The future of small birth units in the UK
(And medium and large ones!)
The fall has occurred entirely among smaller units, those conducting less than 2,000 deliveries annually. As a result a much greater proportion of births take place in larger units.
Liverpool Women’s Hospital with 8,084 deliveries in 2003.
Some people say this is a good thing.
It may be safer to give birth in a large unit with all facilities on
site, and it may be easier to staff a small number of big units than many
smaller ones. It smoothes out the
peaks and troughs in delivery numbers and you can often get away with a smaller
number of staff on site per delivery. Some
staff, typically those who are most technologically orientated, like working in
Others say it is a bad thing. Women have to travel further to deliver. Big units may be more impersonal and over-interventionist. They may be difficult to manage. Staff may leave difficult problems for others to deal with, if they know there are many other people around. And many staff prefer working in small units, so the staffing problem may actually be worse.
We can argue either way. However, no-one can dispute that centralisation in the NHS has gone much further than elsewhere in Europe.
Twelve English units deliver more than 5000 births per annum and the largest, Liverpool Women’s hospital, delivers over 8,000.
contrast the largest Maternity unit in Germany, the Humboldt maternity
department in Berlin, has just over 3,000 deliveries per year.
Few others have more than 2,000. The
Höchst hospital, the largest maternity unit in Frankfurt had 1,800 deliveries
in 2004. The Jeanne de Flandre Hospital in Lille, the largest
Maternity Hospital in France, has just over 4,000 births a year. The largest
unit in Belgium (Flanders) conducted 2641 deliveries in 2004.
The following table is taken from Wildman et al 2003 supplemented by Dept of Health data 2004 for England, Direction de la Recherche des Etudes de L’Evaluation et des Statistiques (DREES) Etudes et Resultats No 225 Mar 2003 for France and Perinatale Activiteiten in Vlaanderen 2003 for Belgium (Flanders).
* Includes delivery units >4,000 not reported separately.
I think this centralisation has gone too far.
Many of the centralising changes in the UK maternity
service have been introduced with the aim of improving safety.
The idea being that large units are better able to provide high quality
neonatal and maternal intensive care without the need to transfer sick babies or
mothers around the country.
A paper from Germany which showed threefold increase in
neonatal mortality between the smallest and the largest hospitals in 2002 is
frequently cited in support of the closure of small units (Heller et al 2002).
However, “small” and “large” are relative terms. The German units were
much smaller than UK; 39 had under 500 deliveries per year, 33 had 500-1000, 14
had 1,000 to 1,500 and only five so called “large units” had 1,500 birth per
year. The actual size of the
“large” units was not given but all five combined delivered only 96,000
women over ten years, i.e. an average of less than 2,000 deliveries each per
year. In England they would be
classified as “small”.
Larger units may also be more difficult to manage
efficiently. Staff may avoid taking
responsibility for clinical decisions or for aspects of organisation in the hope
that others will do it for them. The
Kennedy commission reviewed three hospitals in which concerns had been raised
about maternity care; Northwick Park in London, New Cross in Wolverhampton, and
Ashford St Peters in Chertsey. In all three, serious deficiencies were identified.
These included poor reporting of adverse incidents and poor handling of
complaints, poor staff working relationships, inadequate training and
supervision of clinical staff, services isolated both geographically and
clinically, and staff shortages with poor management of temporary employees.
It is implausible that similar findings would not have been made in many
other hospitals had they been subject to similar detailed review.
So let’s compare the NHS with the rest of Europe.
Crude mortality comparisons.
The NHS has not generally achieved better perinatal or maternal mortality figures than other comparable European countries (OECD 2005). However, it is probably unfair to use routine maternity statistics to compare the quality of care. These statistics are often collected in different ways, and with different degrees of accuracy and many maternal and fetal deaths are either unavoidable or related to social factors outside the control of the maternity system.
