[A decision problem in perinatology: continuous monitoring of the fetal heart during labor].

Review of evaluative studies on electronic fetal monitoring (E.F.M.) during labour shows that the conclusions of the five published randomized controlled trials have not been applied. There is a discrepancy between the widespread use of the method and the lack of proof of its efficacy in the controlled trials. This discrepancy can be related to methodological problems (insufficient number of cases, judgment criteria, compared types of care). It also shows that the conclusion concerning the possible net benefit or net harm of E.F.M. depends on the absence of such benefit, on the unit chosen to combine advantages and disadvantages. Based upon the same set of estimates for advantages and disadvantages the conclusion is in favor of E.F.M. when the chosen unit is utility whereas the conclusion is different in a cost-benefit analysis.