Arch Womens Ment Health (2002) 5: 49–58 DOI 10.1007/s00737-002-0134-6
Original contribution Care in a mother-baby psychiatric unit: analysis of separation at discharge F. Poinso1, M. P. Gay1, N. M. C. Glangeaud-Freudenthal2, and M. Rufo1 1 2
Psychiatric Parent-child Unit, Sainte-Marguerite Hospital, CHU Marseille, France INSERM U149, Villejuif, France
Accepted February 23, 2002; Published online September 17, 2002 © Springer-Verlag 2002
Summary Joint psychiatric admission to a Mother-Baby Unit (MBU) enables a mother to obtain care for psychiatric disorders and simultaneously receive support in developing her identity as a mother. This care is meant to prevent attachment disorders and mother-baby separation. Outcome at discharge, however, may differ according to the mother’s admission diagnosis. Demographic data, clinical features of parent and child, and clinical outcome of 92 consecutive admissions of mothers and their children to a MBU in Marseille were collected over a period of eight years (1991–1998). Separations occurred in 23% of the joint admissions. Women with acute postpartum psychoses and major depressive disorders had better outcomes than those with chronic psychoses: at discharge, the latter were more often separated from their children. In those cases, however, MBU admission provided time to arrange the best placement for the child. Outcome was less predictable for non-psychotic personality disorders and depended not only on the mother’s disease but also on her family and social context. Keywords: Mother-baby postnatal mental disorders.
unit;
Mother-baby
separation;
Introduction The psychiatric hospitalisation of a mother with her infant requires a very specialised treatment unit (Guedeney et al., 1986; Cazas et al., 1990; Buist et al., 1990; Guscott and Steiner, 1991; Kumar et al., 1995; Bell et al., 1994; Bell et al., 1995; Brockington, 1996; Durand, 1997; Poinso and Rufo, 2000). In France, these units are most often managed by child psychiatry departments, even when the mother is mentally ill (Sutter and Bourgeois, 1998). We outline here the operation and goals of such a unit. We also analyse, based on 92 consecutive mothers and their
infants admitted into this unit, the situations that led to separations. Broadly, mother-baby hospitalisation can be indicated in one of four situations: – The first involves acute disorders, apparently psychotic, of the puerperium (acute postpartum psychosis). Postpartum psychoses occur in one or two deliveries per thousand (Cox, 1998; Kendell et al., 1987; Brockington et al., 1982; Pritchard and Harris, 1996). They are often mood disorders (confusional depression, atypical manic states) that fall within the framework of affective psychosis (Kumar et al., 1995). It is thus necessary to treat the mother’s acute episode and, at the same time, prevent any unfavourable consequences of an early and prolonged separation of mother and child, in particular, severe attachment disorders (Bowlby, 1977). – The second concerns various disorders: anxious, neurotic, and depressive reactions that can affect the mother-child relationship (transactional spiral) (Stern, 1981). This category includes depressive syndromes, sometimes very serious (major depressive disorders). Depression without psychotic symptoms often develops more insidiously and later, more than a month after delivery. This postnatal depression affects 10–15% of mothers (Cox et al., 1993; O’Hara and Swain, 1996; GlangeaudFreudenthal, 1999). Although its repercussions on the infant’s cognitive development are substantial (Murray, 1992), it seldom requires hospitalisation.
50
A young mother’s difficulty in expressing her need for treatment is particular to this period of life (Lussier et al., 1995; Poinso et al., 1996). Thus, hospitalisation of mother and baby in this situation is an extreme measure that is considered only when multiple problems are combined: emotional isolation, problems in relationships or at work, neurotic anxiety revived by real events etc. – In the third category, an early mother-child bond may be in jeopardy due to serious medical, social, and psychiatric situations. This involves cases in which the mother suffers from schizophrenia or from serious non-psychotic personality disorders. – The fourth category involves children with functional symptoms (sleeping or eating disorders) or developmental delays (psychomotor retardation) whose families benefit from a mother-child hospitalisation setting, for short-term inpatient treatment. In the latter category, the parents normally are not mentally ill; it is the child who has a disorder. Our aim is to describe care in the Marseille motherbaby unit and to compare the women in our series who were separated from their babies to those who were not.
Sample and methods Sample All joint admissions of children with their mothers to the mother-baby unit at the Marseille (France) University Hospital Centre were collected over a period of eight years (1991– 1998). The unit’s capacity during the week, day and night, is four “families”. A limited number of mother-baby pairs can also be treated during “day” or half-day hospitalisation, from Monday through Friday. The father may be hospitalised, when he is present and willing (Poinso and Soulayrol, 1998). During the study period, 100 children (including some siblings, and one pair of twins) and 92 mothers were admitted to our MBU. Seven mothers were hospitalised with a baby and an older child and one with a pair of twins. Among these hospitalised sibling groups, only the twins were separated from their mother. We were able to meet 69 fathers during this period. Twenty-one fathers asked or agreed to stay in the MBU with the mother and baby. Our information about the population is not exhaustive because of the isolation of some mothers and their reluctance to provide us with information. The difficulties were even greater for fathers. Mothers’ characteristics (Table 1): The mean age of the mothers was 30 years (range: 17–47 years). More than a third of the mothers (39%) had a history of psychiatric hospitalisation. Very few had already had a child placed (4%). Fourteen mothers (18%) were living in a mother-baby hostel at the time of admission. Only one third (35%) were employed. More than a third of the women were neither married nor living with a partner at their admission to the unit.
