prices (to the disadvantage of other general practition-
ers and patients) are therefore disappearing. The
national tariff should also reduce transaction costs. Pri-
mary care trusts will act as the contracting agent for
commissioners in practices, streamlining the process
and reducing costs still further and the additional costs
of practice based commissioning will be funded from
savingsagainst the budget. Unlike
doctors,the new practice based commissioners will not
be legally autonomous and will remain under the aegis
of primary care trusts.This will provide an opportunity
to marry the dynamism of devolution with the strategic
planning and public accountability of primary care
trusts. Practices that commission services in the new
scheme are less likely to be seen as “wild cards” as they
were in the 1990s.12
Nor will the patients of these new commissioners
enjoy access to a far wider range of services. There are
now guarantees of greater choice for all patients and,
from 2008, patients will have unrestricted choices to
see any provider meeting minimum quality standards
and the NHS price tariff.1Furthermore, since the
demise of fundholding, the NHS has worked to
decrease variation in services across the country.
National service frameworks, guidelines from the
National Institute for Clinical Effectiveness,inspections
by the Healthcare Commission, and new national core
and developmental quality standards are all aiming to
regulate the services available to NHS patients. This
new environment might allay the fears of all save the
most hardened critics of fundholding.
Richard Q Lewis visiting fellow
King’s Fund, London W1G 0AN
Competing interests: The author has advised NHS organisa-
tions on practice based commissioning as a paid consultant.
1Department of Health. NHS improvement plan—putting people at the heart of
public services. London: DoH, 2004.
Department of Health. Practice based commissioning. Engaging practices in
commissioning. London: DoH, 2004. (http://www.dh.gov.uk/assetRoot/
04/09/03/59/04090359.pdf (accessed 6 Oct 2004).
Lewis R. Primary care led commissioning—harnessing the power of the
Commission for Health Improvement.What CHI has found in primary care
trusts—sector report. London: CHI, 2004.
Department of Health. Reforming NHS financial flows.Introducing payment
by results. London: DoH, 2002.
NHS Alliance/Primary Care Report. Clinician engagement: a national
survey. Retford: NHS Alliance, 2003.
HM Treasury. Public service agreements 2004. www.hm-treasury.gov.uk/
1 October 2004).
Goodwin N. GP fundholding. In: Le Grand J, Mays N, Mulligan, JA, eds.
Learning from the NHS internal market—a review of the evidence. London:
King’s Fund, 1998.
Propper C, Croxson B, Shearer A. Waiting times for hospital admissions:the
impact of GP fundholding. CMPO Working Paper Series No 00/20, 2000.
10 Dusheiko M, Gravell H, Jacobs R, Smith P. The effects of budgets on doctor
behaviour:evidence from a natural experiment,discussion paper. York: Depart-
ment of Economics, University of York, 2003.
11 Baxter K, Bachmann M, Bevan G. Primary care groups: trade-offs in
managing budgets and risk. Public Money Manage 2000;Jan-Mar:53-62.
12 Glennerster H, Matsaganis M, Owens P. Implementing GP fundholding:
wild card or winning hand? In: Robinson R, Le Grand J, eds. Evaluating
the NHS reforms. London: King’s Fund, 1994.
Safety of antipsychotic drugs for pregnant and
breastfeeding women with non-affective psychosis
Multicentre cohort studies are needed as randomised trials are impractical
affective psychoses are less fertile than controls, partly
because of hyperprolactinaemia secondary to anti-
psychotic drugs.3 4The use of atypical drugs such as
clozapine and olanzapine, which do not have this
effect, may increase fertility rates. Weighing risks
against benefits in treating pregnant and breastfeeding
women with antipsychotics requires assessment of
clinical effectiveness versus risk of toxicity to mother,
fetus, newborn, and the developing child. Withholding
antipsychotic treatment may expose mother and fetus
to more harm than benefit as, in addition to
behavioural disturbance which may put both at risk,
physiological changes associated with psychosis could
affect fetoplacental integrity and development of the
central nervous system. Subsequently, poor clinical
outcomes for mothers may result in poorer interaction
between mother and infant.5But the impact of antipsy-
chotic medication on the fetus is also unclear as no
randomised controlled trials have evaluated the use of
antipsychotics in pregnancy.6
The only large controlled studies of antipsychotics
in pregnancy have been conducted on women with
round 2% of women develop a non-affective
psychotic disorder, and more than half of
them have children.1 2Women with non-
medication (generally drugs such as phenothiazines)
are used for these patients than for patients with
schizophrenia, and residual confounding is common-
place. Conclusions are therefore limited. A meta-
analysis reported that exposure to low potency
antipsychotics during the first trimester was associated
with a small additional risk of congenital anomalies.7
Antipsychotics can also produce toxic effects in
newborn infants medicated in the womb, including
respiratory depression and neonatal behavioural
abnormalities such as extrapyramidal movements
and difficulty with oral feeding, although these
effects usually resolve within days. The few small
studies examining longer term neurobehavioural
effects in children exposed to antipsychotics in the
womb have reported no behavioural or intellectual
Even less is known about any adverse effects of
atypical antipsychotics. On comparing case registry
data (n = 37) with historical controls the rates of
adverse effects with olanzapine during pregnancy were
not found to be higher.8However, data from the
National Teratology Information Service indicate an
increased incidence of malformations of 10% (com-
pared with an expected incidence of 2-3%).9However,
as only three cases were seen, no clear relation can be
BMJ VOLUME 329 23 OCTOBER 2004 bmj.com
established.9One survey of pregnant women taking Download full-text
clozapine reported a higher than expected rate of mal-
formations (five cases in 61 pregnancies).10Reduced
variability of the fetal heart rate before and during
labour has also been noted in one woman taking
clozapine.11Case reports and data from the National
Teratology Information Service do not indicate a prob-
lem with quetiapine and risperidone during preg-
nancy. No data are available on newer drugs such as
aripiprazole. An increased risk of neural tube defects
has also been proposed as many atypical antipsychotics
are associated with weight gain and subsequent folate
deficiency.12This association has not been investigated
Thus very little evidence exists regarding the use of
antipsychotics during pregnancy. When a woman with
schizophrenia presents pregnant, wanting to come off
medication, many psychiatrists will be concerned
about the high risk of relapse during pregnancy if
medication is discontinued. Most will suggest that ide-
ally she should be maintained on the lowest possible
dose of antipsychotic during and after pregnancy, with
fetal screening throughout pregnancy. The woman is
likely to be concerned about the risk of medication to
the fetus. But neither the psychiatrist nor the pregnant
woman will be able to find data on the size of the effect
of these risks.
