Traditional healers’ and biomedical practitioners’ perceptions of collaborative mental healthcare in low- and middle-income countries: A systematic review

Abstract

Access to mental healthcare in low- and middle-income countries (LMICs) is one of the greatest challenges in public health today. One suggestion for improving accessibility is through collaboration between biomedical practitioners and traditional healers. This paper reviews studies of traditional healers’ and biomedical practitioners’ perceptions of collaboration. We conducted a systematic review of online databases, selected journals, and reference lists for relevant studies. Eligible papers were assessed using a tool designed for this review for quality and study characteristics, and qualitative data demonstrating participants’ views were extracted. A total of 14 papers from seven countries were included. The published literature on this topic is relatively homogenous and studies are of variable quality. The findings suggest that, despite differing conceptualisations of mental illness causation, both traditional healers and biomedical practitioners recognise that patients can benefit from a combination of both practices and demonstrate a clear willingness to work together. There are concerns about patients’ safety and human rights regarding traditional methods and some healers are sceptical about the effectiveness of Western psychiatric medication. Despite keeping the inclusion criteria open to all LMICs, 13 of the studies were conducted in Africa, seven of which were in South Africa. This limits the applicability of the findings of this review to the wider LMIC context. The paper concludes with recommendations for research and practice.

Keywords

collaboration community care low- and middle-income countries mental health traditional healer

Introduction

According to the World Health Organisation (WHO), access to mental healthcare in low- and middle-income countries (LMICs) is one of the greatest challenges in public health today, with almost 90% of people with severe mental disorders receiving no treatment (WHO, 2015, p.8). Barriers to care are wide-ranging, including a lack of funding and of trained medical professionals, the centralisation of services in urban areas to the detriment of rural areas, and a lack of prioritisation of mental health in public health leadership (Saraceno, 2007). Despite the fact that psychiatric illness represents a monumental burden on the economy and compounds human suffering, financial and professional resources for therapeutic support are lacking, prompting the need to find innovative ways of reducing this treatment gap (Kohn, Saxena, Levav & Saracen, 2004). A landmark series of articles in The Lancet in 2007 called for mental health services to be ‘scaled up’ across the so-called ‘developing world’, spawning a flurry of research and policy bodies that would attempt to improve access to care. However, these initiatives have been criticised for being rooted in Eurocentric assumptions about illness, with some arguing that they will be ineffective unless steps are taken simultaneously to address people’s help-seeking behaviour (Cooper, 2016). This view has been echoed by leading academics in the field, who have urged policymakers to ‘think outside the box’ in regards to mental health care (Patel, 2012a). In an attempt to develop services within local frameworks, the WHO has suggested working more closely with ‘informal’ healthcare providers, including traditional healers, to widen access to care (WHO, 2015).

The relationship between biomedical and indigenous forms of care has historically been tense, characterised by a belief that the two systems would not be able to work together due to their diametrically different understandings of mental illness (Calabrese, 2013). It has also been widely believed that patients’ use of traditional healers simply reflected the lack of an alternative. However, recent research shows that they are used even when psychiatric facilities and medication are readily available and it is now generally accepted that the appeal of healers lies in their ability to understand patients’ illness experience within their cultural framework (Read, 2012; Burns, 2015; Heaton, 2013). This is particularly true of mental illness, where social and cultural determinants play a heightened role. As widespread usage of traditional healers shows no sign of abating, it seems logical to streamline the efforts of formal and informal providers to create a mental healthcare model that represents patient help-seeking preferences. Furthermore, aligning professional and lay narratives of illness can improve positive clinical outcomes, and thus, tailoring the healthcare system to users’ demands should improve patient care (Lakes, 2006; The Health Foundation, 2014). This was also shown in the ethnographic documentary ‘Breaking the chains’ (Colucci, 2015, 2016).

Traditional healers have been successfully integrated into HIV care as educators and counsellors since the 1990s (UNAIDS, 2006, p. 10). In mental healthcare, a growing body of research in LMICs demonstrates that lay people or community health workers can be trained to deliver psychosocial interventions for people with a range of mental disorders (Thornicroft, Ruggeri & Goldberg, 2013). Commonly known as ‘task sharing’, it is a popular method for addressing shortages of specialist health resources and its evidence base is consistently positive (Patel, 2012b, p. 8). However, research and best practice guidance for working specifically with traditional and faith-based healers is scarce.

The WHO (2012) proposes four approaches to adopting traditional practices into primary care:

1. Tolerant: a select group of traditional practitioners are allowed to practice in some capacity;

2. Inclusive: traditional healers are recognised but not integrated into policy, regulation, or education;

3. Parallel: both traditional and conventional forms of healing are practiced simultaneously in the healthcare system;

4. Integrated: the healthcare and medical education system incorporates both traditional and conventional practices.

In addition, the following principles have been highlighted by several studies as crucial in establishing a collaborative relationship: education, trust, mutual understanding, cross-referral, and shared working spaces (Osafo, 2016; UNAIDS, 2006; Pretorius, 1991).

Whilst useful, the above framework illustrates the contentious nature of the conversation around traditional healing. Some argue that the idea of ‘allowing’ traditional healers to practice in their native country is steeped in complex colonial rhetoric. Similar criticism has been addressed to the body of global mental health research that emphasises ‘scaling up’ services or ‘increasing mental health literacy’, a concept that generally translates as being ‘literate in a medico-scientific system of knowledge’ (Cooper, 2016, p. 710). On the other hand, romanticising traditional healing and delegitimising of psychiatric treatments, many of which have significant benefits for those struggling with mental illness, is problematic. This paradox was summed up by the WHO, which stated that the body of work surrounding traditional healing is underlined by “uncritical enthusiasm or uninformed scepticism” (2002).

A more nuanced understanding of the role that traditional healers play in their respective societies is needed. Recent research has shown that there may be “subjective benefits” that patients reap from seeing a traditional healer that may not be quantifiable in a Western framework; this must be explored in more depth (Nortje, Bibilola, Gureje, Seedat, 2016). The aim of this paper therefore is to systematically review the qualitative literature regarding the perceptions of traditional healers and healthcare professionals towards collaborative mental healthcare. Qualitative studies have been chosen as they are most effective in exploring the subjective views and perceptions of a particular topic. This will hopefully increase our understanding of how collaboration could be implemented in practice.

Methods

Search strategy and definitions

The following databases were searched in August 2017: Pubmed, PSYCHinfo, Web of Science, Scopus and PsycEXTRA. Five journals deemed most relevant to the subject matter were searched: Transcultural Psychiatry; International Journal of Culture and Mental Health; Culture, Medicine and Psychiatry; Medical Anthropology; and International Journal of Social Psychiatry. Reference lists were searched manually.