F. Poinso et al. Table 1. Mothers’ characteristics and history (N ⫽ 92)
N
(%)
Employed Education level (primary school) Born in France Marital status (single mother) Parity (primiparous) Own housing History of previous psychopathology History of obstetric complications No antenatal medical care Drug detoxification of infant at birth Other children placed in foster care
29 28 70 33 56 64 36 12 3 3 4
(35) (33) (85) (39) (67) (82) (39) (14) (3) (3) (4)
Table 2. Children’s age distribution and disease on admission Age distribution
N ⫽ 100
⬍6 months 6–12 months 1–2 years ⬎2 years
48 9 14 29
Disease distribution
N ⫽ 100
No psychiatric disease Eating or sleeping disorders Learning or motor skills disorders Reactional disorders Personality disorders Autistic disorder Behaviour disorder Mental retardation Other psychiatric disease
52 14 13 7 8 1 2 1 2
Fathers’ characteristics: Sixty-nine fathers (75%) were seen at our MBU at least once. The mean age for fathers was 33 years (range: 21–55). Fifty-eight fathers (84%) were employed, 18 had at least a primary school education (28%) and 53 were born in France (82%). Eleven (12%) of the fathers we saw had the following psychiatric problems, based on their history of psychiatric hospitalization: schizophrenia (N ⫽ 1), major depression (N ⫽ 2), and personality disorders (N ⫽ 8). We had no information on this subject for 33 fathers. The other 23 fathers (25%) never came to our MBU for a variety of reasons: father unknown (N ⫽ 8), dead (N ⫽ 1), or did not want or could not participate in MBU care (n ⫽ 14). Children’s characteristics (Table 2): Of the 48 children younger than 6 months old, half were younger than 1 month (n ⫽ 25), a quarter were between 1 and 3 months (n ⫽ 11) and a quarter were between 3 and 6 months (n ⫽ 12). Twenty-nine children were older than 2 years at admission, they were admitted for eating or sleeping disorders, learning disorders or reactional disorders, except one four year old child who was admitted with her mother, during the latter’s schizophrenic episode. The mean age for children at admission was 18.5 months. The sex distribution was 52 boys and 48 girls. More than half of all the children presented no symptoms.
Mother-baby psychiatric unit: separation at discharge
Measures and diagnostic instruments Mothers (and fathers, when they were seen at MBU) were independently diagnosed according to ICD-10 by two psychiatrists at the Marseille MBU. When mothers or fathers were already patients in a psychiatric department, diagnoses were made after consensus among the psychiatrists. Demographic data, clinical features of parent and child, and clinical outcome were collected. The children’s diagnoses were based on developmental assessments performed by two psychiatrists and a psychologist in our MBU and on the child nurses’ observations of behaviour (including eating and sleeping disorders) and development. These were collected and discussed weekly until consensus was reached.
Results Description of care in the Marseille mother-baby unit The goal of this unit is to care for a mother with her baby, to encourage the development of a good relationship between mother and child, and to treat the mother’s postpartum illness. To the extent that we do not offer hospitalisation over the weekend, separation from the unit is planned in the initial contract. This requires that a family network be mobilised. Usually, in acute postpartum pathology, the father and the grandparents take over for the weekend. Chronic mental illness is often accompanied by the deterioration of family bonds, and in these cases weekend arrangements are more difficult to organise. When it is clear that the parents cannot care for their child and there is no family network, a Children’s Court judge orders the child’s separation from the mother during weekends, usually at the beginning of the hospitalisation. Generally, the need to rely on family and social networks available to the baby and mother becomes progressively more evident during hospitalisation. The initial isolation of the mother-baby dyad is often misleading. The mother’s circle, whose members often have their own problems, requires time to come forward in an adequate and supportive manner. A hasty decision by the team of what a mother and her baby require prevents progressive solutions and flexible childcare arrangements that protect the child’s development (help from the grandmother, for example). The caregivers are willing to work with whoever comes: father, grandparents, uncles, and aunts. They establish a multi-parent situation during a transition period, leading to effective long-term family support. The work of this unit thus consists in helping the parent develop an identity, through real experiences,
51
as close as possible to everyday life. This could prevent the discrepancy noted by David (1987) and Lamour (1989), between the discourse of a schizophrenic mother and her psychiatrist in comparison to a mother’s attitudes as observed by caregivers closer to the children. This situation leads to rifts between the caregivers and makes it difficult to promote either the bond or the separation. Continuity, on the other hand, leads some mothers to accept their problems more realistically and to search for appropriate solutions with the team. Our intervention is based on the continuous presence of a nurse for mother and child. During the hospitalisation, the baby is never alone, and the mother rarely, only if she wishes to be. The first principle of support preserves the parents’ role in the child’s daily care. The mother, sometimes the father, prepares and gives the baby a bottle, bathes and changes the child, carries, rocks, and puts the infant to bed. The child-care giver adopts a reserved attitude but is available and remains present at times perceived as anxiety provoking (bath time or meals, for example). They must nonetheless accept the mother’s request to take her place, for a given moment or more regularly. The limits of our ‘neutrality’ (non-intervention) depend essentially on two factors: a) the protection of the child, if the mother appears inadequate, violent, or insufficiently reassuring b) the mother’s request for intervention. Our goal is for the care-giver to provide support in the development of a maternal identity, exemplifying and serving as a role model for attitudes, behaviours, as well as methods that the mother can imitate. For example, a mother may use the same tone as the caregivers in speaking to the baby, or hold the baby for its bath in a similar manner. The relationship between the mother and the caregivers assumes that parental images are projected. Consequently, the caregivers’ feedback must be carefully worked out, to avoid disqualification and to create a positive identification with the caregiver in her maternal function. Discussion of the caregivers’ interventions and their observations is essential during weekly team meetings. A MBU has a privileged view of early mother-child interactions. Our direct observation of babies is based on the principles of the method outlined by Bick (Bick, 1964; Houzel, 1994). This type of observation is intended to contain the emotions and the anxious tension between a mother and a baby. In addition, we use an observation scale,
52
similar to the Bethlem Mother-Infant Interaction Scale (Kumar and Hipwell, 1996), developed and standardised by the Marcé Society (Durand et al., 1999). This scale will be used in all MBUs in France and Belgium participating in this research group. The therapeutic setting is founded on the consistency of the team, a consistency that does not prevent some degree of differentiation. Within this unit, different types of psychotherapeutic approaches can be offered, as a function of the mother’s characteristics. Mother-baby psychotherapy in the strict sense of the term (Cramer, 1989; Lebovici and Weil-Halpern, 1989) is rarely possible. Even when an analysis of transference is not an integral part of the treatment, mother-baby psychotherapy requires that the parents link past and present events and reintegrate the parts that they have projected onto the baby. In our experience, this type of psychotherapy is applicable to only a minority of women with neurotic rather than psychotic problems. More often we treat parental psychopathology as well as their lack of insight into their condition. Psychotherapy for these cases must thus avoid interpretation and instead should be based upon daily experience and practice. This makes it possible for the woman, supported by and observing the caregivers, to develop a self-representation as a parent, to recognise the child’s current needs, and to anticipate the child’s development. For seriously ill mothers, the therapist must sometimes speak to the child (if and when age appropriate) to translate the maternal affect, which is marked by at times confusion between the mother’s needs and those of her child, or by fluctuations in her investment in the child with alternating approaches and withdrawals. In some cases, the mother wants to see the psychologist or psychiatrist alone, for herself: “It’s me who has problems, not my child”. Fundamentally, the therapeutic approach, even in the child’s absence, focuses on parenthood and bonding with the child (Brockington et al., 2001). Drug treatment is often necessary. If the mother has a psychiatrist, we prefer that this physician manages the prescription and the surveillance of the medication. When we meet the mother before any other psychiatrist has, we provide her medication during the acute period. Our intervention is on the whole based on flexibility and offers a variety of treatments (hospitalisation during the week or for briefer periods, as well as home visits), within the
F. Poinso et al.
limits imposed by the need to protect the child. Treatment is discussed with the family, in partnership with medical-social agencies. The length of admission to our MBU ranged from one week to more than a year. The mean duration of admission for full time hospitalisation (in-patient except for weekends) is 8 weeks. Short-term admissions are usually due to the mother’s refusal to continue with the joint admission or for practical reasons, such as relocation, and sometimes following an early transition to outpatient care. In-patient hospitalisation may be followed by a part-time hospitalisation as outpatient care. Overall nearly 50% of mother-baby pairs were cared for in our unit, as either in- or out-patients, for 3 to 6 months. MBU relation with social services and justice Both admission and discharge usually require coordination with a medical-social agency, DGAISS (Direction Générale des Affaires et des Interventions Sanitaires et Sociales) and with the Children’s Court judge. DGAISS and court social worker interventions continue throughout the hospitalisation. In France, each administrative region (95) has a DGAISS that administers child placement (in foster families or institutions). Nonetheless, the Children’s Court judge decides, after an investigation by a court social worker, whether children remain in the family or are placed in foster care. In France, medical ethics clearly separates the mission of an expert providing a judgement from the mission of a clinician providing treatment. When an assessment is needed, it is performed jointly with the other medical and social agencies involved with the family. It never, however, takes the form of an expert evaluation submitted to the judge. We inform the DGAISS and the judge of our proposal but treatment remains our exclusive focus (Chardeau, 1999; Poinso and Rufo, 2000). The framework, the ethic of our care, and its setting are based on the explicit agreement with our patients regarding the type of care we provide. Therefore, treatment is first discussed with the patient and her family. Parents who reject or abuse their children usually disagree with the basic framework of our unit and are not referred to us. A judge cannot commit patients to our unit against their will. That may be why we have few cases in which physical child abuse is a real risk. Neglect or the pathological interaction of a mentally ill mother or parents, on the other hand, pose a clear
