Abstract
Background:
Family interventional programmes are effective adjuncts to pharmacotherapy in patients with schizophrenia. Modification in content of such programmes in response to local challenges is considered important, but has not been fully explored in Africa.
Aims:
To assess the feasibility and acceptability of an interventional family study for people with schizophrenia and their families in a socially deprived urban community in South Africa and to explore the contextual factors that could influence implementation of the intervention.
Method:
A psychiatric nurse facilitated semi-structured interviews with four multi-family groups, each comprising adult outpatients with schizophrenia and their caregivers. Six sessions were held per group. Thematic analysis was applied.
Results:
Three themes emerged: stigma and abuse; substance abuse comorbidity and caregiver burden of multiple stressors. Many of these stressors relate to the challenges of an impoverished urban environment.
Conclusions:
Multi-family groups with a psycho-educational and behaviour modification frame are acceptable. Negative symptoms are seen as protective in areas of community violence. Modification of traditional models of family therapy to include factors related to poverty, violence, caregiver burden, stigma and limited health care access should be considered in this setting.
Introduction
Schizophrenia is a severe, disabling illness characterized by a range of symptoms including psychosis, apathy and withdrawal, mood changes, as well as cognitive impairment, which result in deterioration in social and occupational functioning (van Os & Kapur, 2009). Schizophrenia imposes a disproportionately large economic burden on patients and their families, health care systems and society, because of its early onset, devastating effects and commonly lifelong course (Glazer & Johnstone, 1997).
Randomized controlled trials demonstrate that a combination of family support and education assists families and people with schizophrenia to lessen the frequency of relapse and to improve compliance with treatment (Pharoah, Mari, Rathbone & Wong, 2006). Despite good evidence from a range of contexts that family interventions are key components in packages of care for schizophrenia, mental health practitioners struggle to integrate family sessions into their caseload (Leff, 2000). There are further logistical difficulties; for example, it is far easier to treat a patient on their own than to arrange for an entire family to attend sessions – family members may face competing care demands and transport difficulties, and may have problems applying for leave from their jobs (Asmal et al., 2011). These challenges may well be exacerbated in low-income contexts (Kritzinger, Swartz, Mall & Asmal, 2011). Possibly because of these challenges, despite a range of published studies on the pharmacological treatment of schizophrenia in sub-Saharan Africa, we have not found any studies exploring the efficacy of family treatments for schizophrenia in this region.
Stigma and discrimination, low mental health literacy and low priority of mental health may all contribute to the challenge of mental health care delivery in sub-Saharan Africa (Chisholm et al., 2007; Lund et al., 2011; Patel et al., 2010). Cities are growing at faster rates all over the world but fastest in Africa where, by 2020, 48% of the population will be urban (UN-habitat & UNHSP, 2005). The urban African environment comes with its own set of challenges such as socio-economic deprivation, economic inequality, substance abuse and social fragmentation (UN-habitat & UNHSP, 2011). As a first step towards formally trialling family interventions for schizophrenia in an urban community of South Africa, we were interested in the question of how these challenges might impact on the provision of a family support service in a real-world, low-income African context. We chose a multi-family group model that included patients and caregivers due to severe resource constraints and to explore the acceptability of this method of treatment. We report here on findings from a qualitative study of a group family intervention in which we provided people with schizophrenia and family members the opportunity to experience and comment on both the planned intervention itself and on the contextual factors that could influence future implementation of the treatment.
Methods
Participants
Participants for this study were recruited from a subset of a larger first-episode schizophrenia study conducted at the University of Stellenbosch, South Africa. Participants with first-episode schizophrenia were eligible for inclusion if they had experienced remission of psychotic symptoms and were over the age of 18. All participants in the larger first-episode schizophrenia study were on treatment with a depot antipsychotic (flupenthixol decanoate). We recruited relatives of the patients’ choice who were over the age of 18. All participants signed informed consent. Relatives and participants with schizophrenia were excluded if they had a severe physical illness that could impede participation or had intellectual disability.
Thirty-four patients enrolled in the larger study had achieved remission and were eligible for inclusion. Reasons for declining to participate were as follows: two patients did not have a relative available to attend; two were abusing substances heavily and relatives felt that treating the substance abuse was a priority; two declined because of transport problems; four declined because of work commitments; one because of physical illness in the family; and three family members did not feel ready for a family intervention. We therefore recruited 20 patients with first-episode schizophrenia and 20 of their relatives. Twelve mothers, two fathers, three aunts, one grandmother, one cousin and one guardian of patients with schizophrenia agreed to participate.
