REVIEW Afr J Psychiatry 2011;14:367-371

Family therapy for schizophrenia:

cultural challenges and
implementation barriers in the
South African context

L Asmal1, S Mall2, J Kritzinger2, B Chiliza1, R Emsley1, L Swartz2

1Department of Psychiatry, Stellenbosch University, Stellenbosch, SouthAfrica

2Department of Psychology, Stellenbosch University, Stellenbosch, South Africa

Abstract
Family therapy is an effective, evidence based intervention for schizophrenia. This literature review explores the impact of culture
on family therapy as a treatment model for schizophrenia and examines how cultural beliefs impact on access to care. Although
there is a good deal of evidence to suggest that certain principles of family therapy such as empathy and psycho-education are
universal, there is a paucity of literature about the role of culture in designing family interventions for people living with
schizophrenia in a culturally diverse setting such as South Africa. It is well acknowledged that cultural ideologies influence families’
belief systems of schizophrenia, expected expressed emotion, and levels of stigma in relation to mental illness. Additionally, in
adapting models designed for first-world settings, consideration needs to be given to aspects such as language, educational level
and accessibility of mental health care facilities. Family therapists are increasingly recognising the need for the study and
implementation of evidence based culture-relevant and culture-responsive therapeutic techniques. These techniques need to be
cost-effective and will require training, supervision, staff support, and management input in order to become generally available.

Keywords: Schizophrenia; Family therapy; South Africa; Culture

Received: 20-08-2010
Accepted: 22-01-2011
doi: http://dx.doi.org/10.4314/ajpsy.v14i5.3

Introduction expressed emotion, stress and family burden are shown to

In 2008, the global prevalence of schizophrenia was 1% reduce the risk of relapse, improve medication adherence
with relapse prevention remaining a challenge to the long- and educate families about schizophrenia.2 Results from
term treatment of schizophrenia in developed and studies done in Europe, Asia and North America indicate
developing countries. An effective, evidence based that family interventions employing these strategies are
intervention for schizophrenia is family therapy. Family common and have demonstrated efficacy in randomised
therapy is a treatment model comprising either single or controlled trials.2
multi-family groups of relatives and/or patients.1 Family The treatment models are loosely based on the concept
interventions that psycho-educate, enhance the capacity to of family which varies and can represent a traditional
problem solve and use strategies to reduce levels of nuclear family (comprising parents and children) or an
extended family (comprising additional relatives such as
aunts, uncles and cousins).3 Throughout the review we refer
to family in its broadest, cultural sense.
Correspondence One model aims to ease the burden of schizophrenia on
Dr L Asmal
Department of Psychiatry, Stellenbosch University patients and their relatives (members of their immediate
PO Box 19090, Tygerberg, 7505, South Africa family such as parents or siblings or extended family such
email: laila@sun.ac.za as cousins or caregivers) by creating an atmosphere for