The search used an operational definition of traditional healers, developed by Nortje et al., as those “who explicitly appeal to spiritual, magical or religious explanations for disease and distress” (Nortje, et al., 2016, p.155). The focus on magico-religious healers was due to the widespread belief in LMICs that mental illness is caused by spiritual/magical factors, and to a large amount of published literature that highlights that these healers tend to be specifically sought out for mental complaints (Robertson, 2006; Nortje, et al., 2016; Abbo, 2011). Further, the authors are particularly interested in how collaboration can prevent the unethical practices associated with faith/traditional healing, which are not widely associated with methods that rely on physical or humoral explanations, such as Ayurveda. Throughout this paper the term ‘traditional healers’ will be used to also include faith-based and indigenous healers.

Following Higgins and Thomas, 2019, multiple terms for each element of interest were entered using Boolean operators . Through a process of trial and error, the following search string was employed: (“traditional healer” OR “spiritual healer” OR “religious healer” OR diviner OR shaman OR “traditional practitioner”) AND (“healthcare professional” OR “healthcare worker” OR doctor OR psychiatrist OR nurse OR psychotherapist) AND (“mental health” OR “mental disorder” OR “mental illness” OR “mental health services” OR “mental healthcare”). Neither the concept of ‘collaboration’ nor ‘perspectives’ was included in the initial search string because it was determined that their inclusion could lead to exclusion of relevant studies.

Whilst interest in the potential of collaborating with traditional healers began in the 1950s, recognised most notably by T.A. Lambo’s ‘village psychiatry’ model, current review focussed on research conducted after 1970, because this is the point at which collaboration entered mainstream public health discourse (Jegede, 1981). The Alma-Ata Declaration saw the WHO formally acknowledge the role of traditional practitioners in primary healthcare for the first time (1978). Whilst publications on the subject matter had been produced in earlier decades, these were largely of an anthropological or observational nature, rather than relevant qualitative studies that actively explored practitioners’ views towards collaboration.

Study inclusion and exclusion criteria

All studies that met the following criteria were included: (1) reported in the English language; (2) related to the treatment of psychological disorders; (3) explored the use of both traditional and biomedical mental health practices; (4) explored the perceptions of either traditional healers or biomedical practitioners; (5) used qualitative methods; and (6) reported data from LMICs. Exclusion criteria were: (1) published before 1970; and (2) did not report primary research.

Data extraction and methodological quality

The full text of each study identified as potentially eligible was examined to ensure that it fulfilled the inclusion criteria. Key study characteristics were extracted from the eligible studies and collected in a table that included information on author name, country, study type, study aim, recruitment, data collection, and sample used. For study findings, this review followed Thomas and Harden’s approach to data extraction, focusing on “all text labelled as ‘results’ or ‘findings’ in study reports” (Thomas & Harden, 2008, p. 4). These sections bore the closest resemblance to the raw data used for the original studies and, where possible, direct quotations were used.

To support comparative analysis, a table was designed that recorded the data of all the studies under identified codes. Visual diagrams were also used to explore the coded data and establish overarching themes. Narrative synthesis was therefore employed as the most effective way of analysing the qualitative material (Popay, 2006, p. 1). This approach allowed the authors to synthesise and analyse the existing material, going beyond the summaries of findings in the individual studies to generate new insights. Overall, the process of extraction and synthesis was an iterative process, which involved moving between the original studies, the code tables, and the mapping diagrams until it was felt that the themes were representative of the concepts raised in the studies. The quality of the included studies was assessed through the use of the adapted version of the Critical Appraisal Skills Programme (CASP) quality assessment checklist for qualitative studies (2014), which comprised ten elements: aims; methodology; research design; recruitment method; data collection; relationship between researcher and participants; ethical issues; data analysis; findings; and value of research.

Results

Literature search

The online search strategy produced 1,527 articles in August 2016, with seven articles identified through other sources. There were 1,312 articles once duplicates were removed. This was reduced to 37 full-text articles that were assessed for eligibility.

Figure 1 illustrates the screening process for eligible studies.

Figure 1.

PRISMA diagram.

Eligible papers

Fourteen studies were eligible. They were published between 1999 and 2016, and included 393 participants across 12 of the papers. Two papers did not offer sample sizes. Participants included traditional healers, faith-based healers, psychiatrists, nurses, healthcare professionals, or healthcare facility staff. Details on age or gender were rarely provided, making it impossible to draw any conclusions about whether either factor altered participants’ perceptions of collaboration. Despite inclusion criteria covering all LMICs, 13 of the studies were conducted in Africa and seven of those were in South Africa. Only one paper, from India, looked at a successful example of formal collaboration; all others explored participants’ views of organic collaboration.

The limited geographic spread of studies highlights not only the lack of research regarding collaboration with traditional healers but also the concentration of research in the region where traditional healers have the most recognition (particularly in South Africa, where they are formally regulated).

All studies used qualitative or mixed methods to collect data, with the majority employing structured/unstructured interview techniques and/or focus groups. The key characteristics of each study are reported in Table 1.

Table 1.

Study characteristics.

