Send to

Choose Destination
Therapie. 2002 Mar-Apr;57(2):175-80.

[Methods of evaluation of risks tied to drug exposure during pregnancy. Advantages and limits of epidemiology, interpretation of results].

[Article in French]

Author information

Registre France Centre/Est de Malformations Congénitales, Lyon, France.


Various epidemiologic designs permit an evaluation of drug teratogenicity, but none is sufficient to establish the risk. Only concordance of results of several studies lead a substance to be considered teratogenic. The usual designs of analytical epidemiology can be used: prospective studies of total populations, case-control, cohort. Case series may be useful when no good epidemiologic study is available. Most of those published were collected by teratogen information services (TISs). These series of exposures are not representative samples of the general population, and when interpreting the pregnancy outcomes, it is difficult to define an unexposed group for comparison. In practice, if 50 pregnancies are reported with early exposure to a given drug and without unwanted effects, one can exclude a strong teratogenic effect, and reassure a woman inadvertently exposed. TISs have published several of such exposure series, and they are often used. On the contrary, if the question arises whether or not to prescribe a drug to a woman in early pregnancy, a much larger sample of pregnancy with known outcomes is needed to answer "yes": this is a situation of population risk assessment, because one would give the same answer to all similar questions, which may be many. Birth-defect monitoring systems are in a better position to do this and several examples can be given (valproate and spina bifida, corticosteroids and orofacial clefts).

[Indexed for MEDLINE]

Supplemental Content

Loading ...
Support Center