African Journal of Psychiatry • November 2011 367

 REVIEW Afr J Psychiatry 2011;14:367-371

problem solving in a mature climate.2 This is achieved by deserving of support, sympathy and special treatment.
discouraging hostility, overt criticism and excessive Furthermore, individuals with ukuthwasa are seen as
emotional involvement of patients by family members.4 The potentially curable. However, some people who experience
term “expressed emotion” (EE) was coined by Leff and ukuthwasa do not have a favourable outcome because they
Vaughn5 as a measure of families’ involvement in people are unable to undergo or complete the training to become
with schizophrenia. Accordingly, EE is regarded as an an indigenous healer. In those cases, a retrospective re-
important reproducible indicator of the relevance of social diagnosis of ukuphambana (madness) may be given.10
factors to the prognosis of schizophrenia.6 An appreciation Themes such as ancestral calling, jealousy, bewitchment,
of the relationship between EE and schizophrenia has been ‘evil eye’ and guilt are common explanatory models for
fundamental to the development of family interventions for schizophrenia in Africa. African patients with schizophrenia
schizophrenia. treated with antipsychotics and admitted to psychiatric
There is, however, a paucity of literature about how the hospitals are likely to believe that the treatment is only
structure and content of these interventions should be symptomatic, such as for behavioural control.12 These
adapted for people living with schizophrenia in a culturally patients may then seek further intervention from indigenous
diverse setting such as South Africa. It is well acknowledged healers in an attempt to address what they believe is the
that cultural ideologies do influence families’ belief systems actual cause of the illness. Families of patients with
of schizophrenia and access to mental health care. Research schizophrenia may also share these beliefs and assist their
has also shown that EE may vary in different cultural groups. relatives in seeking help from indigenous healers.
The implementation of family therapy in accordance with a Individualism and individual autonomy are prominent
patient’s culture at a conceptual, practical and philosophical features of the Western medical framework and the person
level should be well considered.7 is viewed as a self-contained entity. In contrast, traditional
This literature review has two aims. We aim to firstly African cultures are described as collective and emphasise
describe how cultural beliefs impact on access to care. the position and responsibility of the individual within the
Thereafter the review will explore the impact of culture on group. Family and kinship patterns assign different roles,
family therapy as a treatment model for schizophrenia. status and power among group members.13 This may
influence decision-making regarding treatment
The intersection between culture, interpretation of interventions and interactions between family members
mental illness and concept of family may differ from traditional western models. An awareness of
The concept of family, interpretation of mental illness and the contrasting concepts of ‘person’ between Western and
expected emotion all vary in relation to different cultural African cultures is important for the successful
groups. implementation of a family intervention.
For example, studies conducted in China that compared
rural and urban populations found substantial differences in Stigma, schizophrenia and culture
perceptions of the aetiology of mental illness. Participants in Culture could determine levels of stigma in relation to
a rural based study were more likely to see mental illness mental illness, a relationship that has been explored in a
as caused by external malevolent spiritual forces and number of South African studies. One such study looked at
families and were therefore more likely to seek help from the attitudes and beliefs of relatives of Xhosa patients with
traditional healers, such as shamans.7 On the other hand, schizophrenia and found that 67% of family members
families in an urban based study were more likely to blame believed that witchcraft or spirit possession was related to
the illness on internal factors, such as the failure to love or the onset of schizophrenia. However this belief did not
an inability to adapt to a competitive environment. necessarily safeguard the patient against stigma. A
Stigmatisation was also perceived to be greater in urban significant number of the relatives believed people with
areas and individuals with schizophrenia were less likely to schizophrenia to be dirty (52%) and dangerous (44%).12
be employed or to find a partner.8,9 A more favourable outcome has been described in
Diagnosis in African indigenous healing may be better patients in developing countries. It has been argued that
conceived as being related to theories of causation of traditional family support structures and higher levels of
illness rather than the classification of symptoms into family tolerance towards people with schizophrenia
categories as in the taxonomical model of the Diagnostic contribute to a better course and outcome in low- and
and Statistical Manual of Mental Disorders or the middle-income countries.14 This was originally reported
International Classification of Diseases.10 For example, spirit more than 20 years ago in a World Health Organisation
possession is the suggested aetiology of Amafufunyana, and collaborative longitudinal study in 10 countries.15,16,17,18
blame for any untoward behaviour is placed outside the More recently, analyzing the results from another World
patient.11 One explanatory theory of Amafufunyana is that it Health Organization collaborative project, the subject was
occurs most often in people who are in a vulnerable or re-examined. Once again, evidence of differences in the
powerless position10 and functions as a socially sanctioned course and outcome of the illness in favor of the developing
model in which experiences of inner conflict can be country centers was found. Potential sources of bias were
contained. The term ukuthwasa, on the other hand, is used examined, and none were able to explain the findings.19
to describe the state of emotional turmoil a person goes However, the notion that schizophrenia has a better
through on the path to becoming an indigenous healer.10 outcome in developing countries has itself been
Since cases of ukuthwasa are not considered to be under an challenged, as have the presumptions of the nature of family
individual’s control, the person who suffers from it is support for people with schizophrenia in developing

