For several decades, visual analysis of the antepartum fetal heart rate tracing has been a primary mode of fetal surveillance despite lack of proof from randomized trials that its use leads to a reduction in perinatal mortality in high risk pregnancies. A meta-analysis of these trials offered the same conclusion. The trials were performed in the early 1980's with fetal monitors with poor signal to noise ratios and poorer testing schemes in comparison to today. Therefore the results may no longer be pertinent, however, it is unlikely that these trials will be repeated because of present medico-legal constraints. Nonetheless, the use of non-stress testing alone is an accepted mode of fetal surveillance because of associated low fetal death rates within 1 week of a normal test; 1.9 to 3.2/1000 patients tested. The expense associated with this low fetal death rate is a low positive predictive value of 25% to 50%. Interrelated with this inefficiency, is the considerable inter- and intra-observer variation in the interpretation of antepartum fetal heart rate tracings. Variation in the interpretation of antepartum fetal heart rate tracings continues to exist despite the increase in experience and knowledge of fetal heart rate monitoring and improvement in fetal heart rate monitors. The end result is a form of fetal surveillance that is of questionable value and inefficient, requiring frequent back up testing to clarify fetal condition.