53
Mother-baby psychiatric unit: separation at discharge
risk to the children. We have an ethical responsibility to report danger to the judge, who may remove the child from the mother or parents. This situation creates complex ethical dilemmas, as Ramsay and Kumar (1996) described. Analysis of separation at discharge Overall, during the study period, 71 babies and their siblings were discharged from the MBU with their mothers and 22 children, including one pair of twins, were removed from their mothers. Type of separation Because we do not offer weekend care, we must ensure a safe and appropriate setting for a mother and her child over these two days. In some cases, the mother and her child go to the family of one parent or another, but in other cases, they spend the weekend apart, with the child in a foster home, pediatric ward or institution and the mother in a standard psychiatric ward. These weekend arrangements are only temporary. Children removed from their mothers at discharge fall into two groups. The first consists of children placed in the unofficial custody of another family member (2 boys and 5 girls). In this group, the family temporarily organised the child’s care, most often with the approval of the DGAISS. Usually, however, the Children’s Court judge subsequently confirmed this placement in a court order. These children were placed with a family member before the age of six months. Five of these mothers were schizophrenic, and two had severe borderline personality disorders.
The other group consists of children removed by judicial order and then placed in foster care (7 boys and 8 girls, including the twins). The separation decision sometimes occurred during hospitalisation in our unit. In this situation, the cases had always been reported to the DGAISS before admission to our unit (during the maternity ward admission, for example). The request for hospitalisation came from social and judicial agencies that wanted to ensure that a separation was as atraumatic as possible for mother and child. In these cases, we require that the mother requests time to develop a bond with the child before admission. Some of these separations also occurred in the one to six months after discharge from the unit, most often when the family had refused outpatient follow-up (consultations or home visits). In both groups (negotiated separations with custody to another member of the family and separations by judicial order into non-family settings), the initial placement measures were always temporary (e.g., an initial court placement is always for less than 6 months), but, in our experience, all have become permanent. When we compared these two groups, we observed that the mother’s illness was distributed similarly in both groups but that family placements were set up more rapidly than foster home placements. This may be due to the greater complexity of the legal procedures required in the second case. Separation according to mother’s illness at admission (Table 3a) Table 3a reports the overall distribution of maternal illness at admission (92 mothers). No statistically
Table 3. Separation on discharge according to type of mothers’ psychiatric disease on admission Psychiatric diagnosis (ICD-10 diagnosis)
3a: On admission Mothers’ psychiatric diagnosis Entire sample N (% of total sample)
3b: On discharge Type of separation from the mother All separations N (% of each disease)
Acute psychosis Schizophrenia Mood disorders Neurotic symptoms Personality disorders No psychiatric disease
11 (12) 16 (17) 16 (17) 8 (9) 24 (26) 17 (19)
2 (18) 12 (75) 1 (6) 0 (0) 6 (25) 0 (0)
Total
92 (100)
21 (23)
* MDP
Placed in foster care by a judge N 2 7 1* 0 4 0 14
Placed within the family N 0 5 0 0 2 0 7
54
significant correlation was observed between the parental disease and the child’s symptoms at admission. As expected, the very young children hospitalised with psychotic mothers seemed to have fewer symptoms than the older children admitted to the unit for reactive disorders (eating or sleeping disorders, etc.) with healthy parents who were seeking help. These children were usually older than children hospitalised with psychotic mothers. Only two of eleven mothers hospitalised for an episode of acute postpartum psychosis were separated from their children. At discharge, both presented residual schizophrenic-type disorders after the acute episode and had mothering difficulties similar to those of schizophrenic mothers. For both women, the acute postpartum psychotic episode was the entryway into schizophrenia. Most mothers with schizophrenia (12 of 16) were not able to live with their children. The 4 mothers with schizophrenia able to raise their child had stable spousal support from a partner, who often had a rigid personality structure, but showed good social adaptation. Separation according to family status and fathers’ illness Of the 15 children placed in foster care by a judge, only 3 had a father who came for at least one appointment to our MBU. One father had schizophrenia; another, not mentally ill, was deported from France and imprisoned in his own country; and the third, also not mentally ill, separated from the baby’s mother during her relapse into bipolar depression. Of the 7 children separated from their mother and placed with other family members (without initial judicial intervention), we knew 3 fathers: two were not mentally ill, one took custody of the child, and the third had a borderline personality disorder. Separation according to child’s age (Table 4) Sixty-eight percent of the children separated from their mothers were younger than 6 months old. Children who were separated from their mothers were younger (mean age 8.9 months) than the overall group at admission (mean age 18.5 months). All the children placed within the family were younger than 6 months old.