Author name (date)	Ae-Ngibise et al. (2010)	Agara, et al. (2007)	Bulbulia & Laher (2013)	Campbell-Hall & Peterson (2010)	Hopa,et al. (1998)	Janse Van Rensburg, et al. (2014)	Kayombo, et al. (2012)
Sample	122 policy makers, health professionals, users of psychiatric services, teachers, police officers, academics, and religious (Christian and Islamic) and traditional healers. Individual details not reported	Traditional healer (n = 15) Church elder (n = 1) Prophet (n = 1) Evangelist/healer (n = 1)	Muslim psychiatrists (N = 7)	Stakeholders from formal health sector; Stakeholders from the NGO sector; Traditional practitioners. Sample size not given	Psychiatrist (n = 2) Medical doctor (n = 7); Traditional healer (n = 6); Psychologist (n = 8)	Psychiatrist (N = 13)	Health facility managers (N = 33)
Data collection	Semi-structured interviews and focus groups	Focus group discussions/key informant interview	Semi-structured interview	Semi-structured interview/focus group discussion	Focus group interview with topic guide	In-depth, semi-structured interview	Face-to-face in-depth interview
Recruitment	Data gathered as part of a situational analysis of the status of mental health policy, legislation. and services in Ghana, conducted as part of the first phase of the Mental Health and Poverty Project. Participants purposively sampled from among the major stakeholders in mental health at the national, regional, and district levels	Convenience sample of local parish. Purposive sample of key informants	Convenience sample	Purposive sample of key stakeholders	Krueger’s list strategy/piggy-back focus group/snowball sample	Purposive sample	Purposive sample
Study aim	1) To explore the reasons underpinning the widespread appeal of traditional healers. 2) To identify what barriers or enabling factors may exist for forming bi-sectoral partnerships	To examine the knowledge, attitude, and practice of mental health care among a syncretic Church’s healers, and their readiness to cooperate with psychiatrists	To explore perceptions of mental illness, particularly in terms of the role of Islam in the understanding of mental illness among South African Muslim psychiatrists practicing in Johannesburg	To explore perceptions of service users and providers of current interactions between the two systems of care and ways in which collaboration could be improved in the provision of community mental health services	To investigate the perceptions of various stakeholders (viz. psychiatrists, medical doctors, psychologists, traditional healers, and consumers) on integration of the traditional and Western healing systems in South Africa	To capture the views of some local psychiatrists on referral and collaboration between South African psychiatrists and religious or spiritual advisors	To get an experience of health care utilization from both urban and rural areas of seven administrative regions in Tanzania
Study type	Qualitative	Mixed methods	Qualitative	Qualitative	Qualitative	Qualitative	Qualitative
Country	Ghana	Nigeria	South Africa	South Africa	South Africa	South Africa	Tanzania
Quality assessment	7	7	6	8	7	7	5
Author name (date)	Keikelame & Swartz (2015)	Kahn & Kelly (2001)	Musyimi, et al. (2016)	Ovuga, et al.(1999)	Shields et al. (2016)	Sorsdahl & Stein (2010)	Teuton, et al. (2007)
Sample	Xhosa-speaking traditional healers (N = 15)	Xhosa nurses (N = 77)	Traditional healers, faith healers and clinicians (registered nurses and clinical officers). Sample size not given	Traditional healers (n = 29)	Key stakeholders instrumental in the set up and maintenance of the programme; Faith-based healers (n = 3); Allopathic mental health practitioners (n = 3)	Diviners (n = 21); Herbalist (n = 1); Both diviner and herbalist (n = 2). All traditional healers were members of the Traditional Healers Organisation	Baganda indigenous healers (n = 10); Religious healers based on Christianity and Islam (n = 10); Psychiatrists (n = 2); Qualified psychiatric clinical officer (n = 3); Trainee psychiatric clinical officer (n = 1)
Data collection	Individual in-depth interview/focus group discussion using semi-structured interview guide	Questionnaire (including open-ended questions)	Focus group discussion	Semi-structured interview	Semi-structured interview	Focus groups with an open-ended semi-structured interview in Xhosa	Informal/unstructured interview with healers. Structured interview using EMIC format
Recruitment	Purposive sample. Access to these healers was gained through assistance from an executive committee member of Traditional Healers Organization	Questionnaires disseminated to all Xhosa nurses at Fort England Psychiatric Hospital, Grahamstown	Not clear	The healers were identified for interview with the assistance of the Chairman of Traditional Healers. 51 traditional healers were identified of whom 29 were randomly selected for interview	Purposive sample	Convenience sample selected from the South African Depression & Anxiety Group, a mental health advocacy group that runs workshops for traditional healers to increase mental health literacy	Snowball/ purposive sample
Study aim	To explore traditional healers’ perspectives on epilepsy and collaboration with biomedical healthcare in Cape Town	To examine the views of a sample of Xhosa-speaking psychiatric nurses on traditional healing and its role in mental health care in South Africa	To form dialogue and establish collaboration among the informal (faith and traditional healers) and formal health workers (clinicians) in enhancing community-based mental health in rural Kenya	To carry out a survey of the beliefs, knowledge, and practice towards mental illness of 29 traditional healers in the Pallisa district of Uganda	To explore the origins, use, and outcomes of a collaborative program between faith-based and allopathic mental health practitioners in India	To use the constructs of the Theory of Planned Behaviour to gain an understanding of traditional healer referral practices of their patients with a mental illness	To analyse the discourse of respondents from different parts of the healing system to establish how they view each other
Study type	Qualitative	Mixed methods (qualitative and quantitative)	Qualitative	Qualitative	Mixed methods case study (qualitative and quantitative)	Qualitative	Qualitative
Country	South Africa	South Africa	Kenya	Uganda	India	South Africa	Uganda
Quality assessment	7	7	6	4	7	7	6

Quality of papers

In line with the CASP guidelines, the first two questions were used to screen studies’ quality, resulting in rejection if they did not receive a ‘yes’ answer to both items (2014). Scales indicating high or low quality were not used; however, the number of ‘yes’ answers indicates the extent to which studies met crucial criteria. All studies received a ‘yes’ answer to the screening questions. On the subsequent eight questions, studies were of mixed quality, with the lowest and highest receiving four and eight ‘yes’ answers, respectively.

Themes

Thematic analysis identified three dominant themes, and six sub-themes, as follows: Recognition of the potential of healers in mental healthcare provision (perception of cultural acceptability; effectiveness in providing psychosocial support); Strategies for collaboration (co-management of patients; capacity building and exposure to one another’s practice); and Perceived barriers to collaboration (perceived illegitimacy of one another’s practice; lack of regulation in traditional healing)

Recognition of the potential of healers in mental healthcare provision

All studies recognized that traditional healers are widely sought out and play a key role in the provision of mental healthcare. Their popularity was attributed to two key factors: a shared cultural understanding with patients of the spiritual cause of mental illness and their ability to provide effective psychosocial support for less severe mental disorders.

Perception of cultural acceptability

It was consistently reported that the popularity of traditional healers was rooted in the fact that their approach to mental illness is “firmly embedded within wider belief systems and are synchronous with dominant constructions of health and illness” (Ae-Ngibise, et l., 2010, p. 561). Biomedical practitioners expressed the belief that healers are sought out because they are from a “more culturally familiar psychiatric milieu” than biomedical professionals (Kahn & Kelly, 2001, p. 42) and can therefore comprehend the meaning of mental illness from a perspective similar to that of the patient. For some healthcare professionals, there was a sense of reluctant acquiescence. Whilst they may not subscribe to the same worldview as healers, they understood why they were sought out. Given the widespread and continued use of traditional healers, some doctors felt that they may as well work together, as explained by this doctor: “Traditional healers have been part of our societies for a very long time and whether we like it or not people with mental problems are going to go to them” (Ae-Ngibise, et al., 2010, p.560). However, some biomedical practitioners seemed genuinely enthusiastic to work with healers because of their position as “gatekeepers of care” in the local community (Shields, et al., 2016, p. 10).

Effectiveness in providing psychosocial support

It was recognised by some healthcare professionals that whilst healers are not able to cure severe mental health issues, they may be effective in providing psychosocial support to patients suffering from less critical disorders (Ae-Ngibise et al., 2010; Campbell-Hall, 2010; Kahn & Kelly, 2001; Teuton, Dowrick & Bentall, 2007). For instance, a psychiatric staff member in the study by Teuton et al. (2007) reports:

[I]f you get an illness like either hysteria or these neurotic illnesses […] in the process of praying, they might be doing psychotherapy without knowing they are doing psychotherapy and the patient gets cured. But not with bipolar, because bipolar is a major psychotic illness which cannot go with that (Teuton, et al., 2007, p. 1268).

This notion was also reflected in healers’ statements, in which they claimed they could provide “words of comfort” to patients (Campbell-Hall, 2010, p. 618). Interestingly, descriptions of their therapeutic strategy were strikingly similar to Western models of talking therapy, in which exposure to, and discussion of, a distressing experience is encouraged to avoid traumatisation. These descriptions suggest that healers perceive themselves as effective counsellors.

