Drinking, Smoking, and Morality: Do ‘Drinkers and Smokers’ Constitute a Stigmatised Stereotype or a Real TB Risk Factor in the Time of HIV/AIDS?

Abstract

This paper follows up an unexpected finding from a community survey that identified drinking and smoking as the most important tuberculosis (TB) risk factor, far ahead of ones commonly associated with TB such as poverty, overcrowded living conditions, and HIV-positive status. It reports perceptions of drinking and smoking from a three-phased study of the stigma associated with TB, consisting of a qualitative pilot study using focus-group discussions (2006), a larger-scale community survey (2007), and follow-up group discussions (2009). The community attitude survey was conducted with a sample of 1,020 adults living in a low-income township in the Eastern Cape Province, South Africa. The study found that the moral and the biomedical understanding of TB risk are intertwined. In the community survey, perceptions of drinking and smoking as TB risk were predicted by fear of contracting TB and being a self-reported born-again Christian. In the follow-up study, heavy drinking and smoking in shebeens (unlicensed township liquor outlets) was associated with a risky lifestyle that can spread both TB and HIV. The paper discusses the similarities and differences in the roles of church and shebeen in providing social support to township dwellers to cope with problems of daily life. It is tentatively concluded that the stereotypical shebeen ‘drinkers and smokers’, alternatively pitied and maligned by moral society, might serve as the scapegoat that deflects pollution from the ‘new’ TB linked to the AIDS epidemic.

Introduction

TB is an ancient disease. Recently it has reached epidemic proportions particularly in developing countries with high HIV prevalence rates. Some experts speak of ‘three’ epidemics. The ‘first’ refers to the resurgence of tuberculosis in the early 1980s. HIV-related TB is known as the ‘second’ epidemic. The surge of multi-drug resistant TB is designated the ‘third’ epidemic (Maurya et al. 2002). The Eastern Cape province of South Africa, where our study site is located, has been hit by all three epidemics. The first epidemic was introduced to the Eastern Cape by migrant workers returning home from the gold mines (Packard 1992). With regard to the second epidemic, currently an estimated 58.8% of TB cases in the Eastern Cape are co-infected with HIV (SAIRR 2008:435). The ‘third’ epidemic arrived in 2006, when cases of extremely drug-resistant TB (XDR-TB) were first detected in neighbouring KwaZulu-Natal province, and then in all other South African provinces, including the Eastern Cape. The study reported here aimed to examine the role of stigma in undermining the control of tuberculosis in a region that has experienced the three TB epidemics in succession. Our study focuses on drinkers and smokers as a possible stereotype or ostracised group in society that is blamed for spreading TB.

The Eastern Cape Province has a long history of tuberculosis. Known as the ‘white’ plague, tuberculosis was brought to South Africa around the turn of the nineteenth century by British and European immigrants seeking their fortunes on the Witwatersrand. Many passed TB onto African men working on the goldmines who were susceptible due to the overcrowded and generally poor working and living conditions in the compounds. Miners too sick to work were sent back to their rural homes. The worst hit by the epidemic was the Eastern Cape, one of the largest reservoirs of mine workers (Packard 1992). Today tuberculosis is endemic in the Eastern Cape. Eight of the 21 TB hospital centres operating countrywide are in the Eastern Cape. One of these hospitals is in Grahamstown East/Rhini, Makana Municipality, the site of the present study. The province’s multiple-drug-resistant TB hospital is located in the nearest coastal city, some 130 km away.

Stigma is thought to be a major factor that hampers efforts to fight tuberculosis (Westaway and Wolmarans 1994; Liefooghe et al. 1995; Kelly 1999; Cassie 2002; Karim et al. 2007). Stigmatised individuals are variously shunned, isolated, ridiculed and cold-shouldered. The consequences of stigma are remarkably similar in different health conditions and cultures (van Brakel 2006). Stigma is known to delay treatment and to lead to non-adherence to treatment of various diseases, particularly self-inflicted ones. This is also the case with TB-related stigma (Godfrey-Faussett et al. 2002; Kanyerere and Aase 2005; Cambanis et al. 2007; Somma et al. 2008).

South Africa has one of the highest TB–HIV co-infection rates in the world; 73% of all TB cases are HIV+ according to the World Health Organisation (2009:145). Given the high incidence of twin infections, it can be expected that the stigma related to HIV also impacts on the control of TB. Until recently, the ruling African National Congress government denied that the HI-virus is the cause of AIDS—ostensibly so as not to project South Africans as a promiscuous people—and the health minister advocated a regimen of beetroot and garlic instead of treatment with antiretrovirals. The stigma related to AIDS has been fuelled by this denialism. It was only in 2004 that the government consented to the rollout of antiretrovirals in public health clinics. Over 300,000 unnecessary deaths have been attributed to the delay (Chigwedere et al. 2008).

Structure of the Paper

The paper is divided into five parts. Following on this introduction, the second part gives a brief outline of the literature on layers of stigma related to the ‘second’ TB epidemic, also referred to as the ‘new’ epidemic (Bond and Nyblade 2006). The third section describes the research design applied in this study that was divided into three phases. The fourth section reports results from the three phases of the study that relate to the risk and stigma associated with drinking and smoking. The final section discusses the conclusions drawn from the findings of the three-phased study and recommendations for future research.

Literature

Stigma Related to the ‘New’ TB

Historically, stigma has been the feature of many diseases, especially those that are sexually transmitted or terminal (Deacon 2005). African studies have found that HIV and TB are seen to be closely linked and highly stigmatised (Gelaw et al. 2001; Godfrey-Faussett et al. 2002). However, several international studies, as well as our own, have found that a lesser stigma is commonly associated with TB than with HIV/AIDS mainly owing to the mode of transmission (Nichter 1994; Balabanova et al. 2006; Macq et al. 2006; Mak et al. 2006; Daftary et al. 2007; Møller and Erstad 2007). Sexually transmitted diseases such as HIV tend to be associated with an immoral lifestyle that underpins stigmatisation. A Russian study found that TB was known as a ‘clean’ disease; HIV/AIDS as a ‘dirty’ disease (Balabanova et al. 2006). A study conducted in Hong Kong that compared the public stigma related to AIDS, TB and SARS (Severe Acute Respiratory Syndrome) found AIDS was associated with the most stigma, followed first by TB, then by SARS (Mak et al. 2006).