African Journal of Psychiatry • November 2011 368

 REVIEW Afr J Psychiatry 2011;14:367-371

countries.20 Studies have consistently found that long-term Theoretically, the attribution of psychotic symptoms to
outcome in schizophrenia is dependent on adequate spiritual forces may lead to relatives being more accepting
treatment. A recent study has again found that patients with of untoward behaviour since it is believed that the patient is
schizophrenia not on treatment have a higher mortality rate not fully in control of their actions29 thereby creating a
than patients treated with any antipsychotic medication.21 It natural ‘low-EE’ environment. Although the belief that
is thus unlikely that patients with schizophrenia in the symptom manifestation is largely outside the patient’s
developing world would have a better outcome when they control may be beneficial in some respects, this model of
have many barriers to accessing treatment.22 Access to illness focuses on just one cultural aspect in the complex
treatment varies within and between developing countries family dynamic in schizophrenia.
and is influenced by a number of factors including Studies examining the relationship between EE and
migration, urbanisation, economic insecurity, social relapse found that Mexican American samples have a lower
inequalities and changes in family structure and support proportion of high EE families than British and US Anglo
networks (such as change in caregiver roles after death American samples (although EE was still associated with
from HIV/AIDS).20 relapse in all groups).24,25 Similarly, in a multicentre family
A retrospective study conducted in Nigeria found that intervention study for first-episode schizophrenia based in
4% of people with schizophrenia were homeless despite Chandigarh, Aarhus and London, Leff et al.30 found a
coming from traditionally structured family backgrounds. significant association between EE and relapse in all
Studies based in Ethiopia, Nigeria and rural China have centres. However, compared with the North Indian sample,
shown homelessness or housing instability in people with the European centres had a much higher proportion of
schizophrenia to be greater than in many developed relatives with high EE (54% compared to 23%). The study
countries.20 In developing countries where public mental investigators postulate that the better outcome at one-year
health resources are limited, families caring for people with follow-up in the North Indian sample may be due to the
schizophrenia may struggle with the strain of coping with a high proportion of relatives with low EE.
chronic and severe mental illness. In the absence of support Alvidrez31 concluded that African Americans were more
for the families themselves, these traditionally supportive likely to attribute the symptoms of schizophrenia to
family networks may break down with time. This may place religious or spiritual factors, which made caregivers more
the patient at risk of victimisation, homelessness and a accepting of odd behaviour and thoughts and less
poorer outcome.20 controlling of psychotic behaviour. Likewise, in a study by
Breitborde et al32, high EE caregivers were more likely to
Culture mediating the relationship between family believe that the ill relatives’ actions were as a result of
therapy and schizophrenia human agency (and thereby more likely to be intentional
Culture has been shown to influence EE and studies have and within the ill relatives’ control) than low EE caregivers.
explored the role of cultural ideas and practices in relapse. The authors postulate that the caregiver infers what
In western based studies, a relapse rate of over 50% is meaning or goal the ill relative attempts to convey during
experienced when patients return to high-EE families periods of illness and this process of understanding is
compared to a 20% relapse rate when patients return to a fundamentally a cultural one. This heuristic framework may
low-EE family setting.23 be a useful way to explore the influence of indigenous
There have been a number of studies among Mexican healing on schizophrenia in South Africa.
Americans.24,25 Kopelowicz et al25 compared the tolerance In a convenience sample of 187 African, European and
threshold of hostility and criticism by others of Mexican Latino American women at a general medical clinic,
Americans and Egyptians with schizophrenia with that of Alvidrez31 found that women who believed that psychiatric
British and US Anglo Americans. They found that Egyptians symptoms had a religious or supernatural basis were less
and Mexican Americans with schizophrenia tolerated likely to attend mental health centres for treatment. In the
higher levels of critical comments and hostility than British same study, it was found that if friends or family had
or US Anglo Americans.25 Similarly, African American previously attended mental health centres, women with
people who experienced caregiver criticism, a key current psychiatric symptoms were 2.5 times more likely to
component of high EE, did not necessarily relapse.25,26 seek treatment for themselves. Certain cultural groups, such
African American caregivers’ intrusive behaviour during as Latino Americans, may to be more likely to believe that
conflict may be interpreted as a sign of sincerity and the family shares responsibility for an individuals
genuine concern and the interactional process is less likely problems33 and that mental illness is best treated with the
to predict relapse symptoms.26 There is a conceivable risk family.34 Studies have found that family members who
of increasing stress should a clinician unknowingly focus on believe that their relatives with schizophrenia are in control
reducing critical and intrusive behaviours in families where of their symptoms are more likely to drop out of family
these types of interactions are positively adaptive.27 therapy.35
Nonverbal behaviours may also be interpreted A study based in Chennai, India, found that women with
differently by clients of different cultural backgrounds. The schizophrenia were stigmatised by the illness itself and by
meaning of maintaining eye contact, the concept of personal the social judgment of divorce and separation.36 Additional
space and how families arrange the seating during a session evidence from Chennai suggests that even when mental
can vary between cultures and the unknowing clinician may health facilities are accessible and free of charge, living in
misidentify salient issues by not fully understanding the an extended family is a risk factor for not being brought in
cultural context of the nonverbal behaviour.28 for treatment.37

