenthal  KBucher  HU Sonographisch nachgewiesene periventrikuläre Leukomalazie (PVL): Inzidenz und assoziierte Faktoren in der Schweiz 1995-1997 [Ultrasonically established cystic periventricular leukomalacia (PVL): incidence and associated factors in Switzerland 1995-1997].  Z Geburtsh Neonatol. 2000;204:68-73.Google Scholar
2.
Ng  EAsztalos  ERose  TBeyene  JWylie  LDunn  M The association of clinical and histologic chorioamnionitis (CA) with cystic periventricular leukomalacia (cPVL) and cerebral palsy (CP) in preterm infants [abstract].  Pediatr Res. 2000;47:318A.Google Scholar
3.
Gray  PHJones  PO'Callaghan  MJ Case-control study of maternal antenatal risk-factors for cerebral palsy in extremely preterm infants [abstract].  Pediatr Res. 2000;47:318A.Google Scholar
×