Strategies for collaboration

Both biomedical and traditional practitioners suggested strategies to improve the likelihood of successful collaboration, with capacity building and the establishment of a referral system to co-manage patients emerging as the strongest factors (Agara, Makanjuola & Morakinyo, 2008; Jansen Van Rensburg, Poggenpoel, Szabo & Myburgh, 2014; Kayombo, Uiso, Mbwambo, Mahunnah, Moshi & Mgonda, 2007; Kahn & Kelly, 2001; Musyimi, Mutiso, Nandoya & Ndetei, 2016; Shields et al., 2016). These were largely in line with existing research and guidance on the topic.

Co-management of patients

In regards to what form collaboration would take, a system of co-referral seemed preferable (Bulbulia & Laher, 2013; Campbell-Hall, 2010; Janse van Rensburg, et al., 2014; Kahn & Kelly, 2001; Musyimi, et al., 2016; Ovuga, 1999; Shields, et al., 2016; Teuton, et al., 2007). Healers tended to want a more integrated system, suggesting that access to rooms in health centres and hospitals would facilitate collaboration (Ovuga, 1999, p. 278). Biomedical participants did not seem keen to share working spaces with healers, with one study stating that establishing joint clinics was rated lowest priority by conventional biomedical practitioners (Kayombo, et al., 2007). Another stated that healers should only visit hospitals to receive health education and orientation, but not to treat patients, “as it would confuse them” (Kahn and Kelly, 2001, p. 44). There was perhaps a fear that working with traditional healers would in some way grant indigenous practices medical legitimacy, which could reflect badly on their own profession (Ae-Ngibise, et al., 2010). Similarly, it is possible that healers wanted to work closely with biomedical providers in order to enhance the legitimacy of their own work, rather than because it would lead to greater co-working.

Although most participants expressed willingness to collaborate in some form, there seemed to be very few formal structures in place (Campbell-Hall, 2010; Musyimi, et al., 2016; Janse van Rensburg, et al., 2014). Reports from traditional healers imply that instances of referring patients were one-sided (Campbell-Hall, 2010; Kayombo, et al., 2007; Keikelame & Swartz, 2015 One psychiatrist stated that “we rarely refer patients to faith healers or traditional healers. The patients can choose to go, but we do not initiate the referral conversation” (Musyimi, et al., 2016, p. 5).

Interestingly, the case study of successful collaboration in India opposes the view that referral should be formalised, demonstrating that encouraging it as an option better reinforced the idea of mutual respect. As one doctor described, “we train faith-based healers how to identify, who to refer, when to refer, and made clear that it is not a compulsion for them to refer. It is their own free will to refer a person” (Shields, et al., 2016, p. 376).

Capacity building and exposure to one another’s practice

Recognising that traditional healers are often sought out as the first point of care, three studies explicitly stated that training should focus on teaching them how to recognise the signs and symptoms of mental illness in order to reduce the delay in getting psychiatric care and function as an “early detection system” (Campbell-Hall, 2010; Kahn & Kelly, 2001, p. 45; Kayombo, et al., 2007; Shields, et al., 2016).

Healers expressed interest in training to improve their patient management skills, indicating that they would be “willing to attend workshops organised by government” (Agara, et al., 2008, p. 118). Some suggested that they would like training to understand “how Western medicine explains and deals with mental health problems” (Campbell-Hall, 2010, p. 619). This idea of improving healers’ understanding of psychiatry was supported by biomedical practitioners, who suggested that an “understanding of psychiatry and the benefits of psychopharmacology should be promoted to healers” (Jansen Van Rensburg, et al., 2014, p. 43). Biomedical practitioners also wished to learn about traditional medicine and undergo training that would “sensitise” them to working with traditional healers (Kayombo, et al., 2007, p. 6).

Study participants expressed that healers should be educated to recognise the symptoms of mental illness and to distinguish “what they could treat […] and when to refer to the Western based health care system” (Campbell-Hall, 2010, p. 619).

The acute stage of illness, in which biomedical practitioners thought it necessary to have Western medical intervention, was often identified by a patient’s aggression.

We tell them [traditional and faith healers] that there are conditions, especially the acute phase, where the person may be very restless or aggressive, and they should know that is not their area. […] After the person has settled, we tell them that the person can go to them where they can take care of the spiritual side (Ae-Ngibise, et al., 2010, p. 654)

Interestingly, this was the same for some healers, who claimed that “violence and aggressive behaviour by patients warranted referral to psychiatrists,” and referenced the Western doctor’s injection that “pacifies even a violent mentally ill person” (Agara, et al., 2008, p. 118; Sorsdahl & Stein, 2010, p. 599).

It is interesting that healers recognised the limits of their own treatment. The recognition that biomedical intervention is needed somewhat contradicts their spiritual health beliefs, suggesting that they are more flexible than previously thought.

Perceived barriers to collaboration

Perceived illegitimacy of one another’s practice

Both biomedical and traditional practitioners expressed scepticism regarding the effectiveness and legitimacy of one another’s treatment of mental disorders (Ae-Ngibise, et al., 2010; Agara, et al., 2008; Campbell-Hall, 2010; Keikelame & Swartz, 2015; Khan & Kelly, 2001; Shields, et al., 2016; Sorsdahl & Stein, 2010; Teuton, et al., 2007). Biomedical concerns were primarily rooted in patient safety and human rights (Ae-Ngibise et al., 2010; Campbell-Hall, 2010; Janse van Rensburg, et al., 2014; Kayombo, Uiso & Mahunnah, 2012; Sorsdahl & Stein, 2010; Teuton, et al., 2007). Biomedical practitioners frequently expressed concerns that traditional healers’ conduct fell short of ethical or professional standards (Ae-Ngibise, et al., 2010; Kayombo, Uiso & Mahunnah, 2012; Khan & Kelly, 2001; Teuton, et al., 2007). Traditional healers equally doubted the effectiveness of psychiatric medication (Sorsdahl & Stein, 2010).

Whilst traditional healers acknowledged the successes of biomedical medication for physical illnesses, namely HIV/AIDs and tuberculosis, they were sceptical about the effectiveness of psychiatric treatment (Ae-Ngibise et al., 2010; Teuton, et al., 2007). They frequently expressed the idea of a ‘band-aid’ treatment that could only treat the symptoms, rather than the root cause, of mental illness. As one healer put it, “Western doctors cannot cure a mental illness. They only help some symptoms” (Sorsdahl & Stein, 2010, p. 600).

Lack of regulation in traditional healing

Both traditional and biomedical practitioners suggested that the lack of regulation was a barrier to effective collaboration (Campbell-Hall, 2010; Hopa, Simbayi & du Toit, 1998; Kayombo, et al., 2007; Keikelame & Swartz, 2015; Khan & Kelly, 2001). Reference was also made to regulating traditional medicine, which was seen by doctors as lacking scientific validity (Kayombo, et al., 2007). Healers expressed frustration that their knowledge was not respected and some presented regulation as a means of professionalising healers so that they could be recognised as ‘legitimate partners’ in healthcare provision (Campbell-Hall, 2010, p. 621). One participant suggested that healers should be registered with a controlling body, either locally or centrally (Hopa, et al., 1998).

