To the Editor: In their national analysis, Robson and colleagues found less favourable outcomes among term singleton babies born in public hospitals than in private hospitals.1 Most health services research is non-randomised, and it is unrealistic to expect studies such as this to be as internally valid as a randomised controlled trial.2 As noted by others,3,4 in the absence of randomisation, it is difficult to untangle the myriad differences between mothers, babies, and the care provided in private and public hospitals (ie, the results might be subject to confounding).

Robson et al statistically accounted for the potential confounding effect of maternal smoking (although these data were only available for about half the mothers), age, Indigenous status, parity, diabetes, hypertension, rurality and method of birth. We replicated their analysis using the Queensland Perinatal Data Collection (July 2005 – December 2007), which included virtually complete data on maternal smoking. Our analysis of 124 300 term singleton babies gave an adjusted odds ratio (AOR) for perinatal mortality of 2.0 (95% CI, 1.4–2.7), which is similar to that reported by Robson et al.

Using more detailed data from the Queensland dataset, we statistically adjusted for other potential confounders — including pre-existing and gestational diabetes, pre-existing and pregnancy-induced hypertension, pre-eclampsia, eclampsia, antepartum haemorrhage, anaemia, depression, urinary tract infection, low birthweight (< 2500 g), socioeconomic status (based on area of usual residence5), alcohol and drug misuse, and artificial reproductive technology — and obtained an AOR of 2.1 (95% CI, 1.5–2.9). We are not implying that adding more and more variables to a statistical model is an appropriate way to account for confounding. Our aim is simply to show that Robson et al’s result is robust to statistical adjustment using the available data; this is not the same as saying the analysis is robust to confounding.

We found that the higher perinatal mortality in public hospitals was greater for neonatal deaths (AOR, 3.1; 95% CI, 1.8–5.6) than for stillbirths (AOR, 1.6; 95% CI, 1.0–2.4). Excluding lethal congenital anomalies4 did not materially change the result (AOR, 1.9; 95% CI, 1.3–2.8). We also stratified our analysis by level of hospital — tertiary referral (neonatal intensive care unit), base (special care nursery), and community — and obtained a similar twofold mortality excess in each stratum.

Although perinatal mortality is uncommon among term singleton babies (1 in 1000 in private hospitals versus 2 in 1000 in public hospitals), any excess risk should be investigated and the reasons for it understood. The results from Robson et al’s article might be due to confounding, but they should not be dismissed and should be investigated with more detailed clinical data. Even if all the excess risk is due to confounding, explicit confirmation of this would be extremely useful. To this end, the Statewide Maternity and Neonatal Clinical Network in Queensland Health will undertake clinical review and classification of term singleton deaths according to national guidelines6 and collaborate with the Australian Maternity Outcomes Surveillance System (AMOSS)7 to enhance prospective monitoring of late gestation perinatal deaths nationally.

To the Editor: Robson and colleagues found higher crude odds of perinatal death in public hospitals — an unsurprising finding, given the maternal demographics in the public system.1 However, after taking into account known risk factors for poor perinatal outcome by “adjusting for the potentially confounding variables available in the NPDC [National Perinatal Data Collection]”, the authors found that the odds ratio actually went up — implying that the pregnancies of mothers in the private system are higher risk. This is implausible.

There are a number of problems with the analysis. The possibility that potentially important maternal information was not included has been acknowledged.1,2 Of greater concern is the omission of a history of low birthweight from the regression model. Since intrauterine growth retardation is a very strong risk factor for perinatal death,3 and since the rate of low birthweight in the public hospitals was double that in the private hospitals, this omission is puzzling.

Furthermore, while the authors’ stated aim was to assess the effect of the private model of obstetrician-led interventional care on perinatal outcomes, they included the method of birth and hospital type as independent explanatory variables. Since the interventional nature of obstetric care is, as acknowledged by the authors, a key component of the model of care provided in a private hospital, adjusting for this variable will lessen the usefulness of the study in assessing the impact of private hospital care on perinatal outcomes. Put another way, in deciding what is best for her baby, a mother who is considering giving birth in the private system cannot cherry-pick only the non-interventional side of the obstetrician-led model of care — she must adopt the entire package.

