A crusade against home birth.

A recent study by Grunebaum et al examined the relationship between place of birth and adverse neonatal outcomes (Apgar of 0 at 5 minutes, and neonatal seizures or serious neurologic dysfunction—hereafter referred to as neonatal seizures) as reported in birth certificate data from 2007 to 2010 for term newborns (n = 13,891,274) (1). Outcomes were analyzed by four practitioner types: hospital physician, hospital midwife, freestanding birth center midwife, and home birth midwife. The authors claim that babies born at home and in freestanding birth centers were at a significantly higher risk of having a 5-minute Apgar score of 0 (RR = 10.55 and 3.56, respectively) and neonatal seizures or serious neurologic dysfunction (RR = 3.80 and 1.88). However, these findings must be interpreted with caution for several reasons. Limitations of birth certificate data for epidemiologic analysis have been widely discussed in the literature, and include concerns about the completeness and accuracy of reporting of specific items on birth certificates, and the inability of birth certificates to provide longitudinal information (such as for planned home births that transfer to the hospital) or information on clinical intentions (2–4). The neonatal seizure variable, for example, is one of several medical variables unreliably reported on birth certificates (4–6). Two detailed studies comparing birth certificate data to medical records in New Jersey and Tennessee yielded sensitivity rates for neonatal seizures of 0.226 and 0.182, respectively (5,6). This means that approximately 80 percent of cases of neonatal seizures identified on medical records are not reported on birth certificates. Data of this poor quality should not be used as the main outcome measure in any study. Although reporting of data on 5-minute Apgar scores in broad categories (such as <7 or ≥7) is a bit better (7), no studies have examined the validity of reporting of 5-minute Apgar score = 0. However, there is substantial evidence that the reporting of this item on birth certificates is very problematic. Watterberg found that although large differences existed between home, birthing center, and hospital settings for reported Apgar scores of 0 and 10, these differences were greatly reduced for Apgar <4, and virtually eliminated for the combined category of Apgar 9 or 10 (8). There appear to be real differences between how physicians and home and birth center midwives perceive and report Apgar scores at the edges of the Apgar spectrum. Physicians are more likely to report fine gradations of either very low or very high Apgar scores, whereas home and birth center midwives are more likely to report Apgar scores of 0 or 10 more absolutely. Apgar score <4 is the more commonly used measure of early neonatal compromise, and has the added advantage of providing greater numbers of cases for analysis. The reported odds ratios for 5-minute Apgar score of 0 and neonatal seizures in the Grunebaum et al study are based on very small numbers of cases, and thus have limited generalizability or clinical relevance. It is also well-established that Apgar scores are poor predictors of neonatal outcomes (9), so even if these data could be improved, they