Integrating Traditional Healers into the Health Care System: Challenges and Opportunities in South Africa

Background: Traditional medicine is widely used around the world to prevent, manage and cure illnesses. For many people living in low-and middle-income countries, particularly in rural areas, traditional medicine is their primary source of health care. Traditional medicines are often used concurrently with treatment from allopathic providers, resulting in concerns related to toxicity, dosing, adherence, and retention in chronic disease treatments. This paper explores opportunities for integrating traditional healers into the allopathic health care system by analysing the challenges hindering integration in order to improve health outcomes for those who routinely oscillate between the two systems. Method : Qualitative interviews were conducted with 30 traditional healers in rural South Africa. Thematic data analysis, using a combination of deductive and inductive coding, was applied using Nvivo Data management software. Results: The study showed that collaboration between traditional healers and the health care system exists in an ad-hoc manner, primarily via health education (allopathic to healer), provision of protective materials to healers (e.g. latex gloves), and the referral of patients to the health system for tests and treatments not available to traditional providers. Despite these collaborative efforts, mistrust and a perceived lack of respect for traditional healer's practices by allopathic providers make formal collaboration difficult. To formalize links between traditional healers and their counterparts providing allopathic health services there is a need for recognition of healers in the health care system through good communication, relationship building, and identification of areas for collaboration between the two parties. Further on, the health care systems need to continue providing aid and equipment to healers, open channels of communication which encourage bi-directional referral systems as well as improved information sharing systems and the continuous training of healers on new treatment guidelines.


INTRODUCTION
Traditional medicine is widely used and valued by patients for its accessibility and alignment with their understanding of disease causation. For many living in low and middle income countries traditional healers are the most accessible and affordable 1,2,3 treatment option, especially in rural areas.
In Africa, Asia, Latin America and the Middle East, 70-95% of the population use traditional medicine for at 4,5,6 least part of their primary health care needs.
Since the 1978 Declaration of the Alma Ata, there have been pledges, resolutions and strategies advocating for integration, where appropriate, of traditional healing practices with national health care systems. Some countries, including South Africa and Mozambique, have developed national policies to maximise the potential contribution of traditional and complementary medicine to primary 7 health care. Unlike allopathic health systems, traditional healers employ diverse methods and vary significantly between regions, making it more 8 difficult to codify their treatment practices.
Traditional practitioners play an important role in treating a wide variety of illnesses, improving quality of life, and supporting patients living with 9 incurable chronic disease. Traditional healers also contribute to a community's way of life and spiritual beliefs. For example, Traditional African Medicine aims to support a patient's overall well-being such as the body, self and society within a framework of dynamic equilibrium. This holistic approach takes into consideration the values, passions, beliefs, social interactions, and spiritual orientation of a 10 person.
In many African, South American and Asian countries, traditional medicine is used in conjunction with allopathic medicine, potentially leading to drug-drug interactions, drug toxicity, 11,12 and/or abandonment of allopathic medications. Given the potentially negative implications on patient health, particularly in the era of the 13 HIV/AIDS epidemic, better integration is needed. While most allopathic and traditional practitioners agree in principle to a partnership, details are difficult to codify. In view of these challenges, this paper explores opportunities for integrating traditional healers and allopathic providers to improve health outcomes among those who use both services. We explore the challenges hindering integration and provide recommendations on how to facilitate integration of traditional healers into South Africa's formal health care system.

Study Location
This study was primarily conducted in the Agincourt sub-district, Mpumalanga province, in north-eastern South Africa. The Agincourt Health and Socio-demographic Surveillance System (HDSS), overseen and maintained by the MRC/Wits Agincourt Research Unit, has been an important site for population demographics and health related research, including longitudinal 14,15,16 census data, since 1992.
The Agincourt HDSS includes 31 research villages and approximately 21,000 households. Its roughly 115,000-120,000 individual inhabitants are primarily impoverished and under-educated and are mainly xiTsonga or Shangaan-speaking. Roughly one-third of the permanent HDSS population is made up of Mozambican refugees, who immigrated to South 17 Africa during the 1980s. Much of the Agincourt HDSS data as well as a database containing data from 10% of the surveillance population is available 18 for public access.

Study Population
Traditional healers who were registered with the Kukula organization and had participated in a prior 19 study were identified. The Kukula organisation has more than 300 registered healers. It allows for the healers to organize themselves, advocate for increased recognition as healthcare providers, and also holds educational sessions for its members. Previous research with the Kukula organization involved the administration of a questionnaire to 221 healers which asked questions regarding their possible occupational exposures to bloodborne 19 pathogens. Based on their responses, the top quartile of the most at-risk healers was identified. Those individuals were then placed in random order using a number generator, and contact was initiated with each healer beginning from the top until 30 interviews were completed. All of the healers who could be contacted participated in the study. All healers were over the age of 18 and were actively treating patients.