Including method of birth as a separate explanatory variable artificially inflates the apparent “riskiness” of the pregnancies of mothers presenting to private hospitals to give birth. These mothers had a 75% increased incidence of caesarean section; however, most of these procedures would be elective rather than emergency. Since this distinction was not made, the women in private hospitals appear high risk despite having had a caesarean section for a low-risk pregnancy. This will have the effect of making perinatal outcomes in private hospitals appear more favourable than they actually are.

As currently analysed, the data are of little value to prospective mums and dads in making the first, and one of the most emotive, of the many decisions they need to make as parents.

To the Editor: Of the many problems with the study by Robson and colleagues,1 the one that concerns us most is the outcome of perinatal death. The headline that this is twice as high in public hospitals has the potential to scare those who have no choice about where they give birth. Both the public and the medical community have a right to expect that this outcome is rigorously and accurately represented. We do not believe this to be the case.

The perinatal death rate is made up of stillbirths and neonatal deaths in the first 28 days of life. Stillbirths will comprise 70% of all perinatal deaths, and more than 90% of stillbirths after 37 weeks’ gestation will occur before the onset of labour.2,3 It cannot be excluded that this relates to model of care but, because less than 10% of these stillbirths will have occurred during delivery, it is most certainly not related to the type of birth hospital or the interventions performed during labour. Most likely, it reflects differences in intrinsic risk that have not been controlled for by Robson et al.1

The neonatal death component of this statistic is probably confounded by ascertainment bias. Most neonatal deaths after 37 weeks’ gestation occur for two reasons: intrapartum hypoxia–ischaemia and severe congenital abnormalities. When the latter are diagnosed antenatally, the mothers will often be transferred to tertiary public hospitals for a second opinion and further management. Babies with undiagnosed lethal abnormalities or severe intrapartum hypoxia will invariably be transferred postnatally to a tertiary public hospital neonatal intensive care unit, where death often occurs after withdrawal of intensive care. The midwives data collection form is filled in at the birth hospital, and the baby is categorised as “discharged”, “transferred” or “died”. The 2006 national perinatal data collection report specifically cautions:

Thus, it is unlikely that the difference in perinatal mortality described by Robson et al1 has anything to do with the birth hospital. To represent it as such in the public arena is irresponsible. Further, for the MJA to misrepresent “perinatal mortality” as “neonatal mortality” in the media release about this article defies belief.

To the Editor: The study by Robson and colleagues1 predictably produced alarming media headlines. However, we feel the study contains potential inaccuracies that seriously undermine the conclusion that the baby toll is lower in private hospitals.

Did the authors ascertain the true impact of fetal abnormality or conditions requiring transfer to tertiary care? Women with threatened preterm labour often transfer to the public system and subsequently remain under the care of public high-risk teams. Babies with a known congenital anomaly detected antenatally by private obstetricians may be similarly transferred. This highest level of care is only available in public hospitals, usually those with dedicated subspecialist services in maternal–fetal medicine, neonatology or neonatal surgery. When babies who have been transferred from private hospitals to the public neonatal intensive care system subsequently die within 28 days of birth, this may be recorded as a public hospital death.

The authors stated that the “major adverse outcome associated with social disadvantage is low birthweight” but that assessment of the effect of social disadvantage was beyond the scope of their study. Not accounting for the effect of low birthweight is unacceptable. The National Perinatal Data Collection records birthweight accurately and is easily used to remove the bias of low birthweight.

An internationally accepted indicator of optimal perineal outcome is the rate of intact perineum. This is known to be lower in private hospitals but was curiously not reported. Further, a prospective data audit in a tertiary hospital revealed that where episiotomy extends to third- or fourth-degree tear, this is only reported as episiotomy.2 Thus, with twice the rate of episiotomy in private hospitals,3 and its association with severe perineal trauma, this outcome could be seriously under-reported in Robson et al’s study.

The authors identify the significantly higher rate of caesarean sections in private hospitals as one of the measurable benefits of this model of care and dismiss any concerns as “orthodoxy”. They have overlooked the significant morbidity and mortality among term neonates associated with the rise in caesarean rates, both elective and following labour. Research using detailed Australian and New Zealand Neonatal Network data to ascertain admission rates of babies of ‘‘low-risk’’ women to neonatal intensive care units and special care nurseries not only found higher rates of admission across each gestational age group after elective caesarean section but also reported the death of two infants.4 Such morbidity and mortality is arguably the most important differential of care experienced by women who give birth in private or public hospitals.