Understanding stigma as a barrier to accessing cancer treatment in South Africa: implications for public health campaigns

Understanding stigma as a barrier to accessing cancer treatment in South Africa: implications for public health campaigns

Tatiana Oystacher¹, Drew Blasco², Emily He³, Debbie Huang⁴, Rebekkah Schear⁵, Devon McGoldrick⁵, Bruce Link⁶, Lawrence Hsin Yang^2,4,&

 

¹Harris School of Public Policy, University of Chicago, Chicago, United States, ²College of Global Public Health, New York University, New York, United States, ³Frances L Hiatt School of Psychology, Clark University, Worcester, United States, ⁴Mailman School of Public Health, Columbia University, New York, United States, ⁵Livestrong Foundation, Austin, United States, ⁶Department of Sociology, University of California, Riverside, United States

 

 

^&Corresponding author
Lawrence Hsin Yang, College of Global Public Health, New York University, New York, United States, Mailman School of Public Health, Columbia University, New York, United States

 

 

 

 

Cancer stigma in South Africa: cancer is one of the leading causes of morbidity and mortality globally [1]. In Africa, about 700,000 people are diagnosed with cancer annually with 47,400 cases resulting in death [1]. The American Cancer Society [2] identified that the majority of deferred cancer diagnoses in Africa resulted from limited awareness of early signs of cancer, lack of biomedical screening and stigma. Cancer stigma often originates from cultural stereotypes that cancer is a fatal disease, is contagious, or is a punishment for immoral behavior [3, 4]. Additionally, cancer stigma has received less attention than communicable disease stigma (e.g. HIV/AIDS) [5]. Further, severe stigma associated with cancer has likely contributed to delays in biomedical treatment [5]. Social consequences of being labeled with a cancer diagnosis (e.g. shame) may deter use of biomedical screening, resulting in treatment delays [6-8]. Public education campaigns to increase knowledge and decrease stigma are crucial in promoting prevention and treatment of cancer in South Africa [9, 10]. Implementing interventions in culturally sensitive ways can enhance the reach of educational messages [11, 12] and increase adherence with treatment [13]. Additionally, successful strategies may include partnering with traditional healers. This has been used to increase diffusion of information regarding HIV/AIDS awareness. In South Africa, estimates range from 60% of people consulting traditional healers [14] to approximately 50% consulting both biomedical providers and traditional healers [15]. South African healers are critical advisors in the use of healthcare in their communities [16-21]. Understanding misconceptions about cancer and best practices to deliver culturally-appropriate educational messages is crucial for successful implementation of educational campaigns. This study investigates cancer stigma as a barrier to help-seeking within a South African community. Additionally, we seek to address a gap regarding knowledge of cultural stereotypes around cancer, common stigmatizing conceptions and range of discriminatory experiences. Our findings may inform interventions aimed to raise awareness, increase biomedical screenings and treatment for those diagnosed with cancer.

 

Study and theoretical framework: this paper presents data collected from 2010-2012 in three South African towns prior to Livestrong Foundation's public education efforts to reduce cancer stigma. We utilize Link and Phelan's [22] "Modified Labeling Theory" as a framework to understand local cancer stereotypes and mechanisms in which stigma manifests. We utilized three concepts (i.e. stereotyping, anticipated discrimination and coping mechanisms) from this framework to understand how individuals are stigmatized and excluded from everyday interactions. Primarily, Link and Phelan's framework has been used to examine mental illness stigma and applied to conditions of HIV [23, 24], obesity [25, 26] and hepatitis C [27]. We propose extending this theory to the illness condition of cancer, which has been relatively understudied [23]. We explore differences in how South African respondents may use local understandings of cancer in contrast to the biomedical model. We also examine ways in which stigma may manifest through different labeling processes as applied to cancer. For example, labeling may occur through formal Western diagnoses, diagnosis by traditional healers, or showing cancer symptoms. Given the prominence of traditional healing in South Africa, we seek to understand how traditional medical practices may be connected to stereotypes and stigma toward cancer and how avoidance of Western treatment labeling may influence use of treatment services. Additionally, we seek to illuminate predominant coping strategies arising from a cancer diagnosis such as withdrawal, secrecy, or how anticipating discrimination may deter someone from biomedical treatment. Finally, we utilize a rare opportunity to analyze the unexamined connection between HIV and cancer stigma in South Africa as a potential effect of prior national stigma reduction efforts for HIV [28-32].