Family groups
Participants were divided into four groups according to language preference. Each group consisted of six to eight participants. Three groups were run in Afrikaans and one in English. Six sessions were held for each group.
Structure of the intervention
These semi-structured sessions were held every two weeks, lasting 90 minutes each, and each session was guided by a topic. The initial session focused on the biology of psychosis. Subsequent sessions focused on medication and side effects; stress and expressed emotion; communication techniques; problem solving; and crisis planning. During these sessions, patients and families were invited to discuss their understanding of the topic of discussion, explore the factors that may influence adherence to a family intervention programme and consider what format such a programme might take.
Development of the session guides
Although family therapy studies vary in terms of aims, session content and number of sessions, a Cochrane systematic review found that any family intervention of more than five sessions was associated with decreased service utilization compared to standard care (Pharoah et al., 2006). The review also found that the common features of family interventions were psycho-education, management to reduce expressed emotion within families, and support (Pharoah et al., 2006). Session guides were therefore developed with this in mind and in collaboration with a psychiatrist with previous experience in conducting family intervention groups for schizophrenia. Colleagues with substantial experience in qualitative research then reviewed the questions and verified that they could potentially elicit sufficient information. The nurse who facilitated the groups had over 20 years of experience in community psychiatry and had received additional training in qualitative interviewing techniques.
Ethical approval
Ethical approval to conduct the study was granted by the University of Stellenbosch Human Research Ethics Committee. The study was conducted in accordance with Research Guidelines issued by the Medical Research Council (MRC) of South Africa. After a complete description of the study to the participants, written informed consent was obtained. Confidentiality was assured for both patients and relatives.
Analysis
All sessions were audio-recorded with permission of the participants, and these recordings were transcribed. The transcriptions were analysed using thematic analysis, guided by the questions in our interview guide (Silverman, 2004). The initial code book was developed by two members of the research team and then reviewed and modified by the remaining team members. Using the revised code book, the researchers coded a number of transcripts together and then iteratively discussed similarities and differences in their codes. Coding discrepancies were discussed until consensus regarding codes was achieved. Attempts were thereby made to achieve some level of internal validity of the codes. Data were managed using NVivo 7, a qualitative coding programme.
Results
We report first on the feasibility then the acceptability of the intervention, and then on key themes that emerged from the data.
Feasibility of family intervention
Twenty patients and 20 relatives were recruited to participate in the study. Overall adherence to the sessions was 75.0% (79.5% among relatives; 70.5% among patients). Attendance rates remained stable without a drop-off as the sessions progressed. Several measures were implemented to enhance adherence: sessions were arranged to coincide with scheduled clinic treatment visits; the study coordinator telephoned a reminder to relatives the day before each session; relatives and patients were modestly reimbursed for transport costs.
In most cases of non-attendance, participants did not provide a reason. Two participants missed sessions because of transport difficulties; one patient and his relative did not attend a session because of negative symptoms; and two participants stopped attending because of work commitments. Due to high unemployment rates among patients and their relatives in keeping with the high unemployment rates of South Africa, poor adherence due to work commitments was not a significant reason for non-adherence.
Acceptability of the intervention format and participant involvement
Patients and relatives agreed that a psycho-educational frame helped to address gaps in their knowledge about the biology of schizophrenia and its treatment. Participants found the content of each session relevant and accessible. While participant discussion by relatives and patients was comparable around certain topics (such as the biology of schizophrenia, communication techniques and problem solving), there were differences in interest between patients and relatives in other topics of the programme. Patients, for example, placed more emphasis on physical and verbal abuse within communities, loneliness and difficulty abstaining from illicit substances. Relatives, however, emphasized hostile behaviour displayed by the patient especially when using substances, poverty, physical illness, lack of support from other family members and community violence as major stressors of being a carer.
Participants did not express concern about speaking in a group setting and no potential participant declined to join the study because of the multi-family format. Relatives felt that the opportunity to share experiences with others who faced similar challenges was valuable and helped to decrease the sense of isolation that many reported experiencing in their communities.
On the other hand, some relatives thought that it would be beneficial to have sessions that did not include the patient with schizophrenia. Relatives reported that these groups were frequently the only form of support available to them and it was sometimes difficult for them to speak freely about the burden of illness when the patient was present. If an implemented family intervention included patients, relatives felt that the intervention should comprise separate sessions for relatives only, or a parallel support group for relatives.