F. Poinso et al. Table 4. Separation on discharge according to child’s age Age group
Separation from mother on discharge Placed in foster care by a judge N
⬍6 months 6–12 months 1–2 years ⬎2 years Total
8* 4 2 1 15
Placed within the family N
All separations N
7 0 0 0
15 4 2 1
7
22
* Including one pair of twins No other child hospitalised with sibling was separated from his mother.
Discussion The hospitalisation of young children with their parents is difficult because the therapeutic space is threatened by the anxiety of team members and also by those outside the unit who initially requested our intervention. The safety of the child should always be paramount even when separation may thwart either the establishment or the maintenance of the motherchild bond. Nonetheless there is no strict correlation between a psychiatric disease and the possibility of caring for and bonding with a baby. Immediate danger to the child is, in our experience, very rare, but that may be because only parents who explicitly agree to a care project that meets the unit’s philosophy are admitted. As we argue below, the mother’s isolation seems to be the most significant indicator of danger and poor prognosis for the mother-child relationship in cases of maternal mental illness. Separation did not occur only in cases where the mother was psychotic. In our series, 6 of 21 mothers separated from their children had non-psychotic personality disorders (borderline state). Maternal psychosis, however, requires a particular type of mother-child separation. The decision is made preventively, on the basis of future deficiencies extrapolated from the mother’s behaviour or discourse, or the presence of current symptoms indicative of a psychotic object relation. The situation is different for mothers with personality disorders: separations are ordered after the observation of serious relational difficulties with the child, and sometimes after violence has occurred. Children whose mothers are mentally ill represent a group at high risk of emotional disorders both during childhood and in adult life. According to Anthony (1969), at the age of 10 years, 15% of the
Mother-baby psychiatric unit: separation at discharge
children of psychotic mothers have global developmental disorders (manifestations of a schizophrenic type), 45% have serious behaviour disorders, 10% are considered vulnerable but can also be intelligent and creative, and 30% have no psychiatric disorders. Studies among adults have essentially considered the incidence of parental schizophrenia. Different studies have consistently established that there is a 10–15% risk for the child of a schizophrenic mother to develop schizophrenia in adulthood (Mednick and Schulsinger 1965; Schulsinger, 1976) and a 20– 30% risk for the child of schizophrenic parents (Rosenthal, 1971; Kringlen and Cramer, 1989). Adoption studies (Heston, 1966; Rosenthal et al., 1971; Higgins, 1966, 1976), however, have shown that the early adoption of these children, from the second week after birth, does not reduce their risk of schizophrenia in adulthood. Moreover, some authors have reported non-schizophrenic psychological disorders in separated or adopted children (more suicides, for example, among children separated from their mothers than among those who were not); this finding indicates that separation induces an additional handicap. Higgins et al. (1997) recently confirmed these findings, although their study contained substantial methodological problems (for example, recruitment bias). Analysis of the problems that the schizophrenic mother has in mothering her child requires a different kind of observation. Duchesne and Roy (1991), reviewing the criteria presented by Bowlby (1977) and by Lebovici (1985), proposed a classification of the risks to the child according to specific traits of the mother’s disease, correlated with various particularities of mothering. For example, mothering by a dissociated and delusional mother is distorted and encloses the child in a relational bubble. In the inadequate care caused by poor continuity (observed among those with intermittent psychotic relapse), the principal risk for the child lies in the non-acquisition of self-consciousness. The inadequate care provided by hebephrenic or catatonic mothers can lead to immediate physical risks for the child and also to overall developmental retardation through a lack of stimulation. For Snellen et al. (1999), the adverse contribution of negative symptoms for disturbed interactions was often not evident until after the positive symptoms had resolved, but the negative symptoms are often treatment resistant.