 

 

Study design: the livestrong foundation interviewed more than 4,500 healthcare providers, cancer survivors, NGOs and experts across 10 countries. Results compelled the Livestrong Foundation to conduct a pilot anti-stigma campaign in South Africa from 2010-2012. Prior to this cancer education campaign, Livestrong Foundation conducted a series of qualitative interviews and focus groups with key community stakeholders to understand cultural conceptions of cancer stigma which we report upon here.

 

Data collection: eleven semi-structured focus groups were conducted in three peri-urban townships within South Africa: Khayelitsha, Soweto, and East London. Townships held focus groups consisting of: traditional healers (3 groups, n = 21 participants); church groups (3 groups, n = 29 participants), non-government organizations (NGOs) (2 groups, n = 33 participants); community leaders, including political leaders and members of community health forums (3 groups, n = 37 participants). Age range of church leaders and traditional healers was 55-75 years, while the age range in other groups was 26-65 years. To encourage open disclosure concerning potentially highly stigmatizing personal, family, or community experiences of stigma, no other personal information was collected. Semi-structured interviews focused on the following: causes of and possible ways of preventing cancer, contact with someone diagnosed with cancer and treatment of cancer patients by the community. In addition, focus group participants discussed how one would be treated and what a person would do/feel if he/she was diagnosed with cancer. Other topics queried included: the role of media in disseminating cancer knowledge, barriers to help-seeking and how access to prevention and treatment might be improved. Follow-up questions were asked as appropriate. All focus group interviews were audio-recorded and transcribed by South African transcription services. Data analysis was approved by the Columbia University Medical Center Institutional Review Board.

 

Data analysis: we based our analytic scheme on Link and Phelan's [21] multidimensional model of stigma. We examined ways cancer stigma might cohere and differ, with key stigma components (i.e. labeling, anticipated discrimination and coping mechanism), which have been predictive of negative outcomes [33-36]. Using directed content analysis [37], we sought to validate and extend conceptual frameworks for cancer-related stigma. Following Link and Phelan's framework as a guide, we first determined analytical schema and initial coding categories. Three research team members (T.O, D.H, E.H) and the principle investigator (L.Y.) coded the 11 focus group transcripts using the initial coding scheme. First, one transcript of each respondent type (traditional healers, NGO's, community leaders) was group-coded to modify initial coding categories. After establishing consensus, each remaining transcript was coded independently by all research team members and any discrepancies discussed. Initial coding categories were modified based on discovery of conceptual relationships between codes (e.g, merging subcodes that represented different aspects of a single coding category). While following a deductive framework, the codebook was expanded inductively to account for new concepts which arose due to culturally-specific experiences of cancer in South Africa. Discrepancies in codes were resolved during consensus meetings. The final codebook is shown Annex 1. Transcripts were analyzed using ATLAS-TI.

 

 

General cancer knowledge and stereotypes: respondents' cancer knowledge varied from no previous knowledge to some knowledge of risk factors, etiology, treatment and prognosis. Respondents cited knowledge of common cancers and risk factors (i.e smoking, drinking, unhealthy lifestyles and excessive sun exposure). Multiple stakeholders characterized cancer as deadly and incurable, leading to fear of diagnosis, hopelessness and depression (Table 1). However, some respondents reported cancer being curable in the early stages. They named methods of medical treatment-"being ironed" (getting chemotherapy) or "being burned" (getting radiotherapy)-and accompanying side effects, such as nausea, difficulty eating and hair loss. Yet multiple groups of respondents revealed uncertainty about causes, transmission and treatment (Table 1). Respondents identified four prominent stereotypical causes for cancer: "physical contagion", "witchcraft", "influence of the culture of White colonizers" (i.e, adopting their diets) and "sexuality/promiscuity". This last cause included early pregnancy, multiple childbirths and even breastfeeding (Table 1). These prominent stereotypes often led to anticipated discrimination.

 

Anticipated discrimination: stemming from these stereotypes, the most prominent topic concerned anticipated discrimination/devaluation. In addition to providing examples of discrimination against people with cancer by community members, respondents also predicted that they themselves would discriminate against their relatives/partners with cancer (Table 2).