Examples of criticisms of the traditional healing system are that it was “not well developed” and that there was an “absence of clear guidelines” (Ae-Ngibise, et al., 2010, p. 563). Whilst some doctors accepted that traditional healers had a role in the healthcare system, they felt that their practice should be subject to the same standards as psychiatric medicine. Some suggested that legislative measures should be taken to regulate traditional medicine, and that a pricing system would help to reduce the chance of financial exploitation (Kayombo, et al., 2007).

Discussion

This review aimed to explore the perceptions of traditional healers and biomedical practitioners towards collaborative mental healthcare. Whilst it was expected that their differing views of mental illness would serve as a barrier to collaboration, this did not stand out as an obstacle. Holding different beliefs about the cause of mental illness did not impact on their apparent willingness to work together. There was recognition by both parties that traditional and biomedical treatment could complement one another. It is obvious, however, that there are very limited structures in place to facilitate co-working. The findings imply that three key factors should be the focus of future research and policy: exploring how an effective system that allows for the shared management of patients could be built; regulating and increasing the evidence base of traditional healing; and developing trust between, and capacity of, all practitioners.

Facilitating shared management of patients

In line with existing literature, these studies demonstrated that traditional healers are widely used in the countries studied (Robertson, 2006; Tilburt & Kaptchuk, 2008). However, traditional healing is not integrated into healthcare policy, regulation, or education, leading to widespread concerns about its unregulated nature (Ae-Ngibise, et al., 2010; Kayombo, Uiso & Mahunnah, 2012; Khan & Kelly, 2001; Teuton, et al., 2007). According to the WHO framework cited in the introduction, current practice therefore seems to adopt a tolerant model in the countries studied (2012). Given that biomedical practitioners require regulation and formalisation of traditional healing practices, this review suggests that an integrated system would be preferable (see introduction for WHO framework). This would allow for traditional healers to be recruited into the healthcare system and provided with the training needed to acquire the skills to identify and manage mental illness in the community. However, this would also lead to a more equal status within the healthcare system and further research should be carried out to explore whether this would cause resistance from biomedical practitioners.

To establish an effective referral system would require agreeing upon referral criteria and designing referral processes. Similar to the multidisciplinary model frequently utilised in modern Western healthcare systems, this collaborative approach will require open dialogue, effective communication, and a mutual appreciation for one another’s practice. This will not be without its challenges. For example, if referral forms are to be used, the use of paperwork may be an alien concept to healers who are accustomed to an oral tradition. The use of written documentation would also require a basic level of literacy that may exclude some healers. This is particularly significant given the concerns expressed by some doctors regarding the lack of literacy amongst healers (Khan & Kelly, 2001).

The willingness to collaboratively manage patients based on the severity of their illness complements recent research that demonstrated that traditional healers could provide an effective psychosocial intervention for common mental disorders, such as depression and anxiety, but that there is “little evidence to suggest that they change the course of severe mental illnesses” (Nortje, et al., 2016, p. 154). That healers have expressed an awareness of the behaviours that require biomedical intervention demonstrates that there is flexibility in their health beliefs and that there is room to develop a framework in which care could be divided, possibly with healers delivering psychosocial interventions and psychiatric treatment being offered for severe mental illness.

By carving out an area in which each practitioner feels they were the ‘specialist’, participants seemed to feel more comfortable relinquishing an element of care to the alternative practice. Significantly, this would allow both kinds of practitioner to retain a sense of autonomy and a proactive role within the healthcare system. This finding is congruent with recent research by Bantjes, who states that for healers working “in the context of global inequality and the dominance of bio-medicine, the protection of spheres of the indigenous becomes very important” (Bantjes, Schwarts & Sithembile, 2018, p. 85).

Regulation and evidence base of traditional healing

The willingness of biomedical practitioners to collaborate hinges on a requirement for evidence-based practice, which whilst understandable, can be problematic. Recent research, such as Calabrese’s exploration of the use of the psychedelic cactus Peyote as a form of “postcolonial healing” in America, demonstrates that traditional forms of therapy can be highly effective when they are embedded and delivered within a unique cultural or socio-historical form of experience (2013). That said, we may not see the same therapeutic benefits if we subject an indigenous intervention to Western standards of analysis, such as randomised control trials. We therefore may need to explore alternative ways of evaluating the effectiveness of indigenous treatments for mental illness.

A recent systematic review by Gareth Nortje et al. (2016) has attempted to develop this field by assessing the quantitative outcomes of traditional healing practices globally. Similarly, organisations such as the Association of the Promotion of Traditional Medicine in Senegal is dedicated to quantitatively measuring the effectiveness of traditional medicine, striving to “reduce health workers’ scepticism and strengthen mutual appreciation, understanding and respect between practitioners of the two health systems of medicine” (Busia & Kasilo, 2010). Research continues to explore traditional healers’ understanding of mental illness and suicidal behaviour, yet the effectiveness of their prevention strategies remains largely untested (Bantjes, et al., 2018). Recent research has highlighted the huge diversity in traditional treatment methods; further research must explore the effectiveness and quality of these treatments before they can be recommended or formally integrated (Kpobi & Swartz, 2018).

Abusive practices within healing centres – as documented, for instance, in Human Rights Watch reports and the ethnographic documentary research Breaking the Chains – need to be investigated and discouraged (Colucci, 2015, 2016). Chaining, beating, and other such abusive practices are an infringement on a patient’s human rights and cause severe suffering and the development of further mental and physical disabilities. However, the widespread view of healers as charlatans is not representative of the entire sector and the widespread dissemination of that view may hamper efforts to bring healers into the fold of mainstream care. Legislation that has deliberately been enacted to prevent malpractice, such as South Africa’s Witchcraft Suppression Act, has been met with widespread opposition from the Traditional Healers Organisation due to the belief that it unconstitutionally suppresses religious or cultural beliefs and also that criminal acts should be dealt with under existing human rights legislation (South African Law Reform Commission, 2016).

Preventing human rights abuses will be one of the most challenging areas to tackle and recent evidence has thrown open its complexities. Chaining has commonly been perceived purely as a method of restraint, or as a result of a lack of medication and safe accommodation. However, a recent RCT in Ghana, which evaluated combined psychiatric and prayer camp care against standard prayer camp care, found that despite a significant reduction in symptoms, there was no significant difference in days in chains (Offori-Atta, 2018). This is surprising and shows that the methods such as chaining and beating are more closely interlinked to the spiritual ideology of mental illness and to concepts of punishment rather than to care. In order to alleviate human suffering but also build stronger ties with traditional practitioners, those methods associated with traditional healing that have shown to be beneficial, such as counselling, should be actively encouraged in the place of beating or forced restraint.