Layers of Stigma

The ‘new’ TB has attracted an additional layer of stigma associated with HIV infection (Gilmore and Somerville 1994; Lawless et al. 1996; Reidpath and Chan 2005; Ware et al. 2005; Bond and Nyblade 2006). In the South African context, the government’s denialist stance towards the AIDS epidemic has served to increase the fear and stigma associated with HIV-infection, that in turn has spilled over onto TB. Daftary et al. (2007) observed that co-infected South African TB patients used a strategy described by Goffman (1963) in his classical study of stigma as ‘covering’. A person with several discreditable attributes ‘covers’ the dominant stigma with a less damaging one. In the South African study, individuals co-infected with TB and HIV/AIDS preferred to disclose TB rather than HIV as it was seen to be less stigmatising. AIDS as the dominant stigma was associated with personal blame and immoral behaviour while TB was less likely to be perceived as being morally reprehensible (Daftary et al. 2007). Other examples in the literature demonstrate that members of marginal groups may adopt an identity that defines them in opposition to a stigmatising label (Collins et al. 2008; Lichtenstein 2008).

Protection from Stigma

Stigmatisation is regarded as a complex social process tied into existing social mechanisms of exclusion which include stereotyping and scapegoating (Alonzo and Reynolds 1995; Link and Phelan 2001; Fassin 2002; Parker and Aggleton 2003; Stein 2003). According to Deacon’s (2005) extensive literature review, disease stigma can be understood as a problem of fear and blame rather than as a problem of ignorance. Importantly, stigma may consist of beliefs that are part of a social process that differentiates against those with a disease in negative social as well as biological terms. Stigma as it is portrayed in Goffman’s (1963) classical description attributes blame, shame and abhorrence, particularly to marginalised groups. People gain an illusion of control by attributing risk-enhancing behaviour to the ‘other’ (Crawford 1994). Attributing medical conditions to lack of personal responsibility helps to distance the ‘moral majority’ from risk (Deacon 2005). Thus, stigmatisation can be seen as a protective response to a threat or adverse situation (Gilmore and Somerville 1994).

‘Othering’: Stereotyping and Scapegoating

Stereotyping, scapegoating, and labelling of the ‘others’ whose behaviour and lifestyles do not conform to cultural norms can function as ‘fast tracks’ to stigmatisation (Gilmore and Somerville 1994). The specific behaviour expected of morally upright citizens may differ according to the cultural setting (Alonzo and Reynolds 1995; Link and Phelan 2001; Lichtenstein 2008). Typically, marginal groups are blamed for engaging in a lifestyle which involves risky behaviour that results in infection, then for perpetuating the epidemic by defaulting on treatment due to these habits (Ware et al. 2005).

The second TB epidemic has often represented AIDS as divine punishment for transgressing religious or moral principles (Gilmore and Somerville 1994). Throughout history, people with infectious diseases have been turned into scapegoats that ritualistically expel ‘evil’ from a community or society. According to Gilmore and Somerville (1994) the scapegoated person must be human enough to carry away the ‘sins’ of the community. The scapegoat must also be dehumanised in order to be isolated, ostracised, or in some way separated from the scapegoating community so it can expel those ‘sins’. Historically, the ritual of scapegoating involved two goats, a good and an evil one. One was sacrificed to the gods, the other exiled into the desert.

Attributing TB Risk to Drinking and Smoking

Schmidt (2008) notes that for centuries, TB has been linked anecdotally with environmental risk factors that go hand in hand with poverty such as indoor pollution, tobacco smoke, malnutrition, overcrowded living conditions and excessive alcohol use. Empirical studies of attribution have found that drinking and smoking are popularly considered a risk factor for TB. For example, a community study conducted in rural India identified addictions such as smoking, alcohol and drug use as a cause of TB and found an association between TB and sexual experience (Atre et al. 2004). Nichter’s (1994) study of illness semantics in the Philippines found that excessive habits of smoking, drinking and sex were thought to place one at risk of TB by weakening the general state of health.

The ‘new TB’ in the HIV/AIDS era carries stronger and more damaging associations with deviant and culpable behaviour because of its association with HIV. Multiple concurrent partners are a major factor driving HIV infection at present (Fisher et al. 2007). Bond and Nyblade writing from Malawi, report that the ‘old TB’ was caused by a range of traditional habits that included smoking and drinking of home-brewed liquor but the ‘new TB’ was more often associated with hanging out in bars and, in towns, with sexual transgressions. “Improper or immoral behaviour is speculated about and often assumed, with judgement passed about how a person contracted TB–HIV” (2006, p. 456). In a South African study of TB-related stigma, Naidoo et al. (2009, p. 61) cite mainly male patients who believe their TB was caused by high risk behaviours such as drinking and smoking, visiting shebeens and womanising. Treatment adherents and health workers cited abstinence from drinking and smoking as the most important determinant of non-compliance to treatment in an earlier Q-methodological study conducted by Cramm et al. (2009) in Rhini, the site of this study.

According to Somma et al. (2008), contemporary TB-related stigma is typically associated with socially unacceptable lifestyles and behaviour with reference to sexual behaviour, dirtiness and promiscuity, alcohol, smoking, and ‘sins’. Smith (2007) notes that TB, AIDS, and smoking, along with other infectious diseases including sexually transmitted diseases (STDs), hepatitis and flu are regularly presented in ‘stigma’ formats in the media, whereas other mainly non-infectious conditions such as cancer and heart disease appear in the ‘challenge’ format which is sympathetic to sufferers. In the scale developed by van Rie et al. (2008) to measure stigma associated with TB and HIV/AIDS, an item attributes guilt to self-inflicted TB that was caused by “smoking, drinking, or other careless behaviours”. In this context, ‘careless behaviour’ could easily be understood as immoral behaviour such as sexual promiscuity.

The Biomedical Risk of TB from Drinking and Smoking

Although excessive drinking and smoking are among the top population-attributable risk factors for TB globally (Schmidt 2008), there is surprisingly little biomedical data on the association of smoking and drinking with tuberculosis. A major problem is that drinking and smoking tend to go hand in hand so it is difficult to disentangle the effects of the two habits (Schmidt 2008). Social gatherings where drinking and smoking take place are common sites for TB transmission (Zuger 1999) and alcohol consumption and smoking may be proxies for frequenting locations that put one at raised risk of close contact with infectious individuals (Harling et al. 2008). In their review of the medical literature, Maurya et al. (2002) found only 16 studies published between 1956 and 2002 that included the keywords ‘smoking’ and ‘tuberculosis’. The evidence from these 16 studies suggested that smoking should be considered an important risk factor for TB. The reviewers note that authors expected most drinkers to be in the TB group because of the known association between drinking and smoking: most drinkers were also smokers (Maurya et al. 2002). More recently, the multilevel analysis of South Africa’s TB epidemic by Harling et al. (2008) concluded that alcohol abuse, cigarette smoking, and low body mass index each independently provided a risk factor for TB even after adjusting for the socio-economic status.

