Patients with psychotic disorders are often assumed to be sub-fertile, and when patients do have children, any child care concerns are usually passed on to social services. The children of patients with psychotic disorders may not, therefore, be of major concern to the responsible medical officer, and community mental health services have not been developed with families in mind. Routine statistics on psychiatric patient episodes do not include information on patients' children — a question covering this issue was included in the pilot draft of the Mental Health Minimum Data Set (approved for England by the Department of Health in September 1999), but was dropped because clinical professionals were reluctant to record this information, as they did not consider it relevant in this context (personal communication, G. Glover, 2000).

There are, however, several reasons for reconsidering whether general adult psychiatric services should know more about patients' children. There is a growing body of evidence to suggest that patients are often involved in the care of children (Reference Gopfert, Webster and SeemanGopfert et al, 1996), that this affects the lives of patients in many ways and that psychotic disorder in parents can also have an impact on their children. I would argue that psychiatric practice should include the collection of routine statistics about patients' children, the development of services which address our patients' family needs and future research into risk factors for poor parenting and the need for children to be placed in care.

Do patients with psychotic disorders have children?

There are few accurate reports of the incidence of pregnancy among patients with severe mental illness (Reference Apfel and HandelApfel & Handel, 1993). It has been traditionally assumed that patients with psychotic disorders have lower fertility rates than other individuals of child-bearing age. This has been thought to be due partly to biological aspects of the disorder itself, neuroleptic medication and also living in separate gender wards in the older hospitals.

With the advent of community care, atypical antipsychotics and changing attitudes more patients may be having children. Recent studies have demonstrated conflicting results on whether fertility is lower in patients with psychotic disorders compared with healthy controls (Reference Pancheri, Romit and MarconiPancheri et al, 1990; Reference JonssonJonsson, 1991; Reference Lane, Byrne and MulvanyLane et al, 1995; Reference Nimganonkar, Ward and AgardeNimganonkar et al, 1997) and although this debate is not resolved, several studies report that patients with psychiatric disorders are involved in caring for their children. For example, 9% of adult female patients with severe mental illness (mostly psychotic disorders) on a mental health database were the primary caregiver for a child (Reference White, Nicholson and FisherWhite et al, 1995). More recently, a study of women in contact with community mental health services found that 59% were mothers (Reference McGrath, Hearle and JennerMcGrath et al, 1999). Women with psychotic disorders are more likely than men with psychosis to be involved in child care. A study of 551 Mental Health Act assessments involving parents living with dependent children found that 72% involved women (Reference Hatfield, Webster and MohamadHatfield et al, 1997).

Impact on patients with a psychotic disorder of having children

It is not clear whether pregnancy makes psychotic disorders better or worse. Epidemiological studies have suggested there may be a slightly reduced prevalence of mental illness, as judged by contact with psychiatric services (Reference Kendell, Chalmers and PlatzKendell et al, 1987), but women with preexisting mental illness may avoid services for fear of losing their children. There is evidence from a prospective study that two-thirds of pregnant women with a history of non-organic psychosis were not in contact with psychiatric services during pregnancy, despite the presence of psychiatric symptoms (Reference McNeil, Kaij and Malmquist-LarssonMcNeil et al, 1984).

Concerns arise during pregnancy about possible adverse effects of prescribed and illicit drugs on the foetus (Reference Cohen, Heller and RosenbaunCohen et al, 1989), but reductions in medication to allay such fears may lead to psychotic relapse. Some women may not engage with antenatal services, although they may be at increased risk of obstetric complications (Reference Goodman and EmoryGoodman & Emory, 1992; Reference Sacker, Done and CrowSacker et al, 1996), which may put their genetically susceptible children at further risk of developing a psychotic disorder (Reference Sacker, Done and CrowSacker et al, 1996).

Post-partum, women with a history of bipolar disorder are at high risk of relapse (Reference Marks, Wieck and CheckleyMarks et al, 1992) and although women with schizophrenia are at less risk of a relapse in the early post-partum period (Reference Davies, McIvor and KumarDavies et al, 1995) the stress of child care may lead to relapse subsequently. There are few mother and baby units nationally (and proportionately fewer internationally) for women with post-partum disorders, despite the recognised importance of mother and baby bonding in the early post-partum period. Admission to hospital for parents of children over one year of age almost inevitably involves separation from the children for the duration of the admission, with few facilities for an appropriate environment for visiting, or help with child care on discharge. There is only one unit in the UK (in Drayton Park in North London) which can admit families when a woman is voluntarily admitted for an acute psychiatric problem (Reference Killaspy, Dalton and McNicholasKillaspy et al, 2000).

