Schizophrenia affects approximately 1 % of the general population and is slightly more common in men. Fertility is lowered among women with schizophrenia (Bundy et al. 2011), but the move away from institutional care, and increased use of second-generation antipsychotic drugs, which have a lower propensity to elevate prolactin, may contribute to increasing rates of pregnancy in this group. Women who switch from older drugs may not be aware of this and may place themselves inadvertently at risk of unwanted pregnancy. Women with schizophrenia are more likely to have unplanned and unwanted pregnancies and less likely to engage well with routine antenatal care. They have more adverse outcomes of pregnancy including increased risk of pregnancy loss and neonatal death (Howard LM 2005)¹⁴. Some of this increased risk may be due to confounders such as high rates of smoking and poorer physical health.

There is an increased risk of relapse of schizophrenia in the postnatal period, but the pattern of relapse into illness does not show the very early high risk found in bipolar disorder (Munk-Olsen et al. 2009). This may be explained by the likelihood that precipitants to relapse relate more to the psychosocial and emotional demands of increased contact with health and social care professionals and caring for a developing child. Although not universally poor, the outcome in terms of the mother remaining the primary carer for her child is often unfavourable (Kumar et al. 1995) leading to great distress for the mother and for those (including health professionals) who support her. Appropriate supports, including social services, should be engaged at an early stage in pregnancy to ensure sufficient help is available to the mother and her family. It is often difficult for women with schizophrenia to cope with frequent contact with health professionals during pregnancy, and there is a risk that they receive suboptimal care due to failure to attend antenatal care.

Bipolar disorder has a lifetime prevalence of 0.4–1.6 %, with peak age of onset for women during reproductive years. Lifetime prevalence is likely to reach 4 % when less severe forms of the disorder are included. Genetic factors play an important role in aetiology, with heritability accounting for approximately 85 % of the variance (McGuffin et al. 2003), but episodes of illness may be triggered by stressful life events. Women with bipolar disorder are likely to be on maintenance therapy, which can include lithium, mood-stabilising anti-epileptics and second-generation antipsychotics. There are teratogenic risks associated with lithium and with the mood-stabilising anti-epileptics and sodium valproate and carbamazepine, but high risk too with regard to relapse of illness on discontinuation. Viguera et al. (2007) found that 71 % of bipolar women who discontinued prophylactic lithium treatment at onset of pregnancy relapsed at some point during that pregnancy, a twofold greater risk than for those continuing treatment.

Pre-existing bipolar disorder is one of the greatest risk factors for postpartum psychosis. Seventy per cent of women with bipolar disorder will experience relapse in the first 6 postnatal months if not taking mood-stabilising agents (Viguera et al. 2000). Irrespective of decisions about medication during pregnancy, all women should be offered prophylactic medication (usually lithium or a mood-stabilising antipsychotic) immediately following delivery. Given that most women with bipolar disorder return to full health between episodes, there is little evidence that they are any less able to care appropriately for their children, except during the acute phase of the illness.