Drinking, Smoking and TB–HIV Co-Infection

A pertinent question in the South African context of TB–HIV co-infection is whether there is an association between drinking, smoking and HIV-infection. South Africa has a double burden of HIV and alcohol abuse (Morejele et al. 2006). Some 5.5 million in a population of approximately 48 million are HIV positive. The amount of alcohol consumed per adult is among the highest in the world (Parry 2005). Problem or binge drinking is strongly associated with the number of sexual partners a person has had and with unprotected sexual activity (Raj et al. 2009). Several recent South African studies support the link between alcohol consumption, alcohol outcome expectancies and sexually risky behaviour that are related to HIV transmission risks (Kalichman et al. 2006, 2007). However, other studies suggest youthfulness and sensation-seeking (Greene et al. 2000) and drug use (Ehrenstein et al. 2004) may need to be factored into the equation. Alcohol use before sex is consistently related to sexual risk practices and should be considered a marker for sexually transmitted infections and HIV risks (Kalichman et al. 2007). A qualitative study of bar and shebeen patrons in an urban South African setting in Gauteng Province by Morojele et al. (2006) recorded the attitudes and behaviours of drinkers and smokers in taverns and township shebeens, the licensed and unlicensed premises where liquor is sold. Fieldworkers observed that many casual sexual encounters occurred while men were drinking. Cross-generational relations between older men, so-called ‘sugar daddies’, and younger women were common in that setting. However, heavy drinking seemed to exacerbate rather than cause the sexual risk behaviours in question (Morojele et al. 2006).

Research Aims

Alcoholism is a serious social problem in the Eastern Cape. It is associated with sexual abuse and violence. Many faith-based organisations and churches ask their members to abstain from drinking and to a lesser degree from smoking. The South African government has successfully banned smoking in public places and has sought to regulate the selling of liquor. Education campaigns warn of the health risks of smoking and the dangers of drinking and driving. Thus, drinkers and smokers might be considered easy targets for stigmatisation.

Drinking and smoking is a topic that has by and large been overlooked in the literature on TB-related stigma in the second epidemic related to HIV. This study aimed to fill the gap. The analysis presented here sought to clarify whether drinkers and smokers are one of multiple layers of stigma linked to the ‘new’ TB. Specifically, our study intended to inquire whether attributing TB risk to drinkers and smokers is a form of ‘othering’, that is, a projection of risk onto a marginal group in society. Following Smith’s (2007) distinction between popular presentations of disease as ‘stigma’ or ‘challenge’, the question is whether ‘drinkers and smokers’ represents the stereotype of a stigmatised marginal group or a social challenge for society.

Research Design

Method

The Study Site

The study was conducted in a low-income residential area of Grahamstown, Makana Municipality, a small university town and educational centre in the Eastern Cape Province that has no industry and a high rate of unemployment. The township of Rhini (estimated 100,000 population), formerly reserved for black residents during the apartheid era, has a high incidence of TB.

The research proceeded in three stages: A pilot study using focus-group methodology was conducted in early 2006 to inform a larger sample survey fielded in late 2007. To assist with the interpretation of the survey results, we conducted a second set of group discussions that focused on risks and stigma associated with drinking and smoking.

Phase 1 Pilot Study: Focus-Group Discussions

Eight focus-group discussions were conducted prior to the main study to inquire whether Rhini residents associated TB with HIV/AIDS and AIDS-related stigma. A total of 59 men (22) and women (37) were interviewed with the aim of covering a wide range of viewpoints. Discussants in the eight groups were recruited purposefully among community health workers, TB patients in hospital, high school pupils, out-of-school youth, middle-aged women, a mixed group of middle-aged adults, older men and women, and traditional healers (see Møller and Erstad 2007).

Phase 2 Main Community Study: Representative Sample Survey

Sample design. An area-stratified proportional-to-size sampling design was applied to select 1,020 survey participants in Rhini. The township was first divided into 26 neighbourhoods. Within each neighbourhood, every tenth household was targeted using a random starting point. A Kish grid was used to select the interviewee, an adult over the age of 18 years who had lived in Rhini for at least 6 months during the previous year. In total, 1,020 respondents were included in the survey.

Fieldwork. Fieldwork was carried out in November 2007. Personal interviews were administered by trained fieldworkers in the respondent’s language of choice using a questionnaire schedule containing mainly closed-ended items. The questionnaire, which was available in the local language, isiXhosa, covered views on risk factors, case finding, adherence to treatment, financial and other assistance to TB patients and their families, and disclosure. Some 50 items explored attitudes and beliefs related to TB and AIDS. The researchers were aware that some of the questions put to respondents might sow seeds of doubt and prejudice and thereby inadvertently amplify stigma (see Somma et al. 2008). To counter this problem, questions were prefaced with facts on TB and prejudicial attitude items were introduced as hearsay. The paper reports only select findings on issues related to drinking and smoking and risk factors.

Sample characteristics. The median age of respondents was 38 years and women represented the majority (73%). The single largest group (40%) had completed some secondary education and 18% had matriculated. Fewer than 7% had received post-matriculation education and training. Only 8% had no formal schooling and a further 30% had received some primary education. Just under half (48%) came from households earning less than Rand 1,000 (circa US$ 150) per month and 70% came from households that received a non-contributory social grant which is means-tested. The majority of respondents were members of a Christian church. Three quarters (77%) of all respondents reported they attended church regularly (42%) or occasionally (35%) and 44% identified themselves as ‘born-again’ Christians. Worth noting is that many respondents had personal experience of TB. Almost a third (32%) reported that a case of TB had occurred in their household. Some 45% had acted as a treatment supporter by assisting a TB patient to take their medicine regularly (Møller 2008).