Key emergent themes
Effect of stigma and vulnerability to abuse
Both relatives and patients felt stigmatized by the community because of the illness. Relatives believed that their communities judged them for being unable to manage the illness better and that there was little understanding within communities of the limited availability of treatment resources. One relative said: ‘There must be help for us parents, because we are being degraded, they look at you in the community, like you are not handling your son properly. Is there anywhere I can go? No, nowhere, doors are closed.’
Relatives felt that patients with schizophrenia were at particular risk of exploitation and violent victimization. South Africa has an uncommonly high crime rate and the Western Cape area, where the study is based, is one of the most violent in the country (Silber & Getten, 2009). Relatives spoke about concerns of violence within their communities and how particularly difficult it can be to navigate for those with schizophrenia. They felt that patients often do not realize when they are at risk of being in unsafe situations and that criminals might target the patient because of the mental illness. For this reason, relatives described negative symptoms as potentially protective mechanisms that kept patients off the streets. Relatives also spoke of needing to be aware of where the patient was at all times even if symptoms were in remission: ‘I don’t know if it’s right, but now I need to always watch out for him. I need to watch out for him against the community.’
Relatives were also concerned that patients are targeted by gangsters and drug peddlers to sell drugs and run errands. Vulnerability to this exploitation was thought to be because of substance abuse, ignorance and difficulty making friends. One relative said: ‘They use him a lot on the street. He helps someone without working, but I feel the man is busy using him.’
Relatives spoke of needing to protect the patient from stigma and abuse within the family itself. Caregivers felt that other family members did not understand the illness and reacted with violence and abuse towards the patient when psychotic behaviour was displayed. As a result, caregivers verbalized feeling overwhelmed by having to protect the patient from stigma and abuse that occurred both within the community and within the family itself. One relative recalled: ‘I had to call the police one night because my husband and my oldest son wanted to hit him with a club.’ Regarding the perpetrators of physical and verbal aggression towards the patient, she said: ‘The family, yes the family. But also many people outside.’
Patients described feeling uncertain about how to deal with abuse and stigma against them. They recognized feeling hurt when taunted and felt that the most appropriate option for them to reflect that hurt was to be verbally abusive or physically violent in response. One patient said: ‘Ok, maybe small children because they like making jokes and maybe big people also talk. Talk that you are crazy and not right in your head and that makes me stressed and then I hit them.’ Similarly, another reported: ‘I tell them things that will hurt them, I will swear at their mother. I will do something, like I hit.’
Psychological
impact of multiple stressors on caregivers.
Along with coping with residual symptoms, related functional decline and substance abuse in the patient, relatives identified a number of extraneous stressors that impacted on their own psychological well-being and their capacity to act as caregivers. Poverty, lack of support from other family members, health conditions in themselves and other family members, and community violence were seen as major contributors to the stress of being a carer. Relatives described being unsupported by other members of the family, by their community and by social services. As a result, relatives felt helpless and overwhelmed by their situation. Some relatives’ experiences are illustrated by the following quotations:
I am alone. I feel alone. I must do everything alone. I can’t manage to do to it all alone.
I have, myself, also been admitted to hospital because of this illness [schizophrenia].Because I can’t cope with managing the stress of it.
I cannot run away from my problems but I want to, I feel like I just want to go.
The father has TB [tuberculosis], the brother has TB and both are so unhelpful. And I ask them if it’s the TB that makes them so or what else is going on?
Impact of substance abuse dual diagnosis
All patients included in the study had achieved remission of psychotic symptoms; however, a number of patients were actively abusing substances. Behavioural disturbance related to substance abuse was frequently the primary concern for relatives involved in caring for these participants. As a result, despite having a dedicated session on it, substance abuse concerns were also expressed during all other sessions by relatives. Relatives raised concerns about patients stealing from the home to buy drugs, and of aggressive behaviour due to substances. Relatives also expressed difficulty differentiating between the symptoms and signs of schizophrenia and substance abuse; for example: ‘[In schizophrenia]… the brain cells are tikked [on methamphetamine].’ and ‘He is drinking or he is drugging, I don’t know but he looks in a state when he comes home.’
Participants felt helpless in the face of the overwhelming availability of illicit substances in the local community. This is illustrated by one relative’s comment: ‘But the drugs are always available near us, so close to people who have children who have such illnesses.’ Similarly, a patient said: ‘Yes, it feels like drugs are everywhere you look. They stand under my flat, just there where I live, on the corner near my sister… and when I am on the road to church there they are selling drugs.’