African Journal of Psychiatry • November 2011 369

 REVIEW Afr J Psychiatry 2011;14:367-371

Implementation barriers concept of the family in relation to schizophrenia should be

Any psychosocial intervention requires finances, time, examined as an additional area of research.
adequate facilities, and trained, motivated implementers all Globally, the gap between showing efficacy and the
of which may be affected by larger attitudinal, knowledge routine implementation of effective services appears
and systemic implementation obstacles.1 The heavy particularly hard to bridge.23 A proposed family
workload of busy mental health professionals in a limited intervention programme adapted for the local context
resource setting needs to be considered as well as the needs to be cost-effective and will require training,
impracticality of large-scale implementation of family supervision, staff support, and management input in order
interventions by psychiatrists or psychologists given the to become generally available. To ensure buy-in from policy
shortage of both groups of professionals in South Africa.38 makers and funders, potential barriers to service provision
Studies have shown that knowledge about evidence based and the design of a cost-effective family intervention
advantages of family psychoeducation (such as reducing programme needs further research. From a clinical
hospitalisation or relapse rates) do not convince most perspective, understanding the patient’s and family’s
clinicians to change their working habits and adopt new explanatory model fosters trust and is essential to a
clinical practices for schizophrenia.39 In the planning of any collaborative treatment plan.41
intervention for family therapy, local operational barriers
need to be considered.1 References
For family members, a number of practical concerns 1. McFarlane WR, Dixon L, Lukens E, Lucksted A. Family
such as transportation, work commitments and competing psychoeducation and schizophrenia: A review of the literature.
demands on energy and time may influence adherence Journal of Marital and Family Therapy 2003; 29(2), 223.
(Kritzinger J, Swartz L, Mall S, Asmal L. Family therapy for 2. Pharoah F, Mari J, Rathbone J, Wong W. Family intervention for
schizophrenia in the South African Context: Challenges and schizophrenia. Cochrane Database of Systematic Reviews 2006; 4,
pathways to implementation. 2010; Unpublished literature 1-108.
review). 3. Doherty WJ, Boss PG, LaRossa R, Schumm WR, Steinmetz SK.
Relatives of people with schizophrenia may also feel Sourcebook of Family Theories and Methods: A Contextual
stigmatised by the illness and consequently may feel Approach. New York: Plenum Press. (1993).
uncomfortable revealing details about psychiatric illness in 4. Brown GW, Birley JL, Wing JK. Influence of family life on the
their families or with being identified as regular attenders at course of schizophrenic disorders: a replication. British Journal of
psychiatric facilities.1 Psychiatry 1972; 121(562), 241-58.
5. Leff J, Wig NN, Ghosh A, Bedi H, Menon DK, Kuipers L, et al.
Conclusions Expressed emotion and schizophrenia in north India. III. Influence
Family interventional programmes are widely used of relatives’ expressed emotion on the course of schizophrenia in
internationally and are effective adjuncts to Chandigarh. Br J Psychiatry 1981; 151,166-73.
pharmacotherapy in patients with schizophrenia.1 It is, 6. Lopez SR, Nelson HK, Polo AJ, Jenkins JH, Karno M, Vaughn C, et
however, crucial when implementing such programmes to al. Ethnicity, expressed emotion, attributions, and course of
consider cultural factors that may influence the expression schizophrenia: family warmth matters. Journal of abnormal
and experience of schizophrenia in fundamental ways. psychology 2004; 113(3), 428-39.
There is a good deal of evidence to suggest that certain 7. Tseng WS, Stretzler J. (Eds.) Culture and psychotherapy: A guide
principles of family therapy such as empathy and psycho- to clinical practice. Washington, DC: American Psychiatric Press,
education are universal and are suitable for implementation 2001: 189-200.
in South Africa.2 Integration of these basic principles into 8. Li SX. Witch doctors and mental illness in mainland China: a
locally based family intervention programmes is suggested preliminary study. Am J Psychiatry 1990.147, 221–224.
regardless of the patient’s and family’s cultural background. 9. Phillips MR. The transformation of China's mental health services.
To ensure successful family based therapy models in China J 1998; 39, 1–36.
South Africa, health care workers would require training in 10. Swartz L. Culture and mental health: A South African view. Oxford
local cultural differences and in the customs of traditional University Press, 1998.
healing systems. Training will also need to raise awareness 11. Lund C. Xhosa speaking schizophrenia patients’ experience of
in clinicians of their own perspectives and biases. their condition: Psychosis or amafufunyana? Psychology honours
Additionally, in adapting models designed for first-world research project, University of Cape Town.
settings, consideration needs to be given to aspects such as 12. Mbanga NI, Niehaus DJ, Mzamo NC, Wessels CJ, Allen A, Emsley
language, educational level and accessibility of facilities. RA, et al. Attitudes towards and beliefs about schizophrenia in
Although family interventions for schizophrenia have Xhosa families with affected probands. Curationis 2002; 1:69–73.
been studied in a range of international settings there is 13. Pacquiao DF. Cultural competence in ethical decision making, in
almost no systematic research on this topic on the African Transcultural concepts in nursing care Andrews MM, Boyle JC
continent. Family therapists are increasingly recognising the (Eds.). Philadelphia, PA., Lippincott, 2008: 411-412.
need for the study and implementation of evidence based 14. Harrison G, Hopper K, Craig T, Laska E, Siegel C, Wanderling J, et
culture-relevant and culture-responsive therapeutic al. Recovery from psychotic illness: a 15- and 25-year
techniques.40 Further research is needed to examine international follow-up study. Br J Psychiatry 2001; 178, 506-17.
whether modifications in content and outcome of 15. Sartorius N, Jablensky A, Korten A, Ernberg G, Anker M, Cooper
international family therapy models is required for the South JE et al.Early manifestations and first-contact incidence of
African context. The role of culture in influencing the schizophrenia in different cultures. A preliminary report on the