Phase 3 Follow-Up Study

The third phase of the investigation consisted of a further set of five focus-group discussions with Rhini residents to shed more light on attitudes to drinking and smoking. A total of 30 discussants, 13 men and 17 women, were recruited purposefully among male churchgoers, female churchgoers, shebeen patrons, youth, and middle-aged men and women. Ages ranged between 18 and 58 years and the majority had received a secondary school education. Discussants represented eight different neighbourhoods in Rhini. Two-thirds (23) reported church membership affiliated to some 15 mainstream and charismatic Christian denominations. Topics of discussion were the key characteristics of drinkers and smokers and motivations for drinking and smoking. Each group was invited to play a ‘word game’, modelled on the classic semantic differential that asks respondents to choose a position on a scale between two bipolar adjectives. The focus groups were presented with a set of 19 bipolar adjectives or descriptors and asked to select the one in each pair that characterises the typical drinker and smoker.

Results

Results on perceptions of drinking and smoking are reported from the three phases of research.

Results from the Pilot Study

Participants in the pilot study represented a mix of generations, socio-economic backgrounds, and bio-medical and lay knowledge of TB. A unique feature of the pilot study was that the facilitators were instructed to avoid explicit mention of AIDS and stigma. Nonetheless, participants in each focus group spontaneously identified stigma as a factor that discouraged individuals to present for and complete treatment. The study found that the stigma associated with TB had come full circle in the Eastern Cape. Apparently, TB was stigmatised in the past when it was known mainly as an incurable ‘Xhosa disease’. TB lost its stigma when a cure was found, only later to be re-stigmatised as a marker of AIDS in what the literature calls the ‘new’ epidemic. Uncertainty surrounding the diagnosis and cure of TB for persons co-infected with HIV contributed to prejudice. By consensus, AIDS was characterised as a ‘killer’ disease. In line with public health messages, TB was projected as a curable and therefore less serious disease if the prescribed course of treatment was completed (Møller and Erstad 2007).

Although most discussants thought anyone could be infected with TB, there was consensus that people who drink and smoke were more at risk than others. TB was regarded as self-inflicted in the case of drinkers and smokers. The resurgence of TB in the province was often linked to human failure: “If precautions are adhered to, TB would be curable”. “It’s good behaviour, taking care of ourselves. Because if people were to take care and behave, the [number of] cases would have dropped by now”. A common perception was that the disease could be fatal in the case of patients not ‘following instructions’ when undergoing TB treatment and ‘enjoying life’, meaning that they engaged in irresponsible behaviour such as excessive drinking and smoking. “The treatment doesn’t go well with drinking” declared the out-of-school youth. Non-compliant cases were almost always discussed in terms of what the disease stigma literature calls the immoral ‘other’ category. The out-of-school youth associated alcohol with TB and repeat infections. “He liked drinking a lot and so TB attacked him again. Most of the people that I see with TB are the ones that like drinking a lot”. Similarly, the health workers also linked smoking and drinking to health relapses: “Some resort to bad behaviours of drinking and smoking after completing treatment”. Apparently, according to the health workers, it was more difficult to get men to take the TB test. “They are very much anti-clinic because they have fears that they smoke and this might be linked to TB”. Male participants confirmed their reluctance to visit health clinics where they were reprimanded by health workers: “We liquor drinkers are spoken badly to in front of everyone”.

The focus-group discussants were of the opinion that TB patients, who smoked or drank while on treatment, could not expect to be cured. Their comments suggested that the drinkers and smokers might belong to a maligned or marked group akin to the scapegoat that is burdened with the ill in society, a theme under investigation in the second and third phases of the study.

Results from the Community Survey

The most significant and unexpected finding from the main survey1 was the importance assigned to drinking and smoking as a TB risk. At the beginning of the interview, respondents were asked to identify the characteristics of people most at risk of TB. The item aimed to explore whether layers of stigma might be associated with TB. Sixteen response options were offered from which respondents could choose four (Table 1). Options included both social and environmental factors considered to be relevant in a high TB prevalence setting: HIV-co-infection, poverty, overcrowded living conditions, low social status indicated by lack of education and employment, and a tainted character or lack of social respectability indicated by non-conformist or socially discredited behaviour such as laziness, promiscuity, not going to church regularly, frequenting shebeens, and excessive drinking and smoking.

Table 1.

Characteristics of people most at risk of TB infection (cumulative percentages)

	Cumulative percentages over 4 responses (n1020 = 100%)
	1st Response	2nd Response	3rd Response	4th Response
People who…
(h) Drink and smoke a lot	19.6	49.1	63.0	71.3
(k) Work in places where there is a lot of dusta	6.1	21.8	47.1	58.8
(m) Are HIV positive	4.9	11.5	26.1	49.8
(g) Live in overcrowded houses	22.5	37.4	44.0	48.0
(e) Live in shacks	14.9	25.3	30.8	34.3
(l) Whose immune systems/body soldiersb are weak	2.1	7.0	21.8	30.0
(o) Have spent time in prisonc	2.7	4.2	10.1	28.9
(p) Were not immunised when they were babies	5.1	6.8	9.0	24.8
(i) Visit shebeens	2.4	7.9	12.9	14.6
(a) Are poor	12.1	13.2	13.6	13.9
(j) Are promiscuous/sleep around	1.1	3.5	7.5	9.4
(c) Are unemployed	2.7	6.2	6.5	6.8
(d) Are less educated	1.7	2.9	3.2	3.5
(f) Live in RDP housesd	1.0	1.3	1.9	2.3
(b) Are lazy	1.1	1.6	1.6	1.6
(n) Do not go to church regularly	0.0	0.1	0.3	0.6

I will read out a list. Please tell us which type of people on this list you think are most likely to get TB? Do not think too long; just tell us what comes to mind

Percentages are based on the total sample, n1020. Respondents gave up to four responses

Risk factors were read out to respondents in the order of (a) to (p)

^aA brick factory is one of the few local industrial employment opportunities

^bThe terminology used in public health messages

^cA prison is located on the periphery of the city

^dSubsidised housing units built by the state for low-income households since 1994

Table 1 shows that over four rounds of responses, the highest proportion of respondents identified people who ‘drink and smoke a lot’ to be at the greatest risk of TB. With 71%, the ‘drink/smoke’ option was endorsed by far the highest proportion of voters, followed by exposure to dust (59%), HIV-positive status (50%), and poor living conditions (48 and 34%, respectively). The ‘drinkers and smokers’ option crowded out characteristics that might be regarded as markers of lower social status associated with joblessness, poverty, and lack of education; or with social deviance or a flawed character (Goffman 1963) as in the case of laziness, having served a prison sentence, engaging in promiscuous sexual relationships, or failing to attend church on a regular basis.