Audits of sub-optimal care
A better method is to compare the frequency of sub-optimal care. In 2003 the Euronatal Working Group compared the frequency of sub-optimal care leading to perinatal death in a range of countries (Ricardus et al 2003). An independent audit panel blindly reviewed 1619 perinatal deaths in regions of ten European countries. Sub-optimal care was defined on the basis of the same agreed “evidence-based” criteria for all countries. The percentage of cases graded as having sub-optimal factors present, which either might, or probably did, contribute to the bad outcome, are shown in the table.
Numbers and percentages of evaluated cases of perinatal death graded as “Suboptimal factor(s) identified which might have contributed to the fatal outcome” or Suboptimal factor(s) present which are likely to have contributed to the fatal outcome.
The NHS appears to have the highest rate sub-optimal care, which might have contributed to the deaths. The authors rightly caution against making the inference that substandard practice is really more common in England than other countries. Nevertheless this hardly suggests that the centralisation achieved by the NHS has resulted in better care.
Measured by value more than half of the potential claims for negligent injury in the NHS arise from maternity care. These are predominantly cases of brain damage or other birth injury allegedly cause by substandard care at delivery. Anecdotally such claims are less common in the rest of Europe although I cannot find any comparative data.
from campaigning groups
Tyler 2002 compared maternity campaigning groups in England, the Netherlands and Germany between 1996 and 1997. In England and the Netherlands almost all organisations had originated as genuinely grass roots organisations founded by individuals with negative experience of the services provided. However, the Dutch groups did little or no campaigning and tended to confine their activities to providing support to individual patients. In contrast, campaigning was the reason for their existence for most English groups, and they were generally vibrant organisations with a regular supply of new members. In Germany there were relatively few such grass roots organisations and instead organisation had tended to be driven by maternity care providers. Again, like the Netherlands the groups did little campaigning, but in Germany the reason seemed to be that the organisations were relatively weak with few new members joining. This may reflect personality difference between inhabitants of different countries but plausibly, it also suggests that, judged by their actions, both Dutch and German patients are less dissatisfied with the maternity care they receive than those in England.
Whatever the data on safety may say, closing maternity units reduces parents' choices. This matters because customers exercising choice is one of the most powerful drivers for higher standards. If a unit provides poor care, people vote with their feet and standards either improve or the unit closes. But if there is only one unit in an area, people have no choice and poor units can continue to exist.
are plans to merge many maternity units in the UK.
I believe these are unwise and should be opposed. Contact
me here if you agree.
I'd also be very interested in any comparative data on medico-legal claims
 Wildman K, Blondel B,
Nijhuis J, Defoort P, Bakoula C. European indicators of health care during
pregnancy, delivery and the postpartum period. Eur J Obstet Gynecol Reprod
Biol. 2003;111 Suppl 1: S53-65.
 Heller G, Richardson DK, Schnell R, Björn Misselwitz B Wolfgang Künzel W Schmidt S Are we regionalized enough? Early-neonatal deaths in low-risk births by the size of delivery units in Hesse, Germany 1990–1999 International Journal of Epidemiology 2002;31:1061–1068
OECD Health Data 2005: Statistics and Indicators for 30 Countries.
Accessed Feb 2006 http://www.oecd.org/document/44/0,2340,en_2649_34631_2085228_1_1_1_1,00.html
JH, Wilco C. Graafmans, S. Verloove-Vanhorick P, Mackenbach JP, The
EuroNatal International Audit Panel, 2003 Differences in perinatal mortality
and suboptimal care between 10 European regions: results of an international
An International Journal of Obstetrics and Gynaecology 110:
 Tyler S Comparing the campaigning profile of maternity user groups in Europe – can we learn anything useful? Health expectations (2002) 5: 136-147
[i] MacFarlane and Mugford 1984
[ii] DoH NHS Maternity Statistics, England: 2002-03
Send mail to email@example.com
questions or comments about this web site.