There is little research into the assessment of the safety of parenting in the context of chronic maternal psychiatric disorders (Reference Appleby and DickensAppleby & Dickens, 1993), but mothers with severe psychiatric disorders are often considered unable to care for their children. Fifty per cent of women with schizophrenia and 10% of women with affective disorder in a mother and baby unit were discharged without their child (Reference Kumar, Marks and PlatzKumar et al, 1995). Another study reported that 60% of children born to 80 female patients with chronic psychiatric disorders were reared by others, most commonly the child's father or an adoptive family (Reference Coverdale and AruffoCoverdale & Aruffo, 1989). It is not clear to what extent mothers with mental disorders retain responsibilities for their children (Reference Mowbray, Oyserman and ZemencukMowbray et al, 1995) and the incidence, in unbiased samples, of mothers with mental illness not caring for their children is not known. Although there is little known about how the loss of the parental role may influence the course of parental illness, anecdotally it is often seen as a difficult loss for the patient to accept. The positive role that being a parent could have for a patient and the factors that may protect families from the negative consequences of a parent with a psychotic disorder are not known.

Impact on children of having a parent with a psychotic disorder

Research into parents with mental illness has usually focused on the impact of the parental illness on the child and associations between parental mental illness and subsequent disturbances in the child are well documented (Reference RutterRutter, 1966; Reference Rutter and QuintonRutter & Quinton, 1984). Genetic transmission, as well as correlates of mental illness such as parental discord, aggressive behaviour or neglect, determine such difficulties. Children of mothers with schizophrenia may have more anxious attachment patterns (Reference Naslund, Persson-Blennow and McNeilNaslund et al, 1984) and those in unsupported single parent families may be particularly vulnerable (Reference WebsterWebster, 1992).

There is little research into the area of child abuse and children in the care system in families with a parent with mental illness. However, there is some evidence that children in care for more than 12 months in England are more likely to have a parent who has had psychiatric treatment, and that these parents are often the most socially disadvantaged (Reference Isaac, Minty and MorrisonIsaac et al, 1986). It is well known that Black patients are more likely to be detained under the Mental Health Act, but psychiatrists may be less aware that there is a disproportionate representation of Black (Black African, Black Caribbean and mixed parentage) children in the care system (Reference Bebbington and MilesBebbington & Miles, 1989; Reference BarnBarn, 1993). Black mothers are more likely to be referred for reasons of mental health by the police and the health service than are White mothers (Reference BarnBarn, 1993). It is therefore likely that future studies may confirm our preliminary findings that Black families with a parent with psychosis are more likely to have children placed in the care system (‘looked after’ children). If this increased risk is confirmed in the children of Black parents, research will need to tease out whether this is due to poorer social circumstances (and the reasons for this), cultural gaps between families and social services, discrimination at the level of housing departments or social services or other factors.

The increase in Black looked after children in the general population may be partly due to the finding that the police and schools are more likely to refer Black than White youngsters for reasons of delinquency (Reference BarnBarn, 1993). (Rates of self-referral to social services do not appear to differ when comparing White and Black families (Reference BarnBarn, 1993).) Socio-economic disadvantage and social workers' negativity have been found to be significant factors in the failure to implement preventative strategies with Black families (Reference BarnBarn, 1993), which may lead to quicker disposal of Black children into the care system. There is also evidence that African-Caribbean children have more rapid referrals into care than other ethnic groups (Reference Barn, Sinclair and FerdinandBarn et al, 1997). Owing to cultural differences, social services' perception of appropriate parental behaviours may differ from that of the parents, and language and communication difficulties may increase the gap between the two groups.

What should mental health services do?

The failure of services to respond to the family needs of parents with mental disorders may be due to ignorance or oversight by mental health professionals, stigma (Reference Blanch, Nicholson and PurcellBlanch et al, 1994) or an unspoken denial of sexual activity and child-bearing (Reference Apfel and HandelApfel & Handel, 1993). Several authors in the past have pointed to the lack of attention to helping mothers with mental illness in managing daily parenting demands (Reference Test and BerlinTest & Berlin, 1981; Reference Bachrach and NadelsonBachrach & Nadelson, 1993). Integrated services are not provided partly because parenting is not generally considered to be a mental health issue unless child protection concerns are raised (Reference Blanch, Nicholson and PurcellBlanch et al, 1994).

Mental health service recipients who are parents should be identified as such. Until this is done we will not know how many of our patients are caring or have cared for their children and what impact this has on their lives, their illnesses and their children. Routine data such as the Mental Health Minimum Data Set (which is due to be operational by 2003, according to the recent National Service Framework (Reference ThornicroftThornicroft, 2000)) could include basic information on the children patients have had, how many are currently living with them and who has parental responsibility for them. Mental health services could include more child care facilities, flexible appointment systems and advice on family planning and parenting. Management plans should include details of what child care arrangements have been made if a patient relapses. More provision for children visiting their parents in psychiatric units could be made, and units could be set up to which parents can be admitted with their children if this is in the best interest of the family.

Future research needs to examine how many of our patients are parents and the impact this has had on their lives, and also the impact of their illness on the lives of their children. Reliable and valid measures of ‘good enough’ parenting should be developed to assess whether children should be placed in care, and service interventions need to be developed and evaluated so that families with a parent with a psychotic disorder can be supported and helped, for the benefit of both patients and children.