In later sections of the interview, a substantial proportion of respondents (47%) stated that abstaining from drinking and smoking was a recipe for success in completing TB treatment. Conversely, a similarly high proportion (47%) identified drinkers and smokers as treatment defaulters. Over a third (37%) thought the youth might be tempted to access a temporary disability grant fraudulently to support their drinking and smoking habits.

When responding to a set of attitude statements, majorities of respondents blamed drinkers and smokers for contracting TB and treatment defaulters for spreading the disease. The majority opinion was that only strict abstinence from drinking and smoking was compatible with successful treatment (Table 2). Opinions were split evenly as to whether treatment defaulters or the AIDS epidemic was to blame for the high incidence of TB in the province (Table 3).

Table 2.

Attitudes to TB and drinking and smoking

Percentage agreeing with statement	%
People who get TB through drinking and smoking get what they deserve	71.2
You will never be cured of TB if you drink and smoke	87.3
If TB patients were allowed to drink and smoke in moderation they would probably complete the course of treatment	10.3
It is mainly irresponsible people who do not take their treatment who are to blame for spreading TB	90.4

Here are some things we’ve heard people say about TB and about being treated for TB. Do you tend to agree or disagree with the following statements: Do not think too long but just tell us what comes to mind first?

n1020

Table 3.

Main cause of the spread of TB: treatment defaulters versus AIDS

Percentage agreeing with viewpoint	%
Some people say there is more TB in the Eastern Cape than 10 years ago because people infected with TB do not take their TB treatment as directed	48.9
Others say there is more TB in the Eastern Cape today than 10 years ago because of AIDS	51.1
	100.0

Here are two different opinions. Which one comes closest to yours?

n1020

The following profile of the respondents who endorsed ‘drinking and smoking as TB risk factor’ was derived from categorical data analysis using chi-square tests of association.

Respondents who named ‘drinking and smoking a lot’ as a TB risk were more likely than others to cite abstinence from drinking and smoking as the reason why TB patients adhered to treatment (49 vs. 41% among others, p < .028). They were inclined to also name HIV positivity as a TB risk factor (52 vs. 45%, p < .061) and to express fear of TB infection and HIV infection (‘two greatest dangers that could happen to you’) (TB: 37 vs. 20%, p < .000; HIV: 57 vs. 49%, p < .011). Although numbers are small, members of the ‘drinking and smoking is a TB risk’ group were more likely to agree that a person with AIDS ‘must have done something wrong and deserves to be punished’ (4.8 vs. 1.7%, p < .005). Above-average percentages in the group also agreed that smokers and drinkers who contracted TB ‘get what they deserve’ (73 vs. 66%, p < .040). However, the majority was of the opinion that the AIDS epidemic (54 vs. 44%) rather than irresponsible treatment defaulters (46 vs. 56%, p < .004) was the root cause of the surge of TB cases in the province.

Exploratory regression analyses were conducted to identify the predisposition to associate drinking and smoking with TB risk. Under scrutiny as possible demographic predictors were factors such as age, education, being a born-again Christian or regular church-goer, and experience as a treatment supporter. Consideration was also given to fear of TB and HIV infection and tendency to place blame on drinkers and smokers and irresponsible treatment defaulters for spreading TB. Initially some 17 variables were considered as possible predictors. By eliminating the most closely associated among these variables, a multinominial regression model with cumulative logit link function was produced that included only seven predictors. The influences of age and education and interaction effects were found to be negligible, so they were removed from the final model which is shown in Table 4.

Table 4.

Multinomial regression model on ‘drinking and smoking a lot’ as TB risk factor

Predictora	Betaa	Significance p<
Born-again Christian	.47	.003
Not a treatment supporter	.36	.021
Perceives TB infection as a great personal danger	.83	.000
Blames drinkers and smokers who contract TBb	.35	.055
Does not blame treatment defaulters for spreading TBb	.71	.025
Does not blame treatment defaulters who get drug-resistant TB	.66	.002
Thinks increase in incidence of TB in the province is due to AIDS (rather than treatment defaulters)c	.29	.060

Likelihood ratio Chi-square = 70.9, df = 7, p < .000

^aThe direction of the sign is given in the description of the predictor variable

^bSee indicators in Table 2

^cSee indicator in Table 3

The regression model results indicate that born-again Christians and persons who fear TB are predisposed to naming drinking and smoking as a TB risk factor. The beta values for these two predictor variables are highly significant. The constellation of beta values for the non-significant predictors in the model suggests that people who see drinking and smoking as a TB risk factor might have a tendency to be harsher on drinkers and smokers who do contract the disease (‘get what they deserve’) than on treatment defaulters. The latter are less likely to be blamed for developing drug-resistant TB or spreading disease.

To sum up results so far, the classical portrayal of stigma refers to the spoiled identity or tainted character. The focus-group results described the ‘irresponsible’, non-compliant TB patient as one who drinks and smokes. In the community survey, the association of TB with drinking and smoking seems to epitomise moral deficiency. Judging from the results of the regression analysis, ‘drinking and smoking a lot’ is more often associated with TB by churchgoers—possible representatives of the moral majority (see Lichtenstein 2008)—than by treatment supervisors. It may not be far-fetched to suggest that ‘drinkers and smokers’ is short-hand for connotations of licentiousness and social deviance. This supposition was explored further in focus-group discussions with residents from Rhini.

Results from the Follow-Up Focus-Group Study

Are ‘drinkers and smokers’ a stereotype, a form of ‘othering’ as described in the literature on disease stigma? With reference to the two-faced identity of the proverbial scapegoat outlined in the literature, do ‘drinkers and smokers’ attract rejection as marginalised ‘others’ or sympathy as lapsed members of the community?

All five discussion groups singled out the heavy drinkers (amanxila) as the people most at risk of contracting TB and spreading the disease. There was a clear distinction between social or the ‘part-time’ and the heavy drinker (nxilayo), the every-day and the weekend drinker, and those who drank at home or in the shebeen. Weekend drinkers would start drinking on Friday evening and finish on Sunday night. A woman in the middle-aged group commented that Africans, unlike some of their white South African counterparts, appeared to be prone to drinking heavily rather than in moderation. “… us blacks who drink right through until the bottle is empty, then follow more and more bottles”.

There was consensus in all groups that the drinkers outnumbered the churchgoers in Rhini. Churchgoers might also drink but their participation in church services on Sundays interrupted their drinking sessions and thus limited their consumption of alcohol. They were not regular shebeen patrons. Churchgoers, who drank, were usually classified as ‘part-time’ drinkers.