 

Cancer linked to physical contagion, sexuality and promiscuity and witchcraft: the belief that cancer is contagious was a source of anticipated discrimination and isolation. Traditional healers and a church respondent discussed fear of contracting cancer through contact with cancer patients (Table 2). This belief resulted in avoidance of sharing items, proving separate bedding and tableware (Table 2). Notably, how a person was treated by their family could influence the attitudes of community members (Table 2). Multiple stakeholder groups endorsed stereotypes that cancer was caused by sexual promiscuity, was linked with cancer of reproductive areas and was acknowledged as punishment for immoral behavior (Table 2). Stereotypes that cancer may be sexually transmitted caused significant shame and justified ending a relationship. Finally, witchcraft stereotypes were attributable to the cancer resulting from a "curse" and someone wishing harm to a "bewitched" person (Table 2). Regardless of stereotype, cancer patients experienced a range of discriminatory occurrences from being perceived as burdensome (Table 2) to being isolated from both their family and broader social network (Table 2). Participants feared they would face severe forms of social discrimination from being labeled as a cancer patient.

 

Mechanisms of Labeling: analyses indicated ways in which a person acquired the label of a "cancer patient", some of which were more definitive than others: 1) labeling through appearance of perceived signs or symptoms of cancer, or treatment side effects; 2) cancer diagnosis from a traditional healer; 3) cancer diagnosis by a western physician.

 

Signs and symptoms of cancer or effects of cancer treatment: respondents named signs and symptoms leading to identification of a person as having cancer in the absence of a medical diagnosis. These included lumps on different parts of the body (including "abnormal things on private parts"), "problems with the throat" (e.g, difficulties swallowing, stabbing pain in the throat, coughing and having a lot of phlegm), visible weight loss and energy loss. Respondents also mentioned back pains and unpleasant odor coming from affected individuals. Further, participants identified cancer signs related to side effects of Western treatments, such as temperature changes, nausea, hair loss and in the case of breast cancer, extirpation of a breast (Table 3).

 

Diagnosis by traditional healers: according to traditional healers' perspectives, they are frequently utilized as healthcare providers first because of their easy access in communities (Table 3) and their perceived efficacy (Table 3), especially when cancer is believed to be caused by witchcraft (Table 3). Use of traditional healers may enable avoidance of the perceived invasiveness of Western cancer treatment (Table 3). While dissenting views existed, cancer diagnosis "labeling" was not as definitive when provided by a traditional healer. Instead, some traditional healer respondents reported that they could not provide a formal cancer diagnosis, thereby referring individuals to western doctors for diagnosis using biotechnology (Table 3).

 

Biomedical diagnosis: the most definitive "labeling" method was receiving a diagnosis from a western physician. This labeling process was initiated by cancer screenings at the clinic and culminated in receiving a biomedical diagnosis. While initiating beneficial biomedical treatment, this label also elicited the above cancer stereotypes. Multiple stakeholders cited that an official biomedical diagnosis (and biomedical treatment) was associated with fear, avoidance of treatment and rejection from others (Table 3).

 

Coping mechanisms in response to labeling and anticipated discrimination: several coping mechanisms, or ways those "labeled" as having cancer responded to their stigmatized status, were identified. Secrecy was the most common response. Participants provided examples of cancer patients refraining from disclosing the disease to their families or community until it was terminal. Multiple stakeholder groups cited that secrecy from family members was attributable to lack of knowledge of the signs/symptoms of cancer, fear of death, shame, concern about being laughed at or feared and fear of witchcraft (Table 4). Additionally, participants reported individuals with cancer might be afraid of stigma and discrimination from their families, spouses and community (Table 4). Less frequently mentioned were negative coping strategies of denial or withdrawal (i.e. not performing checkups due to not knowing what might be discovered or of being diagnosed with what is perceived as an incurable disease) (Table 4). Respondents openly discussed whether cancer may be treatable during its early stages. Additionally, they identified that being secretive and not seeking prompt treatment may lead to worse health outcomes. Less frequently, multiple stakeholder groups suggested openness, acceptance and self-education as positive ways of coping with a cancer diagnosis (Table 4). While less common, these provide examples of more effective coping strategies that might be built upon for future interventions.

