Epidemic disease, and AIDS in particular, is a disease of the body – but it is the presenting symptom. The manifestations of AIDS, illness and death, reveal the fractures, stresses, and strains in a society. This chapter shows that while at the most proximate level the chance of HIV transmission may depend on biological determinants, there are other factors that need to be considered, in particular social and economic poverty and inequality.
The overview of global epidemiology in Chapter 1 showed a range of epidemics. In a few places prevalence has peaked and fallen; in others it has risen to unexpected levels and remains high. There are settings where all the factors that would facilitate HIV spread seem to be in place yet there is no epidemic. While biomedical factors are critical, it is ultimately behaviours that will determine the shape of the epidemic. These depend on social and environmental factors – the position people occupy in society, their economic status, and how they are perceived and value themselves. Where vulnerabilities converge, we see the most serious epidemics.
If we can explain existing epidemics, can we predict new ones? Where prevalence has fallen, can we understand what happened in order that it can be replicated elsewhere?
Infections and epidemics don’t happen randomly. Some diseases are limited to certain geographical environments – for example, to catch malaria a person must be in a malarial area and bitten by a mosquito carrying the parasite. People must be exposed to the pathogen. Even then, for an infection to take hold, the immune system must be unable to resist the disease-causing organism. This is true of all infectious illnesses: we see it every year with the common cold, when most people are exposed to the virus, but some individuals manage to stave off infection while others fall sick. Thus an individual’s immune status is important in determining whether or not they are infected and how severely they are affected.
Drivers of disease are mostly social and economic. For example, living in poorly ventilated, crowded rooms increases the risk of exposure to and of contracting TB. Being undernourished and/or lacking vitamins and micronutrients will increase susceptibility to a plethora of illnesses and means people are sicker for longer.
Disease, globally and nationally, flourishes where there is poverty. In the rich countries of the world, the greatest disease burden is found among the poorer populations: those who are ill-nourished, poorly housed, and less well educated. With regard to nations, broadly speaking it is in the poorer ones where disease is more likely to thrive. Critical factors that can mitigate against poverty are social services and public health. Thus Cuba, or Kerala state in India, have healthier populations than their richer more inequitable neighbours.
On occasion there is a convergence of vulnerability that results in epidemic outbreaks. For example, prior to 1991, cholera had not been seen in the Americas for 100 years. In that year an outbreak began in Peru, after a ship in Lima harbour pumped its bilges of water that was contaminated by the Vibrio cholerae bacteria. The disease took hold in the city’s slums, and then spread from slum to slum across the Americas. Obviously ships had discharged contaminated water before, both in Lima and other ports. However, in 1991 the slums had grown rapidly, the inhabitants were the victims of a decade of economic crisis which resulted in falling incomes and increased inequality. People were poorly nourished and lacked access to basic infrastructure including water, sanitation, and health services. The result was a regional cholera epidemic.
In order for a person to become infected, they must be in contact with HIV with sufficient exposure for the infection to take hold. Once contact has occurred, biomedical factors are the key determinant of whether or not a person will be infected.
The most important biological influences are the virus sub-type and the genetic make-up of those exposed. Some sub-types are believed to be more infectious than others. This may be partly why Southern Africa, where sub-type clade C is found, has such a serious epidemic. The genetics may be important, operating at the individual or population level, making some individuals or groups more or less susceptible. This is controversial, as it is sometimes interpreted as a form of ‘genetic determinism’ instead of the natural ‘reality’ that results from the diversity of humankind. Evolution is not ‘kind’ or ‘cruel’, although we may wish to construe it as such. The importance of the virus type and genetics are areas of continuing scientific research.
The stage of infection is crucial. For several months after infection, there is an intense battle between the immune system and the virus. During this period, the semen, vaginal fluids, and blood contain many virus particles, increasing the chance of infection for sexual partners and people who share injecting equipment. There is then a period when the body rallies and the viral load is low. As the infection progresses, it will slowly climb and the CD4 cell count will fall, as shown in Figure 5 of the previous chapter.
Once the epidemic gains a hold in a society, it has a built-in momentum. The more people with early-stage infection, the greater the chance of someone having sex with such a person and being infected, so that a vicious cycle develops.
