Study did a good job, however......
Although, the study by Evers et al  has been discussed by others
recently, concern about results and conclusions remain.
The authors main conclusion included that the Dutch obstetric system based
on risk selection was not effective and possibly contributes to high
perinatal mortality. This conclusion was based on the findings that
delivery related perinatal death was significantly higher among low risk
midwife supervised pregnancies than among obstetrician supervised
However, we think a serious inconsistency consists that needs to be
Our main concern regards the study design and the results stratified in
low and high risk groups as determinants of obstetric care.
Therefore, we want to refer to the "obstetric indication list"  that
was used for stratification into risk determinants, as presented in the
flow chart (page 3). This list gives indications for referral to an
obstetrician in secondary care in case complications occur (1) during
pregnancy, (2) during labour or (3) in the postpartum period. The reason
for referral during labour is clearly established and includes generally,
signs of fetal distress and unexpected obstetric problems.
As shown by the presented flow chart, about 29% (5492/18686) women were
referred during labour to secondary care according to these indications.
However, the authors stratified the determinants of obstetric care into
women that started labour in primary care and into women that started
labour in secondary care. The incidence of delivery related death was
calculated as such (OR: 2.3, 95% CI [1.1; 4.8]). However, the incidence of
delivery related death in women who started labour and ended delivery in
primary care compared to starting labour in primary care and ended
delivery in secondary care was "not calculated" (table 3). In our opinion
one may question this exclusion, because analyzing these results may be
helpful in answering the research question. After all, women were referred
for serious problems "durante partum" according to the "obstetric
indication list" by first line midwives in order to be treated by second
line obstetricians. So, to investigate medical and maternal misfortune
correctly results in women referred to secondary care during labour should
be assigned to second line care, simply because it was all agreed to do
Having this in mind, calculating data from table 2 shows interesting
results. For instance: Intrapartum stillbirth in primary care (12/13194:
incidence 0.91/1000) compared to secondary care (10/22231: incidence
0.45/1000) shows an OR of 2.02 (95% CI[ 0.86; 4.75]). Higher, but not at a
statistically significant level.
In addition, calculating data from table 3 shows that delivery related
perinatal death demonstrated no difference in newborns delivered in
primary care (14/13194: incidence 1.06/1000) compared to secondary care
(22/22231: incidence 0.99/1000) (OR 1.07, 95% CI [ 0.5; 2.1]).
Besides, delivery related perinatal death is interpreted as a summation of
intrapartum stillbirth and obstetrically related death. According to a
population based study in Scotland by Pasupathy  60 % of delivery
related perinatal deaths were ascribed to intrapartum anoxia and 40% to
other causes. This study by Evers et al (table2) showed 12 (85%)
intrapartum stillbirths from the 14 delivery related deaths in primary
care deliveries, remaining 2 obstetrically (or other causes) related
deaths. Similarly, from the 22 delivery related deaths in secondary care
10 deaths (45%) are assigned to intrapartum stillbirths, remaining 10
obstetrically (or other causes) related deaths. This finding is of
interest because it is in agreement with what may be expected. After all,
obstetrical problems require professional clinical care with
unfortunately, a higher expected perinatal death. So, one may conclude
that: "the midwives did a good job" by referring to the obstetric clinic
correctly. If the authors should have analyzed all the possible
associations their conclusions would have been more balanced and more in
line with previous studies in the Netherlands that showed no elevated
perinatal death or serious perinatal morbidity in low-risk women.
Lastly, the authors did not include late neonatal deaths (> 7days <
28 days) in the perinatal death rates. This is surprising, because deaths
due to events in labour may occur beyond the early neonatal period.
Excluding these findings may have influenced their analyses substantially.
Regarding the above, the conclusion that the delivery related perinatal
death rate of normal term infants was higher in women who started labour
in primary care compared to secondary care delivery should be phrased with
more nuance. Additionally, the final statement that the Dutch system of
risk selection in relation to prenatal death at term is not effective
cannot be concluded from this study. The reverse may be more in line with
the presented findings.
E.Meijer MD, PhD.
Institute for Risk Assessment Sciences, Utrecht University
H 't Hart MD
1. ACC Evers, HAA Brouwers, CWPM Hukkelhoven et al. Perinatal
mortality and severe morbidity in low and high risk term pregnancies in
the Netherlands: prospective cohort study. BMJ 2010;341:c5639. (2
2.Commissie verloskunde van het CVZ. Verloskundig vademecum 2003.
College voor zorgverzekeraars, 2003.
3.Pasupathy D, Wood AM, Pell JP, Fleming M et al. Rates of and
Factors Associated With Delivery-Related Perinatal Death Among Term
Infants in Scotland. JAMA. 2009;302(6):660-668.
4. de Jonge A, van der Goes BY, Ravelli AC, Amelink-Verburg MP, Mol
BW, Nijhuis JG, Bennebroek Gravenhorst J, Buitendijk SE. Perinatal
mortality and morbidity in a nationwide cohort of 529,688 low-risk planned
home and hospital births. BJOG. 2009 Aug;116(9):1177-84
5. Whitfield CR, Smith N, Cockburn F, Gibson A. Perinatally related
wastage: a proposed classification of primary obstetric factors. Br J
Obstet Gynaecol. 1986;
Competing interests: No competing interests