Data Collection
Between April and June 2019, we conducted 30 structured in-depth, in-person interviews. These interviews were conducted in xi-Tsonga by a trained qualitative fieldworker. Healers were asked about their current protective practices, their attitudes towards partnering with local health services to improve this protection, and about performing HIV counselling and testing (HCT) on their patients. Interviews took place at the home or other preferred meeting place of the traditional healer and were conducted individually. All interviews were audiorecorded, then transcribed and translated to English from xi-Tsonga.

Data analysis
Transcripts were imported into Nvivo 12 data management software for data management. Thematic analysis was performed through the process of coding in six phases in order to create recognised and meaningful patterns. These phases are; familiarization with data, generating initial codes, searching for themes among codes, reviewing themes, defining and naming themes. For this paper, analysis was focused on the linkages between traditional healers and the health care system. The main themes identified included: challenges to successful integration, current traditional healer and health system collaborative practices, as well as opportunities and recommendations for future collaboration (See  table 1). Data was familiarized by reading and re-reading of the transcribed interviews while paying attention to patterns and occurrence. Generation of initial codes was done by documented information on use and integration of traditional healers and the health system. This involved listing items from the data set that had a reoccurring pattern. The lists of themes were combined to the coded data with proposed themes. Some themes emerged from the data in cases where the following issues occurred; repeated ideas, indigenous terms, metaphors, analogies as well as similarities of the participants' linguistic expression. The researchers critically looked at how the themes supported the collected data.

Ethics approval
This study was approved by the Vanderbilt Institutional Review Board (IRB # 190395), and the University of Witwatersrand Human Research Ethics Committee (Protocol #M160447), as well as the Mpumalanga Department of Health's Research Ethics Committee. All participants provided written informed consent using a document in their language.

Opportunities for Collaboration
All participants expressed the belief that traditional healers and health care providers should work together to effectively treat their patients but were waiting for healthcare providers to give them the direction for working together.
We were once invited to a meeting and we were told that we will start to work with the healthcare workers at the hospitals so that we can be able to treat the patients that have "tindzhaka" people who have slept with women who have terminated the pregnancies the non-medical way; anything that concerns the traditional healing. We agreed to that and our names were written down, but we have not heard anything since then. (Female 67 years old)

Referral Initiatives
In efforts to increase acceptability of allopathic referrals, healers revealed they inform their patients that they also consult healthcare workers for conditions that they cannot treat themselves. Healer's revealed that when patients come for a consultation, they look for symptoms of HIV and, if indicated, refer the patient to the health facility for an HIV test. One man noted that "When I look at the patient and suspect that the patient may be HIV positive; I take the patient to the clinic and ask the healthcare workers to do all the tests..." (Male 76 years old) Patients often reveal their HIV status to the referring healer and healers indicate that they keep their patient's HIV status confidential once told by the health care providers.
I also assure the patient that the results will be kept confidential between me and the healthcare worker. After knowing the results, it is then that I will come home with the patient and treat them. (Female 52 years old) Participants revealed that they refer their patients to the health facility not only to get an HIV test but to have them also checked for other medical conditions such as diabetes and liver disease.
Usually, when a patient comes for a consultation, I take them to the clinic to do the HIV tests before I can start with the treatment. What happens when I get to the clinic is that I ask for the healthcare workers to do all the necessary tests for my patient. Tests are not done necessarily to find out if the patient is HIV positive or not, there are sugar diabetes and liver diseases and I shouldn't prescribe the bitter or strong medicines because they interfere with the treatment that the patient is taking. (Female 45 years old) Teaching healers when to refer a patient for allopathic assistance requires continuous education and a procedure to support referrals, which are systems that have yet to be codified.
We work hand in hand with the clinics and if I have patients that need medical attention, I am able to refer them to the clinic. For instance, if a patient comes with dehydration, I have to refer that patient to the clinic. (Male, 54 years old)

Educational Initiatives
Participants revealed that as traditional healers they are currently given "health talks" at the health facilities. During these health talks, traditional healers are given information about HIV, requested to refer patients suspected of living with HIV to the health system, and encouraged to share health information with traditional healers not present at the training.
Even today I went to the clinic, I was given the health talk and there is no single day where I went to the clinic and came back without the health talk. (Female, 60 years old) In addition to the health talks and exchange of health information, participants said they are invited to workshops where training is offered to them on how to effectively treat support patients without interfering with their medical treatment. Now healers are part of the health system and we attend the meetings and workshops where we are trained on what to do on our patients. We are told that; for example, if a patient comes to me for initiation I should not just begin with the initiation process without knowing their health status. The patient needs to go to the clinic first to be tested for all the diseases and to see if they are not dehydrated. (Female, 56 years old)