The virus has to breach the natural defences of the body, the skin or mucous membranes. Risk is higher for women as semen remains in the vagina after unprotected intercourse. This partly accounts for the greater number of women infected in heterosexually driven epidemics. The danger is increased by tearing in the vagina, which may occur during abusive sex or rape, especially in younger women whose vaginas are not mature, and thus interventions that delay sexual debut reduce transmission. Condoms provide a barrier, but are not female-controlled.
Sexually transmitted infections (STIs) are an important biological co-factor. Those that cause genital ulcers such as herpes, chancroid, and syphilis create a portal for the virus to enter the body, and at the same time the presence of the cells HIV seeks to infect, CD4 cells and macrophages, is increased. In a person with an STI, the number of virus particles released into blood, semen, and other body fluids increases, even if the infection is asymptomatic. An HIV-infected person is more likely to be infected by STIs and the severity and duration of these infections will be increased.
After sexual transmission, the next most important route of HIV infection is mother-to-child transmission (MTCT) with infants exposed through birth or breast-feeding. The viral load of the mother influences the probability of infection of the infant – the higher the load, the higher the risk. However, if a women has advanced disease, the chance of falling pregnant and carrying a child to term is decreased.
Other biomedical drivers include the use of unsafe blood and blood products and nosocomial (hospital-acquired) infections. In India in 2001, of the risk/transmission categories listed by the National AIDS Control Organization, 4.1% of AIDS cases resulted from contaminated blood or blood products. The practice there of paying blood donors may result in contaminated blood being collected, paid donors being more likely to live on the margins of society and to be infected. In China, it was estimated in 2002 that 9.7% of HIV cases were transmitted through illegal and unsafe practices associated with blood plasma collection.
While ‘nosocomial’ usually means infections acquired in hospital, with regard to HIV/AIDS it is taken to mean all infections transmitted in health care settings. If equipment is not adequately sterilized, then there is a danger of patient-to-patient transmission. Health workers are at risk through accidents involving body fluids such as needle stick injuries. All those caring for AIDS patients, including in the home, face some degree of danger and this rises if carers don’t have adequate protective equipment such as gloves. Sharing drug-injecting equipment is an efficient way of spreading HIV, and this is the main driver in some settings.
One under-researched area is the effect of ill health from other causes on HIV transmission. There is evidence to suggest that any other infection will cause the viral load of HIV to rise rapidly and remain high for some time afterwards. For example, research shows that when an HIV-positive person has malaria the amount of virus in the blood increases tenfold, and thus such a person will be more infectious to his or her sexual partners. They may not want to have sex while they are sick, but when they recover their sex drive will return and infectivity is still high. Research in Kisumu in Kenya estimated that 5% of adult HIV infections were linked to malaria, and conversely, HIV infection increases susceptibility to other diseases, the Kenyan research also suggesting 10% of malaria cases were due to HIV. An HIV-positive individual with any other infection is likely to be sicker for longer and may be more likely to die.
Whether or not a man is circumcised, a biomedical solution, is important, as was discussed in the previous chapter. The routine offering of circumcision for male infants delivered in health care settings makes sense, but will take 20 or more years to impact on HIV prevalence. Had this happened in 1985, we would be reaping the benefits today.
In order for biomedical factors to come into play, a person has to have sex, or share needles with someone who is infected. There are a range of behaviours that increase risk. The AIDS epidemic has taught us unexpected lessons about human sexuality. The frequency of sexual intercourse does not vary greatly from country to country. There are a wide and intriguing variety of sexual practices, most of which are harmless and many are considered ‘normal’. The behaviours that facilitate the spread of HIV are complex and dynamic, but global data suggest it is common for people to have more than one partner in their lifetime.
If someone does not have sex or sticks to one uninfected partner, then that person won’t be sexually exposed to HIV (or any other sexually transmitted infection) provided their partner is also faithful. This applies in all sexual relationships – hetero- or homosexual. In societies where polygamy is practised, then as long as all parties are faithful, the same protections apply. Early AIDS prevention posters which, in most countries, said unequivocally ‘Stick To One Partner’ had to be adapted for Swaziland where polyandry is accepted and the king also has many wives – here, the posters had the less than catchy message: ‘Be Faithful in Your Polygamous Family’.