 

Stigma's impact on accessing treatment: multiple stakeholder groups revealed distrust toward western health services or a lack of understanding of biomedical procedures, leading to fear. Several traditional healers and church leaders blamed western medical epistemology for misdiagnosis, inability to help and even for unethical experimentation on Black people. Additionally, NGO and community leader groups expressed concern that western medical professionals were not providing sufficient accurate information to communities (Table 5). Although direct stigma perpetuated by individual health providers and/or organizations of power (e.g. employers, government agencies, hospitals) were not observed, these commonly-held perspectives portrayed Western modes of treatment and of understanding disease, as an extension of an oppressive, colonialist government. This not only had negative consequences for early detection and treatment but also was seen as a tool to suppress the Black population (Table 5). The lack of consistency about the role of traditional healers in cancer treatment contributed to a disconnect between traditional healing and Western medicine. Therefore, even when traditional healers were utilized, patients might not be referred to western biomedical treatment. NGO groups and church healers held more pessimistic views towards traditional healing. NGO groups cited traditional healers as being detrimental to public health. Church leaders spoke about their belief that information provided by traditional healers often delayed Western medical diagnosis (Table 5). Community leaders rarely expressed negative views of traditional medicine and recognized the frequent use of it as a response to lack of biomedical information or mistrust of Western medicine (Table 5). Traditional healers saw themselves as valuable partners in the cancer-treatment process due to their accessibility in the local community and ability to facilitate referrals. Some of them expressed the desire to augment their own knowledge with Western medical knowledge. However, traditional healers were also concerned that Western providers would not cross-refer to them and that they were not recognized as legitimate by the biomedical field (Table 5).

 

HIV/TB education campaigns and future cancer campaigns: past education campaigns such as with HIV/AIDS and TB made promoting education and reducing stigma around cancer more complex. Participants identified the success achieved by involving multiple community and healthcare organizations in HIV and TB awareness campaigns, thus laying a foundation for cancer education. First, multiple stakeholder groups stated that previous campaigns established methods of outreach including community screening (Table 6), workshops (Table 6) and involving churches (Table 6) and youth in community education (Table 6). Second, members of church groups cited valuable outreach strategies including nurses entering communities to provide education regarding symptoms and check-ups (Table 6) and implementing testing for both HIV and cancer (Table 6). Third, prior campaigns have made significant efforts to access hard-to-reach populations by using appropriate idioms understood by the community (Table 6). Finally, delivery of an educational and anti-stigma campaign for cancer could be modeled after prior HIV campaigns, which have focused on showing real people successfully fighting HIV (Table 6).

 

Conflation of beliefs about cancer with beliefs about HIV/TB: the after effect of prior campaigns also appeared to have an unintended consequence of conflation of HIV/AIDS and TB with cancer among community members. Multiple stakeholder groups were unable to differentiate between the symptoms of HIV/AIDS and cancer (Table 6). Multiple respondents stated that because a cancer diagnosis is perceived by the community as eliciting the same deadliness as HIV/AIDS, a cancer diagnosis may lead to comparable negative responses, including isolation and abandonment (Table 6). However, comparisons with HIV/AIDS may also be beneficial. An NGO group respondent stated that reducing stigma for HIV contributed to reducing cancer stigma (Table 6). Multiple stakeholder groups, however, perceived structural discrimination against providing cancer education due to the healthcare system's focus on information and access to HIV/AIDS and TB prevention and treatment (Table 6).

 

Use of media: media outreach, including television, radio, or newspapers, was identified as a central way to educate people about cancer. Multiple stakeholder groups emphasized stories of real people and celebrities who spoke about their experience fighting cancer (Annex 2). However, factual information about causes, symptoms, course and treatment methods were lacking (Annex 2). Multiple stakeholder groups cited that previous HIV educational programs had successfully used community outreach through local churches, schools, and traditional healers, which could be leveraged for providing biomedical information and antistigmatizing messages about cancer (Annex 2).

 

 

Our study identifies three distinct ways community members may be labeled with a cancer diagnosis. We further identify the resulting anticipated discrimination and negative coping mechanisms employed, including secrecy. Our study builds on previous research illustrating the negative consequences of being labeled, including delays in accessing biomedical treatment in South Africa [4, 7, 8]. Respondents named a variety of signs and symptoms perceived to be caused by cancer leading to a cancer label, even in the absence of a biomedical diagnosis. Generally, while respondents agreed that diagnosis by a biomedical physician was the most definitive owing to biotechnology, this label also elicited fear due to negative stereotypes and potential rejection by other community members. Participants often deemed a diagnosis provided by a traditional healer as less conclusive. Traditional healers were often utilized by community members due to etiological beliefs of witchcraft, their accessibility, their perceived effectiveness in treating cancer and the lack of invasiveness of traditional treatment. Further, we found evidence that negative stereotypes corresponded with anticipated discrimination and the desire to be secretive regarding one's diagnosis. Although these links were not explicitly stated, these findings are consistent with empirical studies of Modified Labeling Theory [33, 34, 36], which demonstrate that these processes are interlinked. Finally, this study highlights an unexplored finding; namely, that previous HIV interventions may unintentionally act as a significant source of etiological misunderstanding and an additional source of stigma in South African communities (see below).