Capacity building

A collaborative approach should serve to improve accessibility to acute care and ultimately bring cost benefits. Delays in access to care are associated with longer hospital stays, poorer health outcomes, and higher costs for both the patient and the healthcare system (Weissman, Weissman, Stern, Fielding, & Epstein, 1999; Kraft et al., 2009). Furthermore, the socioeconomic cost of long-term mental health issues to both patient and society should not be underestimated. Early intervention is therefore essential in creating a cost-effective system; both in terms of monetary and individual value. Integrating healers, so frequently sought out as the first point of contact by patients and/or their carers, and enabling them to recognise the signs of acute mental illness or suicide that require immediate support, could serve to reduce delays in accessing acute interventions. In the long-term, this will alleviate some of the financial burden on healthcare services in LMICs.

The findings in this review suggest that collaboration must go further than training healers to merely assist biomedical practitioners. Whilst the task sharing approach can bolster healthcare services in areas with few resources, it “co-opts [healers’] cultural acceptability in order to deliver conventional treatment, [but] makes little use of their unique skills and specific advantages” (Gujere, 2015, p. 8). Instead, healers’ localised cultural insight and status in the community should be utilised, as it has been in the Ebola epidemics (Maclean, 2016). This is supported by the documentation of a successful collaboration in Shields et al.’s (2016) paper, in which there was no attempt to impose ideological change onto healers. The programme instead focused on recognising symptoms of mental distress without requiring of healers that they subscribe to the biomedical paradigm of disease aetiology. This suggests that educative measures should focus on the acquisition of practical skills, recognising the signs and symptoms of acute mental illness, encouraging signposting to mainstream care, and recognising the points at which biomedical intervention would be beneficial.

It could be assumed that healers would not be receptive to the idea that biomedical intervention is necessary, as it would involve them recognising a limit to their spiritual powers and would appear to threaten the entire rationale for their practice. However, the fact that healers identified aggressive patients as requiring psychiatric assistance demonstrates that the ideology underlying their practice, as with most spiritual ideologies, may be more fluid than previously thought. This nuance in beliefs about health is not limited to healers, but also has been seen in nurses in Nicaragua who work in a collaborative system. Regarding the importance of plurality in the healthcare system, they emphasised “ends before means, practice before theory, and the well-being of the patient before strict biomedical reasoning, [they] did not find biomedicine to be contradictory to other healing systems in their daily work” (Wedel, 2009, p. 54).

Research has also shown that resistance to biomedicine may stem from fear of lost livelihood (Shields, et al., 2016; Kayombo, et al., 2007). This is highlighted in the example from India, in which healers made the majority of referrals and the clinic ensured that patients were referred back to healers for follow-up care (Shields, et al., 2016). This gave reassurance that healers would not lose their role in the healthcare system and community.

Where this may pose difficulties, however, is with regard to psychotropic medication, because there is a tendency for healers not to believe in its effectiveness. This view was often related to the observation that such medication did not work immediately and to the belief that the causative agent had not been addressed when patients relapsed on stopping medication. This negative view of medications could lead to difficulties in collaborative management of patients, because healers’ notions of mental illness as a curable disease – rather than as a chronic disorder might lead them to encourage patients to stop taking their medication once symptoms have subsided. These concerns mirror a recent study that found that the failure of antipsychotics to achieve a permanent cure “casts doubt on their efficacy and strengthens suspicions of a spiritual illness which would resist medical treatment” (Read, 2012, p. 448). Educational measures should clearly explain the rationale behind psychotropic medication, including: dosages, the proposed mechanism of action, areas of knowledge limitations, side effect profile, and expected length of time before effects are felt. There should also be a focus on the dangers of drug interactions and polypharmacy to increase patient safety.

Implications for research and practice

As the global mental health movement continues to develop strategies that enhance the skill set of non-medical professionals to deliver effective interventions, there should be an increased focus on the contribution that traditional healers already make towards health provision in LMICs. Efforts should be directed towards increasing both biomedical and traditional practitioners’ skillset in a culturally sensitive way so they are more readily equipped to provide effective care. Task sharing models could be adapted to build on pre-existing traditional practices. Mainstream healthcare training should include guidance that sensitises providers to local expressions and treatment of mental illness.

A 1981 World Health Forum report stated that “only lip service seems to have been paid to promoting the process of integrating the traditional practitioners into the general medical services” (Ramesh & Hyma, 1981, p. 498). Despite increasing awareness and advocacy for collaboration, almost 40 years later, this sadly remains the case for mental health care. We must heed the findings gathered from working with healers in recent Ebola epidemics and HIV/AIDS care. Further research should focus on documenting existing examples of successful collaboration in mental healthcare and comparing them cross-culturally to develop best practice guidance. Due to the complex nature of indigenous healing in each country, it is unlikely that one single best practice model will be developed that can be followed and implemented across all LMICs. A recent study by Kpobi and Swartz (2018) has emphasised this need for a nuanced approach to collaboration. They clearly demonstrated that different groups of traditional healers exhibit varying perceptions of their own power and ideological proximity to biomedicine; these factors may impact on their motivation for, and understanding of, collaboration; thus traditional healers should not be conceived as one homogeneous group (Kpobi & Swartz, 2018). Local piloting of collaborative care programmes, such as that discussed in Shields et al.’s (2016) paper, must be encouraged. This review highlights that research regarding collaboration with healers is concentrated in Africa and mainly in the wealthiest part of the countries studied. Further qualitative research should be conducted across other LMICs in order to gain a fuller understanding of how collaboration is perceived internationally and in diverse religious and spiritual contexts.

Limitations

This review has a number of limitations. By restricting the search strategy to LMICs, valuable studies that explored traditional healing in the Navajo, Maori, and Aborigine populations were excluded. Secondly, most of the included studies employed purposive sampling method, with several working with an organising body to recruit healers (Agara, et al., 2008; Keikelame & Swartz, 2015; Ovuga, 1999; Sorsdahl & Stein, 2010). Whilst the latter grants access to healers who would otherwise be inaccessible, it may have led to selection bias, compromising methodological validity.

Only two studies critically examined the “role, potential bias and influence” that the researcher may have had on participants during data collection (CASP, 2014). Given that traditional healers often express concern that “their knowledge will be stolen by the West”, leading to lost livelihood and dilution of traditional practice, it is crucial that investigators acknowledge the impact that their presence could have on healers’ participation in research (Kayombo, et al., 2007, p. 8). The positive attitude towards collaboration in these studies may therefore exaggerate the extent to which healers would work with biomedical practitioners in reality. This is supported by a recent study that found that although 99% of healers reported a willingness to refer to biomedical services, only 43% were doing so in everyday practice (Peltzer, Mngqundaniso & Petros, 2006).

Conclusion

This paper demonstrates that, contrary to historical belief, biomedical and traditional healthcare systems are not entirely incompatible and that, when faced with a lack of resources, both biomedical and traditional practitioners have expressed a willingness to work together in order to provide a holistic service that reflects patient behaviour, preference, and beliefs, with a shared common goal of improving patient outcomes.