Pharmacological treatment after delivery depends
on the mother’s symptoms and whether she is breast
feeding. Antipsychotic drugs are excreted in breast
milk, but to date breast fed infants have not shown
signs of toxicity or impaired development in most case
reports of antipsychotics, although manufacturers
advise avoidance of these drugs during breast feeding.
Recent recommendations from the World Health
Organization and the UK government, stressing the
desirability of breast feeding exclusively for six months,
make it imperative that women know whether they can
do this safely while on medication.
Littleis known about
adverse effects attributable to olanzapine, but long
term data are required.10 13Dev and Krupp reported on
four infants breast feeding from mothers taking
another agranulocytosis, which resolved with dis-
continuation of breastfeeding.10Mothers on clozapine
should thus not breast feed. The evidence for prescrib-
ing antipsychotic medication while breast feeding is
therefore based on even fewer reports. Although
mothers are advised to continue the same medication
given during pregnancy when breast feeding, to avoid
drugs with a long half life,and to time feeds to coincide
with trough concentrations of drugs in breast milk, this
is based on common sense and not on any high level
The lack of evidence is partly attributable to ethical
constraints restricting randomisation of pregnant and
breastfeeding women. In addition, drug companies do
not usually sanction the evaluation of antipsychotics in
pregnant or breastfeeding women in randomised con-
trolled trials. This raises ethical and clinical dilemmas
regarding appropriate management of an already vul-
nerable group of women.
Many women are likely to be taking atypical anti-
psychotic drugs at conception. This provides an
opportunity to investigate short and long term adverse
effects of antipsychotic drugs on pregnant women and
their offspring, including neurobehavioural outcomes
in these children. Although randomised controlled tri-
als may be impractical, coordinated multicentre
evidence. Obstetricians, psychiatrists, and general
practitioners need robust evidence on which to base
prescribing of antipsychotic drugs for pregnant and
breastfeeding women, and women must be provided
with dependable information to support them in deci-
sions about medication during pregnancy and after the
Louise Howard senior lecturer
Health Services Research Department, Institute of Psychiatry, London
Roger Webb research fellow
Kathryn Abel senior lecturer
Centre for Women’s Mental Health Research, University of
Manchester, Manchester M13 9PL
Competing interests: None declared.
1McCormick A, Fleming D, Charlton J. Morbidity statistics from general prac-
tice: 4th national study 1991-1992. Series MB5, no 3. London: HMSO,
Howard LM,Kumar C,Thornicroft G.The psychosocial characteristics of
mothers with psychotic disorders. Br J Psychiatry 2001;178:427-32.
Howard LM, Leese M, Kumar C, Thornicroft G. The general fertility rate
in women with psychotic disorders. Am J Psychiatry 2002;159:991-7.
Smith SM, O’Keane V, Murray R. Sexual dysfunction in patients taking
conventional antipsychotic medication. Br J Psychiatry 2002;181:49-55.
Riordan D, Appleby L, Faragher B. Mother-infant interaction in postpar-
tum women with schizophrenia and affective disorders. Psychol Med
Webb R, Howard L, Abel K. Antipsychotic drugs for non-affective
psychosis during pregnancy and postpartum. Cochrane Database Syst Rev
Altshuler LL, Cohen L, Szuba MP, Burt VK, Gitlin M, Mintz J. Pharmaco-
logic management of psychiatric illness during pregnancy:dilemmas and
guidelines. Am J Psychiatry 1996;153:592-606.
Goldstein DJ, Corbin LA, Fung MC. Olanzapine-exposed pregnancies
and lactation: early experience. J Clin Psychopharmacol 2000;20:399-403.
University of Newcastle upon Tyne, NHS. National teratology information
service. (NTIS) www.nyrdtc.nhs.uk/Services/teratology/teratology.html
(accessed 29 Jul 2004).
10 Dev VJ, Krupp P. Adverse event profile and safety of clozapine. Rev Con-
temp Pharmacother 1995;6:197-208.
11 Yogev Y, Ben-Haroush A, Kaplan B. Maternal clozapine treatment and
decreased fetal heart rate variability. Int J Obs Gyn 2002;79:259-60.
12 Koren G,Cohn T,Chitayat D,Kapur B,Remington G,Reid DM,et al.Use
of atypical antipsychotics during pregnancy and the risk of neural tube
defects in infants. Am J Psychiatry 2002;159:136-7.
13 Gardiner SJ, Kristesnsen JH, Begg EJ, Hackett LP, Wilson DA, Ilett KF, et
al.Transfer of olanzapine into breast milk,calculation of infant drug dose,
and effect on breast-fed infants. Am J Psychiatry 2003;160:1428-31.
BMJ VOLUME 329 23 OCTOBER 2004bmj.com