Both drinking and smoking predisposed to contracting TB, but drinkers, who were usually also smokers, were seen to be most at risk. Smoking tobacco and dagga (cannabis) increased pleasure and enhanced the effects of alcohol. Heavy drinkers drank in shebeens in a social atmosphere conducive to bonding with other drinkers, the perfect breeding ground for catching and spreading air-borne diseases such as TB (“It likes a closed space”). Drinkers shared glasses, cigarettes and zol (cannabis). Relaxed patrons under the influence of alcohol became careless in their personal habits—reference was to spitting and coughing without regard for others. Patrons who visited various shebeens spread diseases from their own neighbourhood to other sections of Rhini.

Smokers were different from other people: their addiction to alcohol and tobacco and dagga, in particular, had mind-altering effects. Their lives focused on drinking so much that they lost interest in eating properly. As a result their weakened immune systems made them susceptible to diseases. Preoccupied with drinking, they failed to take responsibility for their families and their homes. Loss of control while under the influence of drink could make laughing stocks even of respectable citizens. Under the influence of alcohol, drinkers lost respect for others and their own dignity.

Asked about their church’s views on drinking and smoking, churchgoers stated that the church preached against drinking and smoking. Drinking was a sin. Shebeen patrons tended to be portrayed as lapsed Christians. A number of the participants in the group of male churchgoers, including a pastor, reported they were reformed drinkers and smokers and spoke from their own experience. Although heavy drinking was considered socially unacceptable by the church, respondents reported that their church gave charity to drinkers’ families and offered counselling to drinkers. However, both younger and older respondents were doubtful whether drinkers could change their ways.

Focus-group discussants made a distinction between those driven to drink to relieve stress and frustration and those who drank for ‘fun’. Both were attracted to and made welcome in shebeens that did not discriminate between the employed and unemployed, educated or uneducated, rich and poor. “Those in a shack drink and forget that they are poor”. “Some would exchange their house for a shack and a drink”. “Whether you drink expensive or cheap liquor the consequences are the same—you get drunk”. The youth noted that people turned to drink after finding out they were HIV-positive because they “would think that their life is over so why not do whatever”.

All groups were divided when it came to blaming drinkers and smokers. Some argued that drinkers and smokers had freely chosen to adopt a lifestyle dangerous to themselves and the wider community. Others blamed the social conditions which drove township dwellers to drink in desperation.

Profile of the drinker and smoker. The semantic differential or ‘word game’ was used to produce a consensus within groups on the profile of the typical drinker and smoker. Table 5 presents the list of descriptors. Each group was asked to identify the most characteristic traits and behaviours of drinkers and smokers on the list.

Table 5.

Characteristic descriptors of ‘drinkers and smokers’ selected in the semantic differential (‘word game’) (English language version)

Characteristic descriptors (in bold typeface):
(1)a	Heavy drinkers b	Social drinkers
(2)	Drink in shebeens	Drink at home with neighbours
(3)	Take drugs	Do not do drugs
(7)	Do not go to church	Regular church goers
(12)	Care about things	Given up caring; beyond caring
(13)	Pick fights	Peace loving
(14)	Selfish	Generous
(15)	Sleep around	Faithful to partners
(16)	Take risks	Play it safe
(17)	Trustworthy	Cannot be trusted
(18)	Respectable	Despicable
(19)	Cursed	Blessed
Non-discriminating descriptors:
(4)	Older	Younger
(5)	Men	Women
(6)	Employed	Unemployed
(8)	Live in a shack	Live in a house
(9)	Poor	Well off
(10)	Lazy	Hard working
(11)	Dirty	Clean

^aNumbering refers to the order of presentation of descriptors

^bDescriptors in bold type were selected as characteristic of drinkers and smokers by all five discussion groups or at least four in five

The character sketch of the typical drinker and smoker below is based on the consensus votes of the five groups of discussants. Results received from the five groups were almost identical.

Distinctive descriptors:

Drinkers and smokers are mainly heavy drinkers, who drink in shebeens and take drugs. They do not go to church. They are generally selfish, they pick fights, and have given up caring. They cannot be trusted as they take risks and sleep around. They are despicable and cursed.

Non-distinctive descriptors:

Drinking and smoking cuts across the social divide: Drinkers and smokers include both young and old, rich and poor, people with paid work and the unemployed, people who live in solid houses and in shacks. Drinkers and smokers include the hard working and the lazy, the clean and the dirty.2

The groups elaborated on their choice of descriptors. With reference to character flaws, heavy drinkers were projected as selfish because they thought only of their own needs. If you drank heavily in a shebeen you were despicable. “Even if it’s a prominent figure like a priest, if he goes there he has lost his dignity and respect”. Heavy drinkers were not faithful to their partners and could not be trusted. Liquor “revives your sexual feelings” and you engaged in risky behaviour while drunk. Drinkers and smokers might even resort to stealing to pay for their addiction. The consequences of drinking and smoking were always bad so heavy drinkers were ‘cursed’. All groups described drinkers and smokers as people who were aggressive and picked fights. However, the shebeen patrons prided themselves on not ever holding a grudge against anyone involved in a drunken fight as they could never remember the details the next day. “We were drunk yesterday, but now we are not drunk and we don’t know what happened yesterday”.

The discussion groups were asked if they wished to add further descriptors to the list provided in the word game. All but the male churchgoer group declined. Thus, it can be assumed that the semantic differential gives a fairly accurate portrayal of the stereotypical drinker and smoker. Worth noting is that before completing the semantic differential, focus-group participants had already referred to several of the 19 descriptors in the list.

Members of the last two groups interviewed were asked to identify the four most apt descriptors. The youth cited the descriptors ‘heavy drinkers’, ‘drink in shebeens’, ‘take drugs’, and ‘take risks’. The male churchgoers focused on descriptors of flawed character: ‘Untrustworthy’, ‘cursed’, ‘take risks’ and ‘pick fights’. In addition, they added a further descriptor, that of injubaqa, a ‘rowdy hooligan’, who is beyond being helped. “They are the ones whose families have given up on them”.

TB and HIV Risk Behaviour Associated with Drinking and Smoking

Spontaneous mention was made in all five group discussions of the risk of HIV-infection associated with drinking and smoking.

In many cases, the risk of TB and HIV infection was seen to be inseparable. The middle-aged group recommended counselling for dagga smokers to “let them know what this does to them like killing their body soldiers and that it weakens them and that it also can cause TB or HIV/AIDS”.