Key behavioural factors are the age of sexual debut, sexual practices, number of partners, frequency of partner exchange, concurrency of partners, and mixing patterns including intergenerational sex. The younger a woman begins penetrative sex, the greater her risk of infection due to the danger of tearing of the vagina. The age of sexual debut is determined by her behaviour and those of her partners, and is influenced by social norms. Globally, data suggest that females have sex earlier than men, but trends for age at first sex are not clear. A meta-analysis of global sexual behaviour concluded trends towards earlier sexual experience are less pronounced than supposed. In developing countries, sexual activity is happening later, but prevalence of premarital sex increases if marriage is postponed. The data show the median age for first sexual intercourse for males was 16.5 in Kenya, Zambia, Brazil, Peru, and Britain. In the USA, it was 17.3. The oldest was 24.5, in Indonesia. For women, median age at first intercourse was lowest in a number of African countries: 15.5 in Ethiopia, Mozambique, Côte d’Ivoire, and Cameroon; the oldest was 20.5 in Rwanda. In the UK and USA, it was 17.5.
Data on sex and sexuality
We are all intrigued by sex and sexual behaviours, but collecting this information is complex. The first major study was by Dr Alfred Kinsey of Indiana University. The Kinsey Reports comprise two books on human sexual behaviour: Sexual Behavior in the Human Male (1948) and Sexual Behavior in the Human Female (1953). When released, this research was controversial, not just for the subject matter, but because it challenged many beliefs about sexuality, including the ideas that heterosexuality, faithfulness, and abstinence were ethical and statistical norms.
A basic problem with sexual behaviour data is that they are self-reported. This means that the data are subject to bias. Most commonly, men over-report and women under-report partnerships and frequency of intercourse. This was well described in the title of an article on sexuality in Tanzania: Secretive Females or Swaggering Males? The meta-analysis published in The Lancet in 2006 shows how little data there are on sexual behaviour, and even less longitudinal information. One source of data is the Durex Global Sex Survey carried out annually since 1996. In 2005, it looked at 41 countries. It is web-based so has huge biases, but gives comparative and longitudinal data.
The question of sexual practices receives more salacious press than is deserved, although it is the area about which we know least. As far as HIV is concerned, some potentially harmful practices are widow inheritance, when a woman is ‘inherited’ by her deceased husband’s brother, and the practice of ‘dry sex’, the use of herbs or other agents to dry out the vagina, which some believe increases (the man’s) pleasure during sex, but the range of practices is immense. It is necessary to be open-minded, identify those that increase risk, understand how they do this, and find out what can be done about them.
The number of sexual partners per se seems less important. Men in Thailand (where the adult infection rate is 1.4%) and Rio de Janeiro (adult infection rate in Brazil is 0.5%) were more likely to report five or more casual partners in the previous year than men in Tanzania, Kenya, and Lesotho (where adult infection rates were 6.5%, 6.1%, and 23.2 % respectively). Adult prevalence in Zambia is 17%, but the 2005 Zambian Study found that over 97% of married women and 90% of married men indicated they had no non-marital partners in the previous year. The same survey found 26% of non-married people reported one ‘non-regular partner’ but only 4% reported two or three.
6. Needle-sharing, a high-risk behaviour
People in industrialized countries do not have significantly more or fewer partners in a lifetime, but their tendency is for serial monogamy. This means that they enter relationships which are maintained for months or years. The relationships involve a degree of commitment, and may be legally recognized as marriage or civil union. Serial monogamy traps the virus within a single relationship and so is not high risk for HIV transmission. The danger of infection increases when people have ‘non-regular’ partners or affairs.
While frequent partner change is hazardous, it is not common anywhere. The greatest risk is concurrency of partnering, when people have more than one partner and the relationships overlap for months or years. Writing in The Lancet in 2004, Halperin and Epstein noted that because infectivity is higher during the weeks and months after infection, concurrent partnering greatly exacerbates the spread of HIV and may be one of the main drivers. When a person in a network of concurrent relationships becomes infected, everyone is at risk. Mathematical models comparing serial monogamy and long-term concurrency showed that, in the latter, HIV transmission would be more rapid and the epidemic ten times greater.