Discussion
Participants expressed a positive attitude to family interventions for schizophrenia and a multi-family group facilitated by a mental health nurse is a relatively low-cost format to implement such an intervention. Certain principles of family therapy, such as empathy and psycho-education, are universal (Asmal et al., 2011) and the results of our study advocate for the integration of these basic principles of family intervention models into locally based care for people with schizophrenia and their families. However, several additional needs were identified that should be incorporated into a programme that is appropriate for this community. Patients described stigma within communities, substance abuse and loneliness as major concerns. Along with substance abuse and hostile behaviour displayed by the patient, relatives identified poverty, physical illness, lack of support from other family members and community violence as major contributors to the stress of being a carer.
Adherence to medication, which is frequently a concern for families in family intervention studies (McDonald, Garg & Haynes, 2002; Zygmunt, Olfson, Boyer & Mechanic, 2002), was not highlighted by participants of this study. The likely reason for this is that all participants with schizophrenia were treated with depot antipsychotic medication together with an assertive monitoring programme. Negative symptoms, particularly apathy, were a source of frustration for caregivers but were largely seen as protective. Social withdrawal meant that patients were not exposed to the hazards of roaming in unsafe neighbourhoods. Communities were seen as a source of external support for relatives as well as stigma.
Relatives and patients regarded substance abuse as a precipitant of the initial psychotic breakdown and a significant contributor to ongoing symptoms. Managing someone with schizophrenia and substance abuse is a considerable challenge for families in terms of care and for the clinician in terms of diagnosis and treatment. These ‘dual diagnosis’ service users are at higher risk of homelessness, medical problems, non-adherence, violence, readmission to hospital and suicide, and care is frequently costly and fragmented (Archie & Gyomorey, 2009; Khantzian, 1985). In developed countries there is evidence that an integrated family-based treatment model, in which the same therapeutic team jointly addresses schizophrenia and substance abuse, may be beneficial in managing these patients. Models for integration of services in South Africa are largely from the HIV/AIDS field. For example, to address high rates of HIV and tuberculosis (TB) co-infection, an integrated TB and HIV/AIDS services has been shown to maximize the potential to address both conditions in a coordinated manner (Corbett, Marston, Churchyard & De Cock, 2006). The cost-effectiveness and implementation challenges of a dual-diagnosis model of service integration for mental disorders in developing countries need further research.
The study aimed to investigate the feasibility and acceptability of a family intervention for schizophrenia, and the factors that could influence the implementation of such an intervention. Therefore, due to the time constraints for each session, one limitation of the study is that we were unable to actively explore positive coping strategies that are employed by patients and relatives to overcome the difficulties described during the sessions.
There are a number of possible reasons why adherence to the intervention was high (75%). First, practical measures were taken to enhance adherence, such as reimbursement for transport costs and synchronizing the family sessions with patients’ routine visits. Second, first-episode cohorts are more likely to engage with services than multi-episode patients (Addington, Collins, McCleery & Addington, 2005). Third, there was a clear desire for information, advice and support among participants.
Conclusion
A family group model for addressing issues faced by people with schizophrenia and their families seems acceptable and feasible in this urban African context. We need to keep in mind that families face additional challenges to that of caring for a relative with schizophrenia, particularly in developing countries. Traditionally, family therapy models have not addressed comorbidities such as depression in the caregiver, substance abuse in the patient with schizophrenia and other medical and psychiatric illnesses in the household. Modification of traditional models of family therapy to address factors such as poverty and its concomitants, violence, stigma, lack of support for caregivers and limited access to medical and mental health care should be considered. If resources permit, further development of the elementary themes of this study can be achieved by expanding the content of each theme over several sessions. The next step in the process is to design a randomized controlled trial of an intervention suitable for the urban African context and to measure the impact of this intervention, taking account of the challenges that this context brings.
Footnotes
Acknowledgements
The authors would like to thank Sister Hettie Benjamin for facilitating the family interviews and Dr Mathias de Fleuriot de la Coliniere for his contribution to the technical content of the groups.
Declaration of conflicting interest
Robin Emsley has participated in speakers/advisory boards and received honoraria from AstraZeneca, Bristol-Myers Squibb, Janssen, Lilly, Lundbeck, Organon, Pfizer, Servier, Otsuka and Wyeth. He has received research funding from Janssen, Lundbeck and AstraZeneca. All other authors declare that there is no conflict of interest.
Funding
Financial support for this study was received from the Subcommittee A, Stellenbosch University and the National Research Foundation, South Africa.