 

Because of the stigma elicited by a biomedical diagnosis of cancer and mistrust of Western hospitals, a powerful theme emerged whereby some respondents indicated a preference for traditional healing. While dissenting views existed, some community members believed in the efficacy of traditional healers to treat cancer. Our results are generally consistent with studies showing that traditional healers are utilized first due to their accessibility [13-21]. Future education campaigns might involve traditional healers in the process of identifying those who may benefit from cancer treatment, and may act as key referral sources for biomedical practitioners. To capitalize on some traditional healers' interest in increasing knowledge of Western medicine, we propose that partnerships between traditional healers and Western physicians may increase referrals to Western biomedical treatment. Developing these partnerships may also allow space for cultural interpretations of illness. Future studies might also examine whether integration of traditional healers into medical services is associated with greater access to treatment, and less stigma. Due in part to the prior implementation of educational campaigns to address HIV, multiple stakeholder groups conflated symptomatology of HIV with cancer symptoms. Our findings build on a prior study where confusion between HIV symptoms and diabetes was reported among comorbid patients in Cameroon [13]. In our study sample, increased stigma appeared to result from beliefs concerning the interconnectedness of HIV/AIDS and cancer, the inability of participants to differentiate between these illnesses and the relative lower institutional priority given to cancer education. Additional cancer stigma due to conflation of stereotypes of deadliness associated with HIV may contribute to negative consequences, including social isolation. Yet positive aspects of prior HIV/AIDS education campaigns also emerged. Due to the perceived similarity of these two illnesses, instances of stigma reduction around HIV/AIDS sometimes also led to reduction of stigma for cancer.

 

Additionally, multiple stakeholder groups cited successful approaches for community outreach for future cancer campaigns, including community screening, workshops and involvement of churches and youth in educating communities. Other than the prior cited study [13], few studies to our knowledge have examined the effects of campaigns following prior educational efforts. To maximize efficacy, future campaigns should take into consideration: a) prior outreach methods; b) preexisting knowledge of other illnesses; c) and how education around new illnesses might interact with preexisting knowledge. Our findings should be considered in light of several limitations. First, data collection occurred in a small community in South Africa and thus cannot be generalized outside this context. Additionally, to protect participants' confidentiality and reduce the possibility of social desirability when inquiring about stigma, limited demographic information was collected. Finally, interviews were conducted at one time point prior to the implementation of the stigma education campaign; we are therefore unable to examine potential changes in stigma during the campaign. This study has several strengths, including the use of qualitative methodology to more thoroughly understand the nuanced nature of stigma. We recruited a sample of diverse stakeholders including traditional healers, church groups, NGOs, community and political leaders and members of community health forums, allowing for a broader sampling of perspectives. Further, to our knowledge, excluding studies examining this issue in connection to cervical cancer [4, 7, 8, 11], this is one of the first studies to examine cancer stigma in South Africa. Finally, this study draws on Link and colleagues Modified Labeling Theory and thus we propose the possible extension of this framework to a new illness condition.

 

 

In conclusion, our study advances the literature by demonstrating the significant negative effects of cancer stigma in South Africa. Using Link and colleagues Modified Labeling Theory, we identified three distinct ways individuals may be labeled with a cancer diagnosis. Consideration of these three labeling mechanisms and their resulting outcomes is important when designing future educational interventions. Future campaigns should examine the possibility of including traditional healers as key collaborators for cancer treatment to increase community member referrals, with the ultimate goal of increasing uptake of biomedical treatment. Finally, when designing new community education campaigns, the potential impact of previous campaigns should be carefully considered.

 

 

The authors declare no competing interest.