55
When the mother has a serious non-psychotic personality disorder, the relationship between infant and mother has fewer profound distortions, but the mother’s investment in the child is questionable: there may be an alternation of movement toward and away from the child. The child is also the target of projections and held responsible for the parents’ episodes of anger or anxiety. Violence may emerge, repeating a mother’s early relationships with her parents. The separation between parents and child, contrary to the situation for schizophrenic mothers, is imposed by the onset of serious events that cause physical danger to the child. In our hospital, as in that described by Stewart (1989), women suffering from personality disorder or substance abuse are more likely to refuse joint admission to the unit than women suffering from a psychotic illness. The impact of acute postpartum psychoses on the mother-child relationship and the child’s development has been studied much less because these disorders are rare: 1–2 per 1,000 births. The series published over the past 15 years (Brockington et al., 1988; Klompenhouwer and van Hulst, 1991; Kumar et al., 1995; Mowry and Lennon, 1998) have modified professionals’ viewpoints about these episodes. The ranking of the frequency of diagnoses for episodes labelled “postpartum psychosis” has changed, just as it has for acute psychoses during adolescence: the most frequent diagnosis now seems to be affective psychosis and not schizophrenic disorder. Joint hospitalisation of the mother and baby allows treatment simultaneously for the mother and for the mother-child relationship while avoiding the early and prolonged separation of mother and child to which these acute situations inevitably lead when this specific type of hospitalisation is not available. Whether the disorder is acute or more chronic, joint mother-baby hospitalisation leads to a new type of situation and to treatment conditions that provide an alternative to immediate separation. The experience of joint mother-baby hospitalisation also demonstrates babies’ capacity to benefit from such a setting. The prognosis for the mother’s relationship with her child seems to be strongly correlated with her capacity to use various treatments: the MBU, the family, Protection Maternelle et Infantile (PMI), Direction Générale des Affaires et des Interventions Sanitaires et Sociales (DGAISS), and Aide Educative en Milieu Ouvert (AEMO, judicial measures). Helping her thus requires us to consider
56
simultaneously: (a) the diagnosis of the mother’s illness; (b) the short-term course of the motherhood process; (c) the father and his potential role (Poinso and Soulayrol, 1998); (d) the stability of the couple’s relationship (or lack thereof); (e) support from family, friends or a broader social circle. Our treatment mission is based on a progressive improvement in the mother-child relationship, an improvement that requires time. The length of hospitalisation is thus longer than in units whose goal is assessment only (Milgrom et al., 1998). Our experience suggests that early separation is not only prejudicial to the quality of maternal attachment but is also a source of suffering for the baby, even before the second half of the first year (contrary to theories of maternal object representation). A newborn establishes ways of interacting with its mother; the baby knows her voice, recognises her smell, and adjusts to how it is being carried. After 6 months, an infant expresses anxiety or refusal of separation more directly, but this does not mean that an earlier separation is inconsequential. Motherbaby units, in approaching the need for a separation prudently, allow it to occur progressively and have an important role in these difficult situations. Finally our study confirms what Riordan et al. (1999) established – that the relational prognoses of schizophrenia and acute psychoses are very different. Acute postpartum psychoses are often related to affective disorders and can indeed be their initial manifestation in young women (Klompenhouwer and van Hulst, 1991; Kumar et al., 1995). The general prognosis, the possibility of effective treatment in this acute period, and the possibility of drug treatment are essential reasons for preserving the bond between the mother and child. Moreover, these acute episodes occur in a very different family setting compared to women with schizophrenia. The family is present, expresses appropriate concern, participates in the treatment, enables home visits over the weekend, and takes over some of the baby’s care. The father is most often present and active, even if there is a risk of onset of psychiatric illness in the postpartum period for him also (Harvey and McGrath, 1988; Lovestone and Kumar, 1993). Women with schizophrenia experience much greater isolation, and the baby’s father is often mentally ill, absent, or unknown. The mother’s bonds with her family, and in particular with her mother, are usually quite frayed. In our study, schizophrenics who were not separated from their child had
F. Poinso et al.
established a sufficiently stable conjugal relationship and were able to take care of their child because the child’s father was also present at home. This was the case for 4 of 16 women. These results are similar to those reported by Hipwell and Kumar (1996). The women who did not have stable support from their partner were separated from their children. Conclusions Units providing for joint hospitalisation of mothers and babies allow time to assess the problems and promote the possibility of a relationship between babies and mothers who suffer from psychiatric problems after delivery. The indications for admission to such a unit are diverse and relate to different situations. When the mother’s illness is an acute postpartum disorder (major depression, depressive or manic state, confusional disorders), the prognosis is generally good, especially if the conditions of hospitalisation permit continuity in her relationship with her child. This allows psychiatric treatment for the illness and supportive care and attention during the process of constructing a maternal identity. For schizophrenic disorders, the prognosis is worse because the disease’s development is not fundamentally modified by the birth of a child (Yarden et al., 1966), and separation is almost inevitable in the absence of a stable relationship with a partner who is the child’s father, as our series shows. If the mother cannot care for her child alone, a period of hospitalisation is the best possible compromise, often with the help of the mother’s family, for the child’s subsequent custody and the maintenance of a bond with the mother. In non-psychotic personality disorders, separation sometimes occurs but the family environment, the seriousness of the maternal illness and the circumstances are too diverse to allow any general conclusions to be deduced. These data lead us to recognise a variety of goals for hospitalisation in a MBU, depending on the mother’s diagnosis. Notably, a MBU can treat an acute postpartum psychiatric illness while developing the relationship between the mother and her child on a daily basis. The treatment proposed for mothers with schizophrenia benefits the relationship between mother and child. This work has enabled us to describe clinical elements while taking into account these babies’ specific needs, in these particular
Mother-baby psychiatric unit: separation at discharge
circumstances. But other measures are available in France (early separation with meetings between mother and child – organised by the DGAISS medical-social agencies). Mother-baby hospitalisation is especially useful when the mother expresses a desire to care for her baby and this does not compromise the child’s safety. A (sometimes long) period of reflection on the conditions of the relationship or of the separation, may then be necessary.