African Journal of Psychiatry • November 2011 370

 REVIEW Afr J Psychiatry 2011;14:367-371

initial evaluation phase of the WHO Collaborative Study on assessment. Jossey-Bass, San Francisco, 2001:50-51.
determinants of outcome of severe mental disorders. Psychol 29. Mzimkulu KG, Simbayi LC. Perspectives and practices of
Med 1986; 16, 909-928. Xhosa speaking African traditional healers when managing
16. Ohaeri JU. Long-term outcome of treated schizophrenia in a psychosis. International Journal of Disability, Development &
Nigerian cohort. Retrospective analysis of 7-year follow-ups. J Education 2006; 53(4), 417-431.
Nerv Ment Dis 1993; 181, 514-516. 30. Leff J, Kuipers L, Berkowitz R, Sturgeon D. A controlled trial of
17. Saha S, Chant D, Welham J, McGrath J. A systematic review of the social intervention in the families of schizophrenic patients: two
prevalence of schizophrenia. PLoS Med 2005; 2, e141. year follow-up. British Journal of Psychiatry 1985; 146, 594-600.
18. Saha S, Welham J, Chant D, McGrathJ. Incidence of schizophrenia 31. Alvidrez J. Ethnic variations in mental health attitudes and service
does not vary with economic status of the country: evidence from use among low-income African American, Latina, and European
a systematic review. Soc Psychiatry Psychiatr Epidemiol 2006 41, American young women. Community Mental Health Journal
338-340. 1999; 35(6), 515-530.
19. Hopper K, Wanderling J. Revisiting the developed versus 32. Breitborde NJK, López SR, Nuechterlein KH. Expressed emotion,
developing country distinction in course and outcome in human agency, and schizophrenia: Toward a new model for the
schizophrenia: results from ISoS, the WHO collaborative follow-up EE-relapse association. Culture, Medicine and Psychiatry 2009;
project. International Study of Schizophrenia. Schizophr Bull 2000; 33(1), 41-60.
26, 835-846. 33. Sabogal F, Marõn G, Otero-Sabogal R, Marõn B, Perez-Stable E J.
20. Cohen A, Patel V, Thara R, Gureje O. Questioning an axiom: better Hispanic familism and acculturation: What changes and what
prognosis for schizophrenia in the developing world? Schizophr doesn’t? Hispanic Journal of Behavioral Sciences 1987; 9, 397-
Bull 2008; 34(2):229-44. 412.
21. Tiihonen J, Lönnqvist J, Wahlbeck K, Klaukka T, Niskanen L, 34. Edgerton RB, Karno M. Mexican-American bilingualism and the
Tanskanen A, Haukka J. 11-year follow-up of mortality in patients perception of mental illness. Archives of General Psychiatry 1971;
with schizophrenia: a population-based cohort study (FIN11 24, 286-290.
study). Lancet. 2009 Aug 22;374(9690):620-7. 35. Birchwood M, Cochrane R, Macmillan F, Copestake S, Kucharska
