Skip to content
View commentaries on this research

This is a plain English summary of an original research article

This NIHR funded study found that antipsychotic use during pregnancy was not associated with worse child outcomes after poorer health and riskier lifestyles were taken into account. For pregnant women these were things like other medications, obesity, smoking, alcohol and taking illicit drugs.

The study also provides further evidence against the use of valproate during pregnancy for epilepsy, bipolar disorder or schizophrenia because of the increased risk of poor outcomes for the child. The risk was double that of taking the alternatives, lamotrigine or carbamazapine during pregnancy.

Mental health problems can become worse during pregnancy or in the period after giving birth. The results of this well conducted study should help inform the difficult decisions of whether to continue, change or stop taking psychotropic drugs during pregnancy, though the reliability of the results is limited by the observational study designs.

Why was this study needed?

Mental health problems, such as bipolar disorder and schizophrenia, can become worse during and in the year after pregnancy. They may progress more quickly and be more serious than at other times.

No psychotropic medication is licensed for use in pregnancy and there is a lack of information about their advantages and disadvantages to either the mother or child. This is due to the absence of randomised controlled trials because of ethical reasons, or observational studies with adequate numbers. Therefore, it is difficult for women and their healthcare professionals to find evidence to make the best decisions on whether to stop these drugs, to keep taking them or to change to similar but safer drugs when they become pregnant.

Psychotropic drugs include antipsychotics and mood stabilisers. According to this large UK study, in the months before pregnancy, 0.21% of women were taking antipsychotics and 0.015% were taking the mood stabiliser lithium. The anticonvulsant types of mood stabiliser (valproate, lamotrigine and carbamazapine) were prescribed to 0.41% of women, mostly to control epilepsy (in 86% of these women).

What did this study do?

This was a multi-component study of 495,953 pregnancies in England between 1995 and 2012.

The authors used anonymous information from a database of GP records to look at the use of psychotropic drugs among women with psychosis, including bipolar disorder and schizophrenia, who became pregnant. They also included women with epilepsy in order to increase the number of women in their sample taking anticonvulsant mood stabilisers. They then used information recorded by the GPs to find out if the drug they were prescribed had any impact on the health of the mother and child following birth and up to 15 months after the child was born.

The study compared the benefits and harms of these types of drugs on the health of the mother and child both when the drugs were prescribed during pregnancy and when the drugs were discontinued during pregnancy. The large sample size lends strength to the study but the findings are limited by the necessity for it to be observational in nature.

What did it find?

  • Antipsychotic use during pregnancy did not increase the risk of caesarean section, prematurity or poor child outcomes after adjusting the results to take into account age, obesity, alcohol problems, smoking, illicit drug use and other psychotropic drug use.
  • Anticonvulsant use during pregnancy compared to no use increased the risk of:
    • Caesarean section from 18.3% to 21.4% (adjusted relative risk ratio [RRR] 1.14, 95% confidence interval [CI] 1.04 to 1.26).
    • Major congenital malformations from 2% to 4.1% (adjusted RRR 2.05, 95% CI 1.53 to 2.74).
    • Poor birth outcomes, including prematurity and low birth weight from 4.4% to 6.2% (adjusted RRR 1.33, 95% CI 1.06 to 1.67).
    • Neurodevelopmental or behavioural disorders from 4.8% to 8.7% (adjusted RRR 1.73, 95% CI 1.42 to 2.09).
  • Valproate use during pregnancy was associated with three times the risk of major congenital malformations compared to no anticonvulsant (RRR 3.15, 95% CI 1.98 to 5.13). The risk on valproate was double that for women on lamotrigine or carbamazepine (RRR 1.85, 95% CI 1.01 to 3.39). No comparison was calculated for lamotrigine or carbamazepine compared to no anticonvulsant. Lithium prescriptions were rare and reduced over the study period. This is in line with guideline recommendations.
  • Most women stopped taking psychotropic drugs before or in early pregnancy and
  • Hospital admission for a mental health issue more than tripled for women just after they gave birth compared to the period just before pregnancy (prevalence ratio 3.16, 95% CI 1.86 to 5.60).

What does current guidance say on this issue?

No psychotropic drug is licensed for use in pregnancy in England. NICE guidance from 2014 states that if a pregnant woman is stable on an antipsychotic and likely to relapse without medication; advice is to continue the antipsychotic. It also states that in regards to use of anticonvulsants for mental health problems, valproate should not be offered for acute or long-term treatment of a mental health problem in women of childbearing potential. Lithium and carbamazepine are also not recommended for women who are planning a pregnancy or are pregnant, except in the case of lithium, when there has been a poor response to antipsychotic drugs. For women taking lamotrigine, monitoring of levels of the drug are recommended during pregnancy and into the postnatal period.

What are the implications?

The risks associated with any antipsychotic use in pregnancy remain unclear. Results showed poorer child outcomes before adjustment for lifestyle factors that were more prevalent in women taking antipsychotics such as smoking, illicit drug use and alcohol problems. The outcomes for each antipsychotic medication were not analysed individually so it remains possible that adverse effects could have been found for single antipsychotic medications if more patients were studied.

It is important to identify and provide support to women to reduce any negative lifestyle factors such as obesity, smoking or illicit drug use. It is also essential to discuss the pros and cons of taking antipsychotic medication early on during pregnancy or preferably before pregnancy is contemplated.

This study provides further evidence against use of valproate for women during pregnancy due to the significantly poorer outcomes for the child compared to women who took other anticonvulsant drugs. This corresponds with NICE guidance on use of valproate.

In this study, the potential adverse effects of carbamazepine and lamotrigine were not considered separately, so there is no evidence to change the NICE conclusion that the risks associated with carbamazepine are higher than for lamotrigine.


Citation and Funding

Petersen I, McCrea RL, Sammon CJ, et al. Risks and benefits of psychotropic medication in pregnancy: cohort studies based on UK electronic primary care health records. Health Technol Assess. 2016;20(23):1-176.

This project was funded by the National Institute for Health Research Health Technology Assessment programme (project number 11/35/06).



NICE. Antenatal and postnatal mental health: clinical management and service guidance. CG192. London. National Institute of Health and Care Excellence; 2014.

NHS Choices. Mental health problems and pregnancy. London: Department of Health; updated 2015.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre

  • Share via:
  • Print article
Back to top