References Anthony EJ (1969) A clinical evaluation of children with psychotic parents. Am J Psychiatry 126: 177–184. Bell AJ, Land NM, Milne S, Hassanyeh F (1994) Long-term outcome of post-partum psychiatric illness requiring admission. J Affect Disord 31: 67–70. Bell AJ, Land NM, Milne S, Hassanyeh F (1995) Postpartum depression – A specific concept? Br J Psychiatry 166: 826–827. Bick E (1964) Notes on infant observation in psychoanalytic training. Int J Psychoanal 558–566. Bowlby J (1977) The making and the breaking of affectionnal bonds. I. Aetiology and psychopathology in the light of attachment theory. An expanded version of the Fiftieth Maudsley Lecture, delivered before the Royal College of Psychiatrists, 19 November 1976. Br J Psychiatry 130: 201–210. Brockington IF (1996) Motherhood and mental health. Oxford University Press, Oxford, pp 552–583. Brockington IF, Winokur G, Dean C (1982) Puerperal psychosis. In: Brockington IF, Kumar R (eds) Motherhood and mental illness. Academic Press, London, pp 37–69. Brockington IF, Margison FR, Schofield E, Knight RJ (1988) The clinical picture of the depressed form of puerperal psychosis. J Affect Disord 15: 29–37. Brockington IF, Oates J, George S, Turner D, Vostanis P, Sullivan M, Loh C, Murdoch C (2001) A screening questionnaire for mother-infant disorders. Arch Womens Ment Health 3: 133– 140. Buist AE, Dennerstein L, Burrows GD (1990) Review of a mother-baby unit in a psychiatric hospital. Aust N Z J Psychiatry 24: 103–108. Cazas O, Dhôte A, Bouttier D, Ginestet D (1990) L’hospitalisation de la mère et de son nourrisson dans un service de psychiatrie adulte. Psychiat Enfant 23: 635–674. Chardeau P (1999) Parentalité difficile, parentalité impossible? Réflexions éthiques sur le traitement des troubles de la relation précoce en unité d’hospitalisation mère-bébé. Neuropsychiatr Enfance Adolesc 47: 502–506. Cox JL (1998) Troubles psychiatriques du post-partum: aspects socio-culturels de la “maladie mentale sévère”. In: Psychiatrie périnatale. PUF, Paris, pp 431–440. Cox JL, Murray D, Chapman G (1993) A controlled study of onset, duration and prevalence of postnatal depression. Br J Psychiatry 163: 27–31. Cramer B (1989) Les thérapies spécifiques mère-bébé et la consultation thérapeutique. In: Lebovici S (ed) Psychopathologie du bébé. PUF, Paris, pp 827–849. David M (1987) Souffrance du jeune enfant exposé à un état psychotique maternel. Perspect Psychiatr (Paris) 26: 7–22. Duchesne N, Roy J (1991) Enfants de mère psychotique: risques développementaux et interactions précoces, revue de la littérature. Neuropsychiat Enfance Adolesc 39: 291–299.
57 Durand B (1997) L’hospitalisation conjointe mère-enfant. In: Guedeney A, Lebovici S (eds) Interventions psychothérapeutiques parents-jeunes enfants. Masson, Paris, pp 163–176. Durand B, Yehezkieli G, Glangeaud-Freudenthal NM, Clerc C, Amzallag C, The Mother-Baby Units Discussion Group (1999) Instrument for weekly nursing assessment of mother-baby interaction. Arch Womens Ment Health 2: 127. Glangeaud-Freudenthal MC (1999) Estimation de la prévalence de la dépression post-partum en France. Devenir 11: 53–64. Guedeney N, Mouren MC, Dugas M (1986) A propos des hospitalisations mère-enfant: l’hospitalisation mère-enfant dans un service de psychopathologie de l’enfant. Psychiatr Enfant 29: 469–531. Guscott RG, Steiner M (1991) A multidisciplinary treatment approach to postpartum psychoses. Can J Psychiatry 36: 551– 556. Harvey I, McGrath G (1988) Psychiatric morbidity in spouses of women admitted to a mother and baby unit. Br J Psychiatry 152: 506–510. Heston LL (1966) Psychiatric disorders in foster home reared children of schizophrenic mothers. Br J Psychiatry 112: 819– 825. Higgins J (1966) Effects of child rearing by schizophrenic mothers. J Psychiatr Res 4: 153–167. Higgins J (1976) Effects of child rearing by schizophrenic mothers: a follow-up. J Psychiatr Res 13: 1–9. Higgins J, Gore R, Gutkind D, Mednick SA, Parnas J, Schulsinger F, Cannon TD (1997) Effects of child-rearing by schizophrenic mothers: a 25-year follow-up. Acta Psychiatr Scand 96: 402– 404. Hipwell AE, Kumar R (1996) Maternal psychopathology and prediction of outcome based on mother-infant interaction ratings (BMIS). Br J Psychiatry 169: 655–661. Houzel D (1994) Les applications thérapeutiques de l’observation directe dans le champ de la psychiatrie. Devenir 