Home Treatment as an Alternative to Hospital Admission for Mothers in a Mental Health Crisis: A Qualitative Study

This study was carried out in the inner London boroughs of Camden and Islington, which have a total population of 373,817. High levels of deprivation give this area one of the highest estimated levels of psychiatric morbidity in the United Kingdom ( 17 ). In these two boroughs patients presenting with a mental health crisis severe enough to require hospital admission can be treated in one of three settings: at home (by one of four CRTs), in a hospital, or at a crisis house for women only ( 18 ). The latter can accommodate 12 women and up to four children for up to four weeks. Local service policy recommends the assessment of needs of children of any patient, and where there are concerns a referral should be made to the local Children and Families Social Services.

We recruited female patients treated by a CRT in the previous 18 months. Inclusion criteria were full or partial responsibility for a child younger than 18 years, proficiency in spoken English, stable mental state at the time of interview, and ability to give informed consent to participate in the study. Patients were not approached to participate in the study if they had recently lost custody of a child, in order to avoid insensitivity and potential distress. We used purposive sampling to recruit patients with different diagnoses, treatment histories, levels of social support, and children's ages. Patients were identified through CRT staff. For patients who were under the care of a community mental health team, the team was contacted to ascertain the patient's most recent contact and mental state. Eligible patients were sent an information sheet and a consent form and were then contacted by phone by CRT staff, community mental health team staff, or the researchers, as appropriate. After full explanation of the study aims and an opportunity to ask questions, informed written consent was obtained from those willing to participate. Interviews took place at a location of the participant's choice. After the interview the women were given two information sheets on parental mental illness, one that was aimed at parents and was produced by the mental health charity Mind ( www.mind.org.uk ) and another that was aimed at children and was included in a Department of Health training manual ( 15 ). Participants were given a £10 voucher in appreciation of their time.

After completing interviews with the mothers, we conducted further interviews with their children, in order to triangulate the findings. We sought permission from all parents of children aged 12 to 18 years to contact these children. Information sheets and consent forms were then sent to the children, and informed written consent was obtained from those willing to participate. Children were given the option of being interviewed alone or with a family member present. They were given a £10 voucher in appreciation of their time. The study was approved by the local research ethics committee.

A semistructured interview was used. The interview topic guide was based on previous research in the area ( 11, 15, 16 ) and on the specific aims of the study, which were to explore the following: patients' experiences of home treatment, children's experiences of home treatment, treatment preferences, and unmet needs.

Interviews with patients were conducted over a six-month period in 2006 and 2007 by one of two researchers (HKh or CM). Both researchers are specialist registrars in general adult psychiatry and were not directly involved in the participants' clinical care. Interviews with the children were carried out by MF, a specialist registrar in child and adolescent psychiatry, with a semistructured interview to explore themes emerging from the maternal interviews.

All interviews were audiotaped and independently transcribed. Information identifying the interviewer was removed. Content analysis was carried out with the support of QSR N6 NUD*IST software ( 19 ). A coding frame was developed that contained themes relevant to the aims of the study. The interview guide topics were the starting point, and they were included as themes in the coding frame at the beginning of the process of analysis. The coding frame was then elaborated and modified as new themes and subthemes emerged in the course of the analysis. The development of the coding frame thus combined deductive elements (based on a priori themes to be explored) and inductive elements (based on themes emerging directly from the data) ( 20 ). Negative case analysis was used to refine the coding framework, by searching for elements of the data that seemed to contradict emerging themes. Interviews were analyzed throughout the study period in groups of two to four interviews, with findings used to refine the interview schedule and guide further analysis.

All patient interviews were independently coded by HKh and CM. All child interviews were independently coded by HKh and MF. A consensus on the final coding frame was established through discussion with all coders.

Of the 35 patients we attempted to contact, 18 (51%) agreed to participate, seven (20%) could not be contacted, and ten (29%) declined to participate. The demographic and clinical characteristics of participants are outlined in Table 1 . Of note is that half of the participants were single parents. Five (28%) had past or current involvement with Children and Families Social Services. Four (22%) had children on the "at need" register through Children and Families Social Services, where a family is assessed as requiring additional support without significant risk issues, and one (6%) had children on the "at risk" register, because there was significant concern about child abuse or neglect.