As evidenced by the widespread use of both systems, patients are already engaged in a pluralistic model. A healthcare system that formally integrates multiple modes of healing could be an effective way of addressing a large treatment gap and limited resources, bringing cost and health benefits.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Erminia Colucci

Bethany Green is a social worker, currently working in a community mental health team in East London NHS Foundation Trust. Bethany is interested in the cultural dimensions of psychological distress and how we can employ social approaches to prevent and treat mental illness at an individual and community level. In 2017, she completed an M.Sc. in Cultural and Global Perspectives in Mental Healthcare at Queen Mary University, where she wrote this paper with Dr Colucci.

Erminia Colucci is Senior lecturer at the Department of Psychology at Middlesex University London and Honorary Senior Research Fellow at the Global and Cultural Mental Health Unit, Centre for Mental Health, The University of Melbourne. She uses qualitative and arts-based/visual methods in her research, teaching, and advocacy work in Cultural and Global Mental Health and Applied Cross-Cultural Psychology. Erminia’s work has largely focused on suicide and suicide prevention, human rights abuses, domestic and gender-based violence, and traditional/faith-based healing in low- and middle-income countries and in people from immigrant and refugee backgrounds. Erminia is Chair of the International Association for Suicide Prevention Special Interest Group on Culture and Suicidal Behaviour and the World Association for Cultural Psychiatry Special Interest Group on Arts, Media and Mental Health, and the Founder of Movie-ment ().

References

Abbo

(2011). Profiles and outcome of traditional healing practices for severe mental illnesses in two districts of Eastern Uganda. Global Health Action 4(1), 1–15.

Ae-Ngibise

Cooper

Adiibokah

Akpalu

Lund

Doke

The MHAPP Research Programme Consortium (2010). ‘Whether you like it or not people with mental problems are going to go to them’: A qualitative exploration into the widespread use of traditional and faith healers in the provision of mental health care in Ghana. International Review of Psychiatry 22(6), 558–567.

Agara

A. J.

Makanjuola

A. B.

Morakinyo

(2008). Management of perceived mental health problems by spiritual healers: A Nigerian study. African Journal of Psychiatry 11(2), 113–118.

Bantjes

Schwarts

Sithembile

(2018). ‘Our lifestyle is a mix-match’: Traditional healers talk about suicide and suicide prevention in South Africa. Transcultural Psychiatry 55(1), 73–93.

Bulbulia

Laher

(2013). Exploring the role of Islam in perceptions of mental illness in a sample of Muslim psychiatrists based in Johannesburg. South African Journal of Psychology 19(2), 52–54.

Burns

J. K.

(2015). Traditional and religious healers in the pathway to care for people with mental disorders in Africa: A systematic review and meta-analysis. Social Psychiatry and Psychiatric Epidemiology 50(6), 867–877.

Busia

Kasilo

(2010). Collaboration between traditional health practitioners and conventional health practitioners: Some country experiences. African Health Monitor 13: 40–46.

Calabrese

J. D.

(2013). A different medicine: Postcolonial healing in the Native American church, New York: Oxford University Press.

Campbell-Hall

Petersen

(2010). Collaboration between traditional practitioners and primary healthcare staff in South Africa: Developing a workable partnership for community mental health services. Transcultural Psychiatry 47(4), 610–628.

10.

Colucci, E. (2015). Breaking the chains (64 min, Indonesia/UK/Australia). Accessed online: https://movie-ment.org/breakingthechains/.

11.

Colucci

(2016). Breaking the chains. Ethnographic filmmaking in mental health. The Lancet Psychiatry 3(6), 590–591.

12.

Cooper

(2016). Research on help-seeking for mental illness in Africa: Dominant approaches and possible alternatives. Transcultural Psychiatry 53(6), 696–718.

13.

Critical Appraisal Skills Programme (CASP). (2014). CASP Checklists. Accessed online: http://www.casp-uk.net/casp-tools-checklists (Accessed 2nd November 2016)..

14.

Gureje, O. Nortje, G., Makanjuola, V., Oladeji, B., Seedat, S., Jenkins, R. (2015). The role of global traditional and complementary systems of medicine in treating mental health problems. Lancet Psychiatry, 2(2): 168–177.

15.

Heaton

(2013). Black skin, white coats: Nigerian psychiatrists, decolonization, and the globalization of psychiatry, Athens, OH: Ohio University Press.

16.

Higgins, J. & Thomas, J. (2019). The Cochrane Handbook for Systematic Reviews of Interventions. Accessed online: https://training.cochrane.org/handbook/current.

17.

Hopa

Simbayi

L. C.

du Toit

C. D.

(1998). Perceptions on integration of traditional and Western healing in the new South Africa. South African Journal of Psychology 28(1), 8–14.

18.

Janse van Rensburg

A. B. R.

Poggenpoel

Szabo

C. P.

Myburgh

C.P.H.

(2014). Referral and collaboration between South African psychiatrists and religious or spiritual advisers: Views from some psychiatrists. South African Journal of Psychology 20(2), 40–45.

19.

Jegede

R. O.

(1981). Aro village system of community psychiatry in perspective. Canadian Journal of Psychiatry 26(3), 173–177.

20.

Kahn

Kelly

(2001). Cultural tensions in psychiatric nursing: Managing the interface between Western mental health care and Xhosa traditional healing in South Africa. Transcultural Psychiatry 38(1), 35–50.

21.

Kayombo

Uiso

F. C.

Mbwambo

Z. H.

Mahunnah

R. L.

Moshi

M. J.

Mgonda

Y. H.

(2007). Experience of initiating collaboration of traditional healers in managing HIV and AIDS in Tanzania. Journal of Ethnobiology and Ethnomedicine 3(6), 1–9.

22.

Kayombo

Uiso

F. C.

Mahunnah

(2012). Experience on healthcare utilisation in seven administrative regions of Tanzania. Journal of Ethnobiology and Ethnomedicine 8(5), 1–8.

23.

Kahn, M. & Kelly, K. (2001). Cultural tensions in psychiatric nursing: managing the interface between western mental health care and Xhosa traditional healing in South Africa. Transcultural Psychiatry, 38(1): 35–50.

24.

Keikelame

M. J.

Swartz

(2015). ‘A thing full of stories’: Traditional healers’ explanations of epilepsy and perspectives on collaboration with biomedical health care in Cape Town. Transcultural Psychiatry 52(5), 659–680.

25.

Kohn, R, Saxena, S., Levav, I., & Saraceno, B (2004) The treatment gap in mental health care. Bulletin of the World Health Organisation, 82.

26.

Kpobi

Swartz

(2018). Implications of healing power and positioning for collaboration between formal mental health services and traditional/alternative medicine: The case of Ghana. Global Health Action 11(1), 1445333.

27.