2: 79–86. Kendell RE, Chalmers JC, Platz C (1987) Epidemiology of puerperal psychoses. Br J Psychiatry 150: 662–673. Klompenhouwer JL, van Hulst AM (1991) Classification of postpartum psychosis: a study of 250 mother and baby admissions in The Netherlands. Acta Psychiatr Scand 84: 255– 261. Kringlen E, Cramer G (1989) Offspring of monozygotic twins discordant for schizophrenia. Arch Gen Psychiatry 46: 873– 877. Kumar R, Hipwell AE (1996) Development of a clinical rating scale to assess mother-infant interaction in a psychiatric mother and baby unit. Br J Psychiatry 169: 18–26. Kumar R, Marks M, Platz C, Yoshida K (1995) Clinical survey of a psychiatric mother and baby unit: characteristics of 100 consecutive admissions. J Affect Disord 33: 11–22. Lamour M (1989) Les nourrissons de parents psychotiques. In: Lebovici S (ed) Psychopathologie du bébé. PUF, Paris, pp 655– 673. Lebovici S (1985) Les enfants de familles psychotiques In: Lebovici S, Diatkine R, Soulé M (eds) Traité de psychiatrie de l’enfant et de l’adolescent. t3, PUF, Paris, pp 375–384. Lebovici S, Weil-Halpern F (1989) Psychopathologie du bébé. PUF, Paris, (882p). Lovestone S, Kumar R (1993) Postnatal psychiatric illness: the impact on partners. Br J Psychiatry 163: 210–216. Lussier V, David H, Saucier JF, Borgeat F (1995) La dépression postnatale, sa détection, sa mesure, son étiologie. Le Carnet Psych 95: 13–15. Mednick SA, Schulsinger F (1965) A longitudinal study of children with a high risk for schizophrenia: a preliminary report. In: Vandenberg S, Orlando F (ed) Methods and goals in
58 human behavior genetics. Academic Press Inc, New York, pp 255–296. Milgrom J, Burrows GD, Snellen M, Stamboulakis W, Burrows K (1998) Psychiatric illness in women: a review of the function of a specialist mother-baby unit. Aust N Z J Psychiatry 32: 680– 686. Mowry BJ, Lennon DP (1998) Puerperal psychosis: associated clinical features in a psychiatric hospital mother-baby unit. Aust N Z J Psychiatry 32: 287–290. Murray L (1992) The impact of postnatal depression on infant development. J Child Psychol Psychiatry 33: 543–561. O’Hara MW, Swain AM (1996) Rates and risk of postpartum depression – A meta-analysis. Int Rev Psychiatry 8: 37– 54. Poinso F, Rufo M (2000) Quelle éthique pour l’évaluation de la relation entre une mère et son bébé? Hospitalisations conjointes mère-bébé et séparations. Neuropsychiat Enfance Adolesc 48: 139–148. Poinso F, Soulayrol R (1998) La place du père dans une unité d’hospitalisation parents-enfants. In: Psychiatrie périnatale. PUF, Paris, pp 377–390. Poinso F, Delzenne V, Thirion X, Sokolowsky M, Samuelian JC (1996) La dépression dans le post-partum. Recherche épidémiologique et clinique depuis l’accouchement. Sem Hôp Paris 3–4: 69–78. Pritchard DB, Harris B (1996) Aspects of perinatal psychiatric illness. Br J Psychiatry 169: 555–562. Ramsay R, Kumar C (1996) Ethical dilemmas in perinatal psychiatry. Psychiatric Bull 20: 90–92.
F. Poinso et al.: Mother-baby psychiatric unit: separation at discharge Riordan D, Appleby L, Faragher B (1999) Mother-infant interaction in post-partum women with schizophrenia and affective disorders. Psychol Med 29: 991–995. Rosenthal D (1971) A program of research on heredity in schizophrenia. Behav Sci 16: 191–201. Rosenthal D, Wender PH, Kety SS, Welner J, Schulsinger F (1971) The adopted-away offspring of schizophrenics. Am J Psychiatry 128: 307–311. Schulsinger H (1976) A ten-year follow-up of children of schizophrenic mothers. Clinical assessment. Acta Psychiatr Scand 53: 371–386. Snellen M, Mack K, Trauer T (1999) Schizophrenia, mental state, and mother-infant interaction: examining the relationship. Aust N Z J Psychiatry 33: 902–911. Stern D (1981) Mère-enfant. Les premières relations. P Mardaga (ed) Bruxelles. Stewart DE (1989) Psychiatric admission of mentally ill mothers with their infants. Can J Psychiatry 34: 34–38. Sutter AL, Bourgeois M (1998) Unités mère-enfant en psychiatrie périnatale. EMC Psychiatrie, Paris, France, 37204L10, 7-1984, 6 p. Yarden P, Max D, Eichenbach Z (1966) The effect of childbirth on the prognosis of married schizophrenic women. Br J Psychiatry 112: 491–499. Correspondence: Dr. François Poinso, Psychiatric Parent-child Unit, Hôpital de Sainte-Marguerite, Service de Psychiatrie de l’Enfant, CHU Marseille, F-13274 Marseille Cédex 09, France; e-mail:
[email protected]