Nonparticipants were more likely than participants to have a psychotic illness (four of 17 nonparticipants, or 24%, compared with two of 18 participants, or 11%), but there were no differences in treatment history, the number or ages of children, living circumstances, or involvement of social services.

Of the ten mothers with children aged 12 to 18 years, four gave consent for their children to be approached for interview and six could not be contacted. Two children of these four consenting patients did not attend arranged interviews and could not be subsequently contacted. Five children from two families composed the final sample (two siblings from one family and three from the other). Children were aged 12 to 18 years and lived either with both parents or with the father. All children had significant contact with the mother during the crisis, and they had also experienced maternal hospitalization in the past.

Mothers' experiences of home treatment. Overall, the positive aspects of being treated at home related to participants' needs as patients, whereas the negative aspects related to their roles as parents. The two most commonly described advantages of home treatment were receiving good-quality care from the CRT and avoiding unwanted hospital admission. One mother said, "They [the CRT] were really good, you know, and they sort of understood as well. They didn't look at you like you were from cuckoo land … they're always polite, always there for you." Another mother described how she preferred home treatment over hospitalization when she said, "[The] hospital is like a German concentration camp, you know."

Disadvantages to home treatment commonly described by patients were all related to difficulties with parenting while in crisis, with the following five common themes emerging. The first theme consisted of difficulty in meeting the children's physical needs and maintaining normal boundaries. One single mother of two young boys said, "I found that there were no boundaries. … I didn't feel like cooking, so what I would do is spend money that I didn't have, and just give them money to buy takeaways and things like that, and letting them play out[side] when normally I would put my foot down."

The second theme involved a lack of emotional connectedness with the children. A single mother of two teenage girls said, "I feel numb, and I feel like I've got no feeling … and that bugs me a lot. It's really hard. I mean, I know I do love them, but I don't feel nothing." A mother of a toddler expressed the same types of feelings when she said, "I just felt that I wasn't really necessarily interacting with her well because I was so switched off. … I just felt within myself that I was failing as a mother."

The third theme involved exposing the children to distressing symptoms or behaviors. A married mother of three young and teenage girls said, "When I self-harmed, she went in her bedroom, and she picked up her dolly blanket and put it over my head, and she said, 'Don't do it, mummy, it's sore.' And she was only two." And a single mother of a teenage boy said, "[My son] said, 'You used to sit out there half naked, and the way you sat in your sarong and you didn't have any underwear on, and it was disgusting.' And I had no idea that I wasn't properly dressed."

The fourth theme consisted of directly incorporating the children in their symptoms. For example, a married mother of four boys aged four to 18 years said, "There was this dreadful thing where I thought about him [one of my sons] being put in an oven, and it completely and utterly freaked me out, to be honest with you. … I'd never in a million years want to put him in an oven, you know."

The fifth theme involved burdening the child with caregiving responsibilities. One single mother of a teenage boy said, "I'm reliant on him physically to go to bed, physically to get up, emotionally because he's my one and only contact. And it's almost like sometimes I am the child, and he's the parent."

Children's experiences of home treatment. We explored participants' perceptions of their children's experiences and needs during home treatment, and we compared themes emerging from the parent and child interviews. Although parents were able to describe their own struggle to meet the children's needs, most appeared to have difficulty in reflecting on what the children themselves experienced, thought, and felt. They described this difficulty as being mainly due to poor communication, both during the crisis and afterward. The shaded box on this page lists common themes for poor communication between parents and children and provides quotes as examples. Some parents were able to describe the confusion, worry, anger, and resentment that their children felt during the crisis. One extreme example was given by a woman whose son once wished she would commit suicide, so as to gain relief from the responsibility of caring for her. This single mother said, "I can't remember if it was 75% or 25% of people who are bipolar commit suicide. And he said, 'Do you know what I thought, mum? And this is really, really what I thought. I thought, yeah, why doesn't she just do it? Then we'll have a funeral, it will all be over, and I'll just be the poor boy who lost his mum.'… [He said this] because of the responsibility of me … the emotional responsibility."