Kraft

A. D.

Quimbo

S. A.

Solon

Shimkhada

Florentino

Peabody

J. W.

(2009). The health and cost impact of care delay and the experimental impact of insurance on delays: Evidence from a developing country. The Journal of Pediatrics 155(2), 281–285.

28.

Lakes

Lopez

S. R.

Garro

L. C.

(2006). Cultural competence and psychotherapy: Applying anthropologically informed conceptions of culture. Psychotherapy (Chic) 43(4), 380–396.

29.

Maclean, R. (2016). Healers cure mistrust in Guinea's health system after horrors of Ebola. The Guardian [online]. Accessed online: https://www.theguardian.com/global-development/2016/sep/28/healers-cure-mistrust-guinea-health-system-after-horrors-ebola (Accessed 30 September 2018)..

30.

Musyimi

Mutiso

V. N.

Nandoya

E. S.

Ndetei

D. M.

(2016). Forming a joint dialogue among faith healers, traditional healers and formal health workers in mental health in a Kenyan setting: Towards common grounds. Journal of Ethnobiology and Ethnomedicine 12(4), 1–8.

31.

Nortje

Bibilola

Gureje

Seedat

(2016). Effectiveness of traditional healers in treating mental disorders: A systematic review. The Lancet Psychiatry 3(2), 154–170.

32.

Ofori-Atta, A., Attafuah, J., Jack. H., Baning, F., Rosenheck, R. J., 2018. Joining psychiatric care and faith healing in a prayer camp in Ghana: randomised trial. British Journal of Pyschiatry, 212(1): 34–41.

33.

Osafo

(2016). Seeking paths for collaboration between religious leaders and mental health professionals in Ghana. Pastoral Psychology 65(4), 493–508.

34.

Ovuga

(1999). Traditional healers and mental illness in Uganda. Psychiatric Bulletin 23: 276–279.

35.

Patel, V. (2012a). Mental health for all by involving all. [Online Video]. June 2012. Available online: https://www.ted.com/talks/vikram_patel_mental_health_for_all_by_involving_all/transcript?language.

36.

Patel

(2012b). Global mental health: From science to action. Harvard Review of Psychiatry 20(1), 6–12.

37.

Peltzer

Mngqundaniso

Petros

(2006). HIV/AIDS/STI/TB knowledge, beliefs and practices of traditional healers in KwaZulu-Natal, South Africa. AIDS Care 18(6), 608–613.

38.

Popay, J. (2006). Guidance on the conduct of narrative synthesis in systematic reviews: A product from the ESRC methods programme. Lancaster University. Accessed online: http://www.lancaster.ac.uk/shm/research/nssr/research/dissemination/publications.php (Accessed 13 November 2016)..

39.

Pretorius

(1991). Traditional and modern medicine working in tandem. Curationis 14(4), 10–3.

40.

Ramesh

Hyma

(1981). Traditional medicine in an Indian City. World Health Forum 2(4), 495–499.

41.

Read

(2012). ‘I want the one that will heal me completely so it won't come back again': The limits of antipsychotic medication in rural Ghana. Transcultural Psychiatry 49(3–4), 438–460.

42.

Robertson

B. A.

(2006). Does the evidence support collaboration between psychiatry and traditional healers? South African Psychiatry Review 9(2), 87–90.

43.

Saraceno

(2007). Barriers to improvement of mental health services in low income and middle-income countries. The Lancet 370(9593), 1164–1174.

44.

Shields

Chauhan

Bakre

Hamlai

Lynch

Bunders

(2016). How can mental health and faith-based practitioners work together? A case study of collaborative mental health in Gujarat, India. Transcultural Psychiatry 53(3), 368–391.

45.

South African Law Reform Commission. (2016). The Review of the Witchcraft Suppression Act 3 of 1957. Available online: http://www.justice.gov.za/salrc/dpapers.htm (Accessed 8th August 2017)..

46.

Sorsdahl

Stein

(2010). Traditional healer attitudes and beliefs regarding referral of the mentally ill to Western doctors in South Africa. Transcultural Psychiatry 47(4), 591–609.

47.

Thornicroft

Ruggeri

Goldberg

(2013). Improving mental healthcare: The global challenge, Oxford: Wiley-Blackwell.

48.

Teuton

Dowrick

Bentall

(2007). How healers manage the pluralistic healing context: The perspective of indigenous, religious and allopathic healers in relation to psychosis in Uganda. Social Science and Medicine 65(6), 1260–1273.

49.

The Health Foundation. (2014). Person-centred care made simple. Accessed online: http://www.health.org.uk/sites/health/files/PersonCentredCareMadeSimple.pdf (Accessed 21st November 2016)..

50.

Thomas, J., Harden, A. 2008. Methods for the thematic synthesis of qualitative research in systematic reviews. BMC Medical Research Methodology, 8(45). doi: 10.1186/1471-2288-8-45.

51.

Tilburt

Kaptchuk

(2008). Herbal medicine research and global health: An ethical analysis. Bulletin of the World Health Organization 86(8), 577–656. Available online: http://www.who.int/bulletin/volumes/86/8/07-042820/en/ (Accessed 13th February 2017)..

52.

UNAIDS (2006). Collaborating with traditional healers for HIV prevention and care in Sub-Saharan Africa: Suggestions for programme managers and field workers, Geneva: UNAIDS.

53.

Wedel

(2009). Bridging the gap between Western and Indigenous medicine in eastern Nicaragua. Anthropological Notebooks 15(1), 49–64.

54.

Weissman

J. S.

Stern

Fielding

S. L.

Epstein

A. M.

(1991). Delayed access to health care: Risk factors, reasons, and consequences. Annals of Internal Medicine 114(4), 325–331.

55.

World Health Organisation. (1978). Declaration of Alma-Ata International Conference on Primary Health Care, Alma-Ata, USSR, 6–12 September 1978. WHO. Accessed online: http://www.who.int/publications/almaata_declaration_en.pdf?ua=1. (Accessed 3rd November 2016)..

56.

World Health Organisation. (2002). WHO traditional medicine strategy 2002-2005. Geneva: WHO. Accessed online: http://www.wpro.who.int/health_technology/book_who_traditional_medicine_strategy_2002_2005.pdf (accessed 3 December 2017)..

57.

World Health Organisation (2008). Integrating mental health into primary care: A global perspective, Geneva: WHOAccessed online: https://www.who.int/mental_health/resources/mentalhealth_PHC_2008.pdf (accessed 15 December 2019)..

58.

World Health Organisation. (2012). Traditional and complementary medicine policy. In: MDS-3: Managing access to medicines and health technologies. 3rd ed. West Hartford, CT: Kumarian Press. Accessed online: http://apps.who.int/medicinedocs/documents/s19582en/s19582en.pdf (accessed 14 November 2016)..

59.

World Health Organisation (2015). WHO comprehensive mental health action plan 2013–2020, Geneva: WHOAccessed online: http://www.who.int/mental_health/action_plan_2013/en/ (accessed 8 August 2016).