22. Saraceno B, van Ommeren M, Batniji R, Cohen A, Gureje O, J, Carriss M. The influence of ethnicity and family structure on
Mahoney J, Sridhar D, Underhill C. Barriers to improvement of relapse in first-episode schizophrenia. A comparison of Asian,
mental health services in low-income and middle-income Afro-Caribbean and, white patients. Br J Psychiatry 1992; 161,
countries. Lancet 2007; 370(9593):1164-74. 783-90.
23. Kuipers E. Family interventions in schizophrenia: evidence for 36. Thara R, Kamath S, Kumar S. Women with schizophrenia and
efficacy and proposed mechanisms of change. Journal of Family broken marriages – doubly disadvantaged? Part II: family
Therapy 2006; 28(1), 73-80. perspective. Int J Soc Psychiatry 2003; 49(3), 233-40.
24. Karno M, Jenkins JH, de la Selva A, Santana F, Telles C, Lopez S, 37. Srinivasan TN, Thara R. How do men with schizophrenia fare at
Mintz J. Expressed emotion and schizophrenic outcome among work? A follow-up study from India. Schizophrenia Research
Mexican-American families. J Nerv Ment Dis 1987 1997; 25, 149-154.
Mar;175(3):143-51. 38. Lund C, Flisher AJ. Staff/population ratios in South African public
25. Kopelowicz A, Zarate R, Gonzalez V, Lopez SR, Ortega P, Obregon sector mental health services. S Afr Med J 2002; 92(2);161-164.
N, et al. Evaluation of expressed emotion in schizophrenia: a 39. McFarlane WR, McNary S, Dixon L, Hornby H, Cimett E.
comparison of Caucasians and Mexican-Americans. Schizophr Predictors of dissemination of family psychoeducation in
Res 2002; 55(1-2), 179-86. community mental health centers in Maine and Illinois.
26. Rosenfarb IS, Bellack AS, Aziz N. Family interactions and the Psychiatric Services 2001; 52, 935-42.
course of schizophrenia in African American and white patients. 40. Koss-Chioino JD, Vargas LA. Through the cultural looking glass: a
Journal of Abnormal Psychology 2006; 115(1), 112-120. model for understanding culturally responsive psychotherapies,
27. Rosenfarb IS, Bellack AS, Aziz N. A sociocultural stress, appraisal, in Working With Culture: Psychotherapeutic Interventions With
and coping model of subjective and family attitudes toward Ethnic Minority Children and Adolescents (Vargas LA,Koss-
patients with schizophrenia. Journal of Abnormal Psychology Chioino JD ). Jossey-Bass, San Francisco, 1992:1-22.
2006; 115(1), 157-165. 41. Kleinman A. (1988). Rethinking psychiatry: From cultural
28. Suzuki LA, Ponterotto JG, Meller PJ. Handbook of multicultural category to personal experience. New York: Free Press.

Open access to
African Journal of
Psychiatry website

www.ajop.co.za
African Journal of Psychiatry • November 2011 371