The main findings of this study were that mothers treated at home were positive about the quality of care they received as patients, but they struggled to provide adequate parenting. They found it difficult to reflect on the children's needs and to access help for the children from professionals. Children described difficulties not highlighted by parents, such as hostility from parents and fear of them. Despite the difficulties, many patients preferred home treatment over hospital admission, where they felt safer and better looked after. In contrast, most children preferred parental admission, because it relieved them of worry. Both parents and children were positive about admission to a crisis house. The findings from this study suggest that although home treatment may meet patients' treatment needs, children may be exposed to a number of risks, such as parental hostility, poor parental insight, poor communication, and lack of access to professional support.

Before the implementation of CRTs, parents in acute crisis who presented to services were hospitalized. Previous qualitative studies show that parents are insightful and articulate about their difficulty in meeting the children's needs in a crisis and about the potential benefits of hospital admission ( 21 ). This is the first study to investigate experiences of home treatment as an alternative to psychiatric admission. We found that although mothers were insightful about their difficulty in providing adequate parenting during home treatment, this did not influence their treatment preference—most preferred home treatment over hospital admission. As with previous studies, our study found that although mothers recognized the child's need for support, they were often reluctant to seek professional help because of fear about custody loss or the wish to protect the child ( 11 ). We found that parents struggled to reflect on the children's own experiences and needs. This mirrors past studies, where accounts of children's experiences by parents and by the children themselves were found to differ, because the parents' accounts were in part colored by their mental state and anxieties ( 22 ). It is of note that although patients in this study focused on the harmful effect of lack of care, children themselves were more distressed by parental hostility and fear of their parents. Past studies highlight parental hostility as an important risk factor in child morbidity, especially when the child has a poor understanding of parental illness and behavior ( 23 ). The risks may be attenuated by offering families psychoeducational interventions, which have been shown in randomized clinical trials to improve communication and reduce child morbidity ( 24, 25 ).

The qualitative design of our study is well suited for exploration of patients' experiences and views. We attempted to make our sample as representative as possible by purposively selecting people with the range of diagnoses and characteristics likely to be encountered by any CRT. The participants had varied past treatment experiences, allowing for in-depth exploration of treatment preferences. Triangulation through interviews with the children allowed for exploration of the emergent themes from an alternative perspective and pointed to potential conflict of preferences and interests between children and parents. However, the findings of our study were limited by the relatively low participation rate, and it is possible that women with a positive experience of home treatment were more likely to participate. Hence our findings cannot be considered fully representative of CRT patients with children. We were able to interview only a small sample of children, but these children raised important concerns not mentioned by their parents, and these views have not been previously explored in other studies. The study focused on mothers, not fathers. The latter group may have unique experiences and needs and warrant an independent study.

This study suggests that mothers with child care responsibility who experience an acute mental health crisis prefer home treatment to hospital admission but that their children may subsequently be exposed to significant risks. When these parents are admitted to a hospital, the care and safety of their children are likely to be considered, especially if the patient is a single parent, and this may involve assessments by statutory organizations such as Children and Families Social Services. We suggest that when these patients are treated at home, the assessment of their children's needs is as important for patients as it is for those treated in the hospital.

Currently a major impediment to parents' seeking help with child care during a mental health crisis is fear of losing custody. Ready access to support from a source other than the agencies that implement child protection legislation may be a better alternative. Where there is a high proportion of CRT service users who have children, staff members who have specialized training in family support may be able to access additional help for the family during the crisis. Relatively simple interventions, such as access to nursery day care or home child care, may be effective at reducing parental distress and risks to the children, particularly for more socially isolated parents. Specialist projects for young caregivers, including children of people with mental illnesses, may also have a role. For example, Cooklin ( 26 ) described how young caregivers found a series of workshops useful and popular. In these workshops young caregivers were encouraged to express their feelings and experiences through storytelling and games. Enhanced access to such projects at the time of a crisis may fulfill some of the main needs of children who are caregivers for parents with mental illness—that is, it gives children a place where they can have a frank discussion of their situations, have contact with others with similar experiences, and receive guidance on how to respond to their parents ( 26 ).

When parents are treated at home for an acute mental health crisis, both parents and children struggle to cope, but they are reluctant to seek professional help. A significant proportion of psychiatric patients have child care responsibility, and there is an international trend toward home treatment for all phases of illness. The challenge for service providers is to develop acceptable and accessible help for families, incorporating interventions that address unmet needs and barriers to help seeking that have been articulated by the families themselves.

This study was funded by the Department of Mental Health Sciences, Hampstead Campus, University College London.

The authors report no competing interests.