Commercial sex, whether heterosexual and homosexual transactions, is potentially risky both for sex workers and their clients. In many settings, in the early years of the epidemic commercial sex workers were ‘core transmitters’. A modelling exercise in Nairobi illustrated this. It assumed that 80% of sex workers were infected and had four clients per day, and 10% of men were infected and had four sexual partners per year. If women sex workers increased their clients’ condom use from 10% to 80%, that was estimated to prevent 10,200 new infections. Increasing condom use among the men to 80% would avert only 88 infections. In Thailand, the early epidemic was spread by sex workers, but the ‘100% condom campaign’, making condom use in brothels mandatory, was effective at bringing HIV spread under control. In Durban, research in the early 1990s found brothel-based sex workers (who used condoms) had negligible HIV infection.
Mixing patterns make it possible for an infection to be carried from one part of a country to another, across national borders, or to be introduced into previously closed circles. Here paths for transmission include both sex and drug use. For example, an oil worker who becomes infected in, say, Nigeria can carry the disease to his home country, then to, say, Indonesia in a matter of days.
7. Warwick Junction in Durban, South Africa: where thousands of street traders serve many more thousands of daily commuters, and where HIV infection is high
A central Asian drug user can fly to any European capital in hours. With such mixing there is also the danger of re-infection and of new strains being created.
Mixing not only takes place across geographic regions but across age groups. Intergenerational sex, usually where men have younger female partners, is common in many societies. In countries where there is a heterosexual epidemic, the pattern is for women in their teens and twenties to have much higher prevalence than their male contemporaries. This is because they are having sex with older infected men, and sometimes this is transactional – for money, food, transport, and school or university fees.
The use of condoms is also a ‘behaviour’. Correct, consistent condom use reduces the chances of HIV infection. When condoms were used in risky settings – among young people in Europe and the USA, and in brothels in Thailand – they prevented HIV spread or turned the epidemic around, but it is difficult to achieve consistent use other than in commercial and casual sexual encounters, and women may not have the power to insist their partners use condoms.
How people behave may determine their risk of infection, but behaviours result from the environment in which people live and operate. This milieu is, in turn, a function of local, national, and international factors; economics, politics, and culture. These are complex and varied and how we view them depends on our own values, backgrounds, and disciplines. The way the epidemic is influenced by these determinants is best illustrated by examples.
In Southern Africa, development of the mines and industry required a large workforce. The dominance of capitalism meant wages were tightly controlled. The colonial history and, in South Africa, subsequent apartheid legislation resulted in black labour being most exploited. Apartheid imposed strict controls on where black people could live and work and meant many South Africans were classified as migrants, effectively foreigners in their own country. Huge numbers of men travelled to work in the mines, factories, and on the farms. Foreign migrant miners were drawn primarily from Malawi, Lesotho, Botswana, Swaziland, Mozambique, and Namibia, and in the 1970s there were close to half a million foreigners employed on contracts in South Africa. In 1985 nearly two million black South Africans were classified as migrants. These people lived apart from their families, in hostel accommodation, and had to return home between contracts.
The effects of this dislocation and disempowerment have been well documented. When people are placed in circumstances in which they cannot maintain stable relationships, life is risky and pleasures are few and necessarily cheap, then sexually transmitted diseases will be rampant. This was true for all migrants. For migrant miners, their work was particularly dangerous, their control over most aspects of their lives was minimal, and they were disempowered in many respects. However, they had regular incomes. When gender inequality and the extreme poverty in the surrounding communities is considered, an ideal setting had been created for the spread of sexually transmitted infections.
During the 1980s, four large surveys were carried out to establish if HIV was present in South African populations outside of known high-risk groups. HIV was found by only one survey. The few cases were Malawian miners. Migration to and within South Africa created the perfect environment for the spread of HIV, not only in labour centres but in the migrants’ home communities. The fracturing of families, changing gender dynamics, and increased poverty were major causes of the high levels of HIV.