Social distancing or ‘othering’ was most evident when discussing the twin risks of TB and HIV-infection associated with drinking and smoking. When commenting on the difference between drinkers and smokers and other people, a member of the youth discussion group identified drinkers and smokers as the carriers of disease: “the ones who bring us ‘things’ while we stay at home”. People at home who “never had a disease like TB in the first place or HIV/AIDS” get infected by “the father who’s a drinker in the shebeen”.

‘Bad habits’ were typically associated with frequenting shebeens. Shebeen patrons lost many of their inhibitions when drunk. Others purposely sought sexual adventure.

In former times, respondents observed, heavy drinking was restricted mainly to men but that had changed. When discussing the words in the semantic differential, the male churchgoers noted that “even the clean ones become dirty because they end up sharing partners the same day and have no time for washing”.

The church women stated that drinkers ended up having intercourse with strangers which they would never do when sober. “You get involved with a stranger and have intercourse hence get positive in different ways, e.g., have igcushuwa (a sexually transmitted disease). …AIDS is the recent one, but it also comes in with that. We non-drinkers, we won’t get that as a result of liquor”.

Similarly, the church men detailed the many dangers of drinking liquor:

“When drunk you’d only think of a woman whom you hardly know and perhaps they already are infected with the virus”.

“In the shebeen they’ll talk to a girl and later leave with them but only realise the following day that… the person… has ‘this thing’ [referring to HIV/AIDS] or the person… was admitted to Themba hospital [the local TB hospital] and it would be very late by then”.

The shebeen patrons in the follow-up study confirmed that they went looking for casual partners in shebeens. “That’s one reason why we go to the taverns. As men we know there are girls to meet there”. They agreed that drinkers and smokers were mainly responsible for spreading TB. Apart from being careless about coughing and spitting, they also “drink and go to bed with different partners”. “You get infected with diseases like HIV and TB from sharing partners”.

Discussion and Conclusion

An unexpected finding in our community survey prompted us to inquire in greater depth into perceptions of the typical South African ‘drinker and smoker’. Survey respondents identified ‘drinkers and smokers’ as being most at risk of contracting TB (Table 1). Drinking and smoking as a TB risk factor was endorsed by a significantly greater proportion of survey respondents than other factors commonly associated with the illness, such as poverty and crowded living conditions.

The regression analyses applied to the survey data identified churchgoers and born-again Christians as the social groupings most likely to associate TB risk with drinking and smoking. Fear of TB and HIV/AIDS (‘the greatest danger in life’) were further significant predictors. Survey respondents who linked drinking and smoking to TB risk were more likely to blame drinkers and smokers (‘get what they deserve’) for spreading the disease than treatment defaulters generally.

The above findings suggested that ‘drinkers and smokers’ might constitute a stereotype, a form of ‘othering’ as described in the literature on stigma, that serves to distance the community from contamination with moral pollution and a health threat. The pilot study that informed the sample survey had already suggested that drinkers and smokers might belong to a maligned or marked group akin to the scapegoat that is burdened with the ill in society. A second set of follow-up focus-group discussions was conducted to shed more light on this idea.

The Drinker and Smoker as Stereotype

The amount of alcohol consumed per South African adult is among the highest in the world (Parry 2005). Focus-group discussants in the follow-up study described drinkers and smokers as contaminated on the ‘inside’ through alcohol addiction, smoking, and drug-taking, which makes them prone to disease. Drinking in a confined space and sharing drinking vessels and smokes puts shebeen patrons at risk of air-borne TB. (“Most of the time they are in a closed space in big numbers which makes it easy for the air-borne TB to spread evenly”.) The shebeen was also described as a high-risk HIV transmission site. All five focus groups at some point in the discussion spontaneously referred to drinkers and smokers losing their self-control and engaging in risky sexual behaviour linked to HIV-infection. This finding compares with the shebeen behaviour observed by Morejele et al. (2006). The results of the semantic differential confirmed that licentiousness was associated with drinking and smoking (see Table 5). Attributions of ‘cursed’3 and ‘despicable’ conforms to the classic definition of stigma as the flawed character.

The Drinker and Smoker as Scapegoat

Drinkers and smokers are ideal candidates to serve as scapegoats in a community that is affected by the second TB epidemic. According to the literature the scapegoat must bear the traits of mainstream society members as well as ones that mark them as outsiders (Gilmore and Somerville 1994). The follow-up focus-group study suggests this is the case. Discussants vacillated between describing smokers and drinkers as different or as ‘one of us’. There was no consensus among focus-group discussants on whether drinkers and smokers who contracted TB or HIV were blameless victims of poverty (Farmer 1996; Erstad 2006) or had chosen a risky lifestyle of their own free will. The youth made a distinction between ‘fun’ drinkers and ‘stress’ drinkers, who drank to forget their problems. The semantic differential outcome suggested that drinking and smoking, including ‘stress’ drinking, is not exclusive to the poor and unemployed; drinkers and smokers included a mix of generations and social strata. The churchgoing discussants tended to regard drinkers and smokers as lapsed Christians. Drinkers and smokers were often portrayed as having strayed from the exemplary moral behaviour prescribed by the church. According to our discussants, the local churches condemned drinking and smoking but extended charity to drinkers and smokers in the hope of winning ‘sinners’ back to the fold. However, it was considered unlikely that drinkers would change their lifestyle, even with the support of others.

Inxila is the derogatory label for heavy drinking. When drunk, shebeen patrons were described as undignified, undisciplined and a threat to the health and moral order of society. The shebeen-patron discussants painted a similar self-portrait of the drinker—interestingly, their self-deprecating auto-stereotype matched the one produced by the other discussion groups completing the word game. Following Smith’s (2007) distinction between popular presentations of disease as ‘stigma’ or ‘challenge’, these findings suggest that there is prejudice against drinkers and smokers whose plight is at the same time regarded as a challenge that needs to be addressed by mainstream society.