 

 

Tatiana Oystacher analyzed and interpreted the data, conducted the literature search, wrote the first draft, commented on the intellectual content, provided extensive revisions to the manuscript and reviewed and approved the manuscript for publication. Lawrence Hsin Yang conceptualized the study write-up, analyzed and interpreted the data including providing the conceptual framework for the analyses and results, commented on intellectual content, provided extensive revisions to all drafts of the manuscript and reviewed and approved the manuscript for publication. Drew Blasco interpreted data, conducted the literature search, commented on intellectual content, provided extensive revisions to the manuscript, and reviewed and approved the manuscript for publication. Emily He analyzed and interpreted the data, contributed to the first draft, commented on intellectual content and approved the manuscript for publication. Debbie Huang analyzed and interpreted the data, contributed to the first draft, commented on intellectual content and approved the manuscript for publication. Rebekkah Schear coordinated the data collection, commented on intellectual content and approved the manuscript for publication. Devon McGoldrick coordinated the data collection, commented on intellectual content and approved the manuscript for publication. Bruce Link made substantial contributions to interpretation of data, advised on the concept and design of the study, commented on intellectual content and approved the manuscript for publication.

 

 

Work on this project was supported by Livestrong Foundation. Debbie Huang's participation in this study was supported by the grant NIH T32 MH013043 (Mailman School of Public Health Columbia University Psychiatry Epidemiology Training Program).

 

 

Table 1: knowledge and stereotypes of cancer

Table 2: devaluation and discrimination

Table 3: labeling process

Table 4: coping mechanisms and emotional response

Table 5: services and treatment

Table 6: interrelation between beliefs about cancer and HIV

 

 

Annex 1: final coding protocol

Annex 2: recommendations from respondents

 

 

1. Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, Mathers C, Rebelo M, Parkin DM, Forman D, Bray, F. GLOBOCAN 2012 v0, Cancer Incidence and Mortality Worldwide: IARC CancerBase 11; Lyon, France: International Agency for Research on Cancer; 2013. Accessed 30 October 2017.


2. American Cancer Society. Global Cancer Facts & Figures 3^rd Edition. Atlanta. American Cancer Society. 2015. Google Scholar


3. Daher M. Cultural beliefs and values in cancer patients. Ann Oncol. 2012; 23(Supplement 3): 66-69. PubMed | Google Scholar


4. van Schalwyck SL, Maree JE, Wright SCD. Cervical cancer: the route from signs and symptoms to treatment in South Africa. Reprod Health Matters. 2008; 16(32): 9-17. PubMed | Google Scholar


5. van Brackel, WH. Measuring health-related stigma: a literature review. Psychol Health Med. 2006; 11(3): 307-334. PubMed | Google Scholar


6. Travis K. Health experts aim to curb potential epidemic. J Natl Cancer Inst. 2007; 99(15): 1146-1147. PubMed | Google Scholar


7. Abrahams N, Wood K, Jewkes R. Barriers to cervical screening: women's and health workers' perceptions. Curationis. 1997; 20(1): 50-2. PubMed | Google Scholar


8. Wood K, Jewkes R, Abrahams N. Cleaning the womb: constructions of cervical cancer screening and womb cancer among rural black women in South Africa. Soc Sci Med. 1997; 45(2): 283-294. PubMed | Google Scholar


9. Link BG, Phelan JC. Stigma and its public health implications. Lancet. 2006; 367(9509): 528-29. PubMed | Google Scholar


10. Keusch GT, Wilentz J, Kleinman A. Stigma and global health: developing a research agenda. Lancet. 2006; 367(9509): 525-7. PubMed | Google Scholar


11. Maree JE, Wright SCD. Cervical cancer: does our message promote screening? A pilot study in a South African context. Eur J Oncol Nurs. 2011; 15(2): 118-123. PubMed | Google Scholar


12. Chirwa S, Mwanahamuntu M, Kapambwe S, Mkumba G, Stringer J, Sahasrubuddhe V, Pfaendler K, Parham G. Myths and misconceptions about cervical cancer among Zambian women: rapid assessment by peer educators. Glob Health Promot. 2010 Jun; 17(2 Suppl): 47-50. PubMed | Google Scholar


13. Awah PK, Phillimore P. Diabetes, medicine and modernity in Cameroon. Africa. 2008; 78(4): 475-495. Google Scholar


14. van Wyk BE, van Oudtshoorn B, Gericke N. Medicinal plants of South Africa. Pretoria. Briza Publications. 1999. Google Scholar


15. Case A, Menendez A, Ardington C. Health seeking behaviour in Northern KwaZulu Natal. Cape Town. University of Cape Town: Centre for Social Science Research. 2005. Google Scholar


16. Kale R. Traditional healers in South Africa: a parallel health care system. BMJ. 1995; 310(6988): 1182-1185. PubMed | Google Scholar