Negative Attitudes of Healthcare Providers towards traditional healers
Several participants indicated that health care providers are disrespectful towards them and their treatment practices, making it difficult to form a fruitful collaboration. One man said, "I think it is because of what they believe in, I mean the religion; to a point where they think the religion of traditional healers is useless." (Male, 41 years old) Healers felt this disrespectful attitude was evident in the behaviour of clinicians. This negative attitude was mostly reported by older male traditional healers, who complained that healers were poorly treated when they visited a health facility.
When you are visiting someone; they have to welcome you. Like now that you have visited me here at home, I have even given you the chair to show that I welcome you…so it is something like that, isn't it?….. And you have time to come and talk to me. So you find that they invite us to the clinic and we go there on that given date and we wait for a very long time until we are told that the person who was supposed to be giving us health talks is not around; so you see that it is not a good way of doing things. It is impossible for us as traditional healers to go to the clinic or the hospital and treat the patients that need to be treated the traditional way because we are not allowed. ……. we are not welcomed. (Male, 51 years old)

High turnover of health care providers
The rate at which health care providers leave and are replaced was reported as another challenge to a substantive collaboration. This high staff turnover of health care providers often left facilities with young providers who were not supportive of a collaboration with traditional healers. One woman noted that she "used to get gloves from a certain nurse who is no longer there" (Female 67 years old) but that this distribution ended when that nurse was transferred.
Healers were selective about the facilities they would refer patients to avoid unpleasant situations. One woman said, "I don't use any clinic; I use one clinic because I am well known there." (Female 36 years old). The value of these relationships can not be overstated; healers were more willing to refer patients to facilities that respected their role in community health than those who treated them poorly.

DISCUSSION
This study has shown that traditional healers in rural South Africa have formed an informal partnership with the health care system, but improved relations, formal referrals, and better systems for sharing information and resources need to be developed. Our results underscore the importance of appreciation and recognition of traditional healers to ensure a successful collaboration. Health care workers often fail to reciprocate the respect which traditional healers give to trained health care 20 workers generally, leading to hurt feelings.
While opportunities for integration exist, the lack of a formal system to allow for long-term mutual engagement has led to mistrust and fragile 21 relationships. For example, while there are efforts from the formal health system to encourage referrals of patients for HIV testing, healers are expected to spend their own time accompanying the patient to the health facility while enduring disrespectful treatment of health care providers. Test results are often not relayed to the healer; such information (if accepted by the patient) would re-enforce mutual trust and result in better and safer care for 22,23 patients.
Other studies in SSA have reported similar changes in referral and co-management of patients between western medicine practitioners 24,25 and traditional practitioners.
Central to the delivery of medical practice is the availability of evidence-based information and best 26 practices. This is particularly true in HIV care 27 where protocols and guidelines change frequently. In our study, the lack of formal processes to share information by health workers with traditional healers could result in negative health outcomes among their patients. Traditional healers revealed that they were often expected to know new guidelines without being formally trained or at least 28 oriented in a timely manner. In addition, while healers are open to learning more about allopathic care systems, providers appear to have less interest in gaining information about traditional diagnostics 20 or treatments.
Individual health workers can be suspicious of traditional practices and are unwilling to engage 29 meaningfully. The lack of policy direction and platforms to support integration, has essentially left the decision to integrate services to individual practitioners and proponents for traditional healer 30 practice. Policy and guidelines for integration of these two separate yet complementary systems are urgently needed to ensure patients safety in LMICs, which have few qualified health workers. Clear guidelines and continuous interactions are needed to monitor the health outcomes and costs of care as a 13 result of collaboration. Cutting off this colonial legacy which actively discouraged use of traditional medicines and religion is an important step towards full independence and use of local sustainable 31 resources. Despite efforts to separate the two systems, patients who are the beneficiaries trust both systems and often chose which conditions to take to which practice and whether or not to disclose 32 the use of traditional medicine use to the doctor. Partnership between traditional practitioners and the health system therefore requires a significant commitment of time and finances at the beginning, and is not reached without controversy and 23 challenges.

CONCLUSION
Our study has confirmed potential for a formal collaboration between traditional medicine and the allopathic health system in SSA. The lack of policy direction and guidelines has made it difficult to bring together these complementary systems which provide parallel services to the same patients and community. Significant suspicion and mistrust remain between tradition and modern medical practices, yet patients trust both. Thus, allopathic providers and healers need to overcome our challenges to develop a functioning collaboration.

Availability of data and material
Data is available on request from authors.

Competing interests
All Authors Declare that they have no competing interest

Funding
The study was funded by NIH

Authors' contributions Acknowledgements
All participants provided written informed consent using a document in their language. They were informed that results of the study could be published but no personal identifiers will be used.
EC, CMA and RW developed the protocol; CMA, TFLM and WM coded the data and conducted the analysis. TFLM and WM drafted the preliminary manuscript. All authors edited and approve of the final manuscript.
We thank the traditional healers who participated in this study.