Church and Shebeen

The church and shebeen represent core social institutions around which township social life is organised. Shebeens may be as accessible as the many churches serving the population of Rhini.4 There are an estimated 181 licensed shebeens/taverns and a ‘dark figure’ of a further 40–50 liquor outlets in Grahamstown East/Rhini5 according to a recent count. If the church and the shebeen represent the sacred and secular poles of township morality, the two institutions also share a number of commonalities. Historically, the African churches and the township shebeen share a proud track record of defiance. The older generation of charismatic churches of Africa were established as independent centres of worship (Centre for Enterprise and Development 2008). It is telling that many go by the name of ‘Ethiopian’6—the only African country that never experienced colonial rule. With reference to shebeens, Parry (2005: 426) notes that the growth of illegal liquor outlets in the second half of the twentieth century served as a form of resistance to apartheid policies instituted to repress the black majority, as did the destruction of the government-run beerhalls in the 1970s. Both church and shebeen offer fellowship and refuge to people experiencing personal and social problems. Hymn-singing and alcohol have the power to transport to another sphere where everyday anxieties fall away. Similarly, a focus-group discussant in the follow-up study described the shebeen as “a place where friends become enemies, …people who are drinkers share some kind of love and care”.

Unsurprisingly given their commonalities, the two institutions compete with each other for a following. Focus-group discussants estimated that drinkers and smokers might outnumber churchgoers and the drinking lifestyle was gaining popularity among the youth. Some members of the focus group drawn from churchgoers claimed they were reformed drinkers while others in the group of shebeen patrons said they were still churchgoers in spirit: “On Sunday, drinkers always sing the church songs which means they do think about church but the drinking overcomes them. …You did not go to the church but the heart is at the church”. The alternative identity of shebeen patrons is suggestive of the ‘covering’ strategy discussed in the stigma literature (Goffman 1963; Daftary et al. 2007; Collins et al. 2008).

The Cleansing Function of Abstinence

Our findings from the follow-up study suggest that churchgoers and drinkers and smokers can be regarded as opposite poles on a morality continuum. Township folk are offered only two choices on how to conduct their lives: you either drink and smoke or you abstain, there is no in-between.7 According to this belief, the church represents salvation; the shebeen the work of the devil. Drinkers must be saved from damnation.

Clues to why drinkers and smokers are maligned in a society that knows no moderation in its drinking habits, is provided by Steinberg’s (2008) insightful inquiry into resistance to HIV testing in Eastern Cape villages. His research found that church people recommended a cleansing process, that involves abstaining from drinking and smoking, to attain purity in body and soul, and to redeem society from ill in the time of HIV/AIDS.

Steinberg was puzzled that a self-appointed HIV-counsellor, a devout Christian, who accompanied many of the people in her village to the health clinic to check their status, advised her charges to quit drinking and smoking before going on antiretroviral (ARV) treatment (2008: 167–181). Steinberg established that there was no sound medical motivation to prohibit all ARV users from drinking. Although drinking while on medication for chronic illness was undoubtedly unadvisable, it did not impede the drug’s work nor aggravate the side effects (Steinberg 2008:180). One of the medical experts Steinberg consulted on the origin of the injunction against drinking and smoking for patients on anti-retroviral treatment, spoke of a prejudice against drinkers. He intimated that AIDS counsellors were creating an exclusive church (sic) around ARVs that excluded many, particularly unemployed men, from benefiting from ARV treatment. According to the medical expert, “it is important for people to believe they are part of a club, perhaps even a church. It is a sense of ubuntu.8 If you want to create a club, you must create rules” (cited in Steinberg 2008:181, first emphasis added). Steinberg speculates about the significance of these rules. Apparently, it is an article of faith among many South Africans that those who smoke or drink while taking ARVs will die. He reasons that to belong to the ‘church’ created by the treatment campaigners one “must submit one’s powers of discipline and restraint to the public test” (Steinberg 2008:181). Thus treatment is a redemption of sorts as is “cleansing your contaminated body of smoke and drink” (Steinberg 2008:182). The sharp accusation of those who drank and smoked while on ARV treatment is akin to announcing that they died of contamination. Steinberg notes that “perhaps it is hard to cleanse and to heal in the absence of an enemy or a scapegoat”. (Steinberg 2008:182, emphasis added).

Our study similarly encountered difficulties in disentangling the moral from the bio-medical understanding of drinking and smoking. Drawing on Steinberg’s interpretation of the prohibition on drinking and smoking for people living with HIV, we conclude that drinkers and smokers may well serve as the scapegoat that will protect the Rhini community from contamination with the ‘second’ TB epidemic. Alcoholism presents a challenge to moral society in all times but even more so in the time of AIDS. It therefore lies close at hand that the church should take strong measures lest the licentiousness of drinkers and smokers pollute fellow congregants and expose them to what has been labelled a ‘dirty’ disease (Balabanova et al. 2006; Lawless et al. 1996).

Conclusion

To sum up, to date little systematic research has been conducted on the impact of smoking and drinking on health-seeking behaviour related to TB. Our study of attitudes to drinking and smoking as perceived TB risk factor sought to fill the gap. The pilot study confirmed that the stigma associated with AIDS affects TB sufferers and acts as a barrier to the detection and treatment of TB (Møller and Erstad 2007). However, results from the community survey suggested that drinkers and smokers rather than treatment defaulters might have become scapegoats that carry the stigma associated with a risky lifestyle that can spread both TB and HIV/AIDS. Although limited in scope, the follow-up focus-group study supported the thesis that drinkers and smokers are shorthand for licentiousness that presents a risk to the moral order in society. Heavy drinking and smoking was associated with shebeen drinking and the risks of contracting HIV as well as TB.

Smoking and drinking are moral risk-factors visible in the community, and thus easy to use in scapegoating. Poverty as a structural risk-factor (Farmer 1996; Erstad 2006) was not mentioned as frequently, possibly because the participants in this study are living in poverty themselves. Although the literature has failed to establish unambiguously the bio-medical connection between drinking and smoking and TB, our study has confirmed the social connection between drinking and smoking and TB and HIV/AIDS. Shebeen patrons are portrayed as the prototypes of the drinkers and smokers whose behaviour is condemned by moral society in a time when the ‘second’ TB epidemic poses a major threat to township society. Our study, along with that of other research in the Eastern Cape, suggests that a cleansing process that involves abstinence from smoking and drinking is seen as the solution to achieve salvation from the epidemic.

Our conclusions on the stereotyping of drinkers and smokers in the time of the twin epidemics of TB and HIV/AIDS are based on a single community survey with a limited sample size. It is possible that the drinker and smoker stereotype is a product of the South African social setting in which poverty and discrimination has driven the politically and economically oppressed to find solace and an independence of spirit in either the church or the shebeen. It is therefore recommended that further research be conducted in a range of settings using the bio-medical or sociological approach or a combination of both.