17. Ashforth A. Madumo: a man bewitched. Chicago, IL. University of Chicago Press. 2000. Google Scholar


18. Janzen JM. Ngoma: discourses of healing in Central and Southern Africa. Berkeley, CA. University of California Press. 1992. Google Scholar


19. Brier M. The shortage of medical doctors in South Africa, Scarce and Critical Skills Research Project. South Africa. Department of Labour. 2008. Google Scholar


20. Nelson JA, Francis SA, Liverpool J, Soogun S, Mofammere N. Healers in a non-traditional role; a focus group study of Sangoma's knowledge of and attitudes to cervical cancer prevention and screening in Johannesburg, South Africa. Sex Reprod Healthc. 2010; 1(4): 195-196. PubMed | Google Scholar


21. Steyn M, Muller A. Traditional healers and cancer prevention. Curationis. 2000; 23(3): 4-11. PubMed | Google Scholar


22. Link BG, Phelan JC. Conceptualizing stigma. Annu Rev Sociol. 2001; 27: 363-385. Google Scholar


23. Fife BL, Wright ER. The dimensionality of stigma: a comparison of its impact on the self of persons with HIV/AIDS and cancer. J Health Soc Behav. 2000; 41(1): 50-67. PubMed | Google Scholar


24. Mahajan AP, Sayles JN, Patel VA, Remien RH, Ortiz D, Szekeres G, Coates TJ. Stigma in the HIV/AIDS epidemic: A review of the literature and recommendations for the way forward. AIDS (London, England). 2008; 22(Suppl 2): S67-S79. PubMed | Google Scholar


25. Carr D, Friedman MA. Is obesity stigmatizing? Body weight, perceived discrimination, and psychological well-being in the United States. J Health Soc Behav. 2005; 46(3): 244-59. PubMed | Google Scholar


26. Lewis S, Thomas SL, Blood RW, Castle DJ, Hyde J, Komesaroff PA. How do obese individuals perceive and respond to the different types of obesity stigma that they encounter in their daily lives: a qualitative study. Soc Sci Med. 2011; 73(9): 1349-56. PubMed | Google Scholar


27. Golden J, Conroy RM, O'Dwyer A, Golden D, Hardouin JB. Illness-related stigma, mood and adjustment to illness in persons with hepatitis C. Soc Sci Med. 2006; 63(12): 3188-3198. PubMed | Google Scholar


28. Merson MH, O'Malley J, Serwadda D, Apisuk C. The history and challenge of HIV prevention. Lancet. 2008; 372(9637): 475-88. PubMed | Google Scholar


29. UNAIDS. Collaboration with traditional healers in HIV/AIDS prevention and care in sub-Saharan Africa: a literature review, Geneva, Switzerland. UNAIDS Best Practice Collection, Key Documents. 2000.
30. Flint A, Payne J. Reconciling the irreconcilable: HIV/AIDS and the potential for middle ground between the traditional and biomedical healthcare sectors in South Africa. Forum Dev Stud. 2013; 40(1): 47-68. Google Scholar


31. Mills E. HIV illness meanings and collaborative healing strategies in South Africa. Social Dynamics. 2005; 31(2): 126-160. Google Scholar


32. Thomas F. Indigenous narratives of HIV/AIDS: morality and blame in a time of change. Med Anthropol. 2008; 27(3): 227-256. PubMed | Google Scholar


33. Link BG, Struening EL, Rahav M, Phelan JC, Nuttbrock L. On stigma and its consequences: evidence from a longitudinal study of men with dual diagnoses of mental illness and substance abuse. J Health Soc Behav. 1997; 38(2): 177-190. PubMed | Google Scholar


34. Link BG, Struening EL, Neese-Todd S, Asmussen S, Phelan JC. Stigma as a barrier to recovery: the consequences of stigma for the self-esteem of people with mental illnesses. Psychiatr Serv. 2001; 52(12): 1621-1626. PubMed | Google Scholar


35. Hatzenbuehler ML, Phelan JC, Link BG. Stigma as a fundamental cause of population health inequalities. Am J Public Health. 2013; 103(5): 813-821. PubMed | Google Scholar


36. Markowitz FE. The effects of stigma on the psychological well-being and life satisfaction of persons with mental illness. J Health Soc Behav. 1998; 39(4): 335-47. PubMed | Google Scholar


37. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qualitative Health Research. 2005; 15(9): 1277-1288. PubMed | Google Scholar