Similar stories can be told of former communist countries. The collapse of communism was not good news for millions of citizens of the Soviet Union and Eastern Bloc. The system had provided many benefits, citizens were assured of employment, education, housing, health care, and even holidays; basic needs, and more, were met. The collapse of these economies has also been well documented. In the Ukraine, the per capita GDP (in purchasing power parity) fell from US$ 6,372 in 1990 to a low of US$ 3,194 in 1998. In 1994 alone, GDP declined by 22.9%. From having full employment, by 2000 the number of unemployed had reached close to three million, 12% of the economically active population. The pattern of societal collapse is seen across the region. Alcohol abuse was always common, but intravenous drug use increased dramatically, especially among the dispossessed and lost youth. The epidemic here has been driven by drug use – but this in turn is the result of economic and social disintegration and the consequent blow to the morale, hopes, and dreams of the younger generation.
In China, the epidemic of HIV among people selling blood, described in Chapter 1, has its roots in the political economy of the country. The peasants from whom the blood was collected are among the poorest, and selling blood is a survival strategy. The collapse of state medicine and introduction of fees meant that a ready, and unregulated, market existed. Ultimately, embracing the globalized economy will have partly driven China’s epidemic.
Gender relations shape risk and behaviours. A woman’s biology puts her at greater risk. Of crucial importance is the lack of power, and violence against women. Girls often feel pressured or forced into having sex. The Reproductive Health Research Unit Survey in South Africa reported that 28% of females and 16% of males aged 15 to 24 either ‘did not want’ or ‘really did not want’ their first sex. In Zambia, the Sexual Behaviour Survey in 2005 found 15.1% of females reported they were forced to have sex, and in 67.5% of cases it was by their husbands/boyfriends.
Some customs encourage early marriage and pregnancy; the marriage of young women to older men; and unequal partnering. These accept male dominance and female subservience. Globally, social norms emphasize female chastity and turn a blind eye to male promiscuity. In most of the poorer world, women are economically dependent on men, and sex work is the most extreme manifestation of this. Enabling female control of reproductive health would help the response to HIV/AIDS.
The relationship between HIV/AIDS and poverty is complex, both at the individual and national level. Botswana is, by most standards, a wealthy country. With a per capita income of US$ 4,372 in 2003, it has the third highest income in sub-Saharan Africa; Senegal by contrast has an income of just US$ 634 per capita. The prevalence rates among adults aged 15 to 49 in these countries are 24.1% in Botswana and 0.9% in Senegal. It would seem that simply being poor does not determine a country’s HIV prevalence; rather, what is crucial is societal equality.
At the individual level, data are also less than straightforward. Demographic Health Survey (DHS) data from Burkina Faso, Cameroon, Ghana, Kenya, Lesotho, Malawi, Tanzania, and Uganda show that, contrary to evidence for other infectious diseases and intuitive expectations, HIV prevalence is not disproportionately higher among poorer adults in sub-Saharan Africa. Indeed in all countries, except Ghana, the trend is the wealthier the person, the more likely they are to be HIV positive. When confounding and mediating factors (such as mobility and urban residence) are controlled for, wealthier adults are at least as likely to be HIV-infected as poorer ones. This is particularly marked for women. In Kenya, 3.9% of the poorest quintile of women were infected; for the richest, it was 12.2%; for men, these figures were 3.4% and 7.3% respectively.
While, at the most proximate level, the chance of HIV transmission may depend on biological determinants, they are only a part of the picture. Developing drugs, vaccines, and microbicides, circumcising men, and putting people on treatment are technical and biomedical responses. Unfortunately, this disease does not lend itself to simple technical solutions.
The real challenge is to change behaviours to reduce risk. Behaviours can be modified, and the evidence suggests that there are a few key interventions that would have a significant impact on the progress of the epidemic. These include reducing concurrent partnering and delaying sexual debut for young women. Beyond this are the messages that have been used since the early days of the epidemic: abstinence, fidelity, and condom use.
Both behaviours and biomedical factors are determined by how a society operates at the macro level: the culture, politics, and economics. These factors are crucial, and most important are gender relations and income equality. The central issue is how people treat and see one another. A society in which people respect the views and choices of others is one in which unsafe sex is less likely to occur.
Preventing HIV transmission requires a greater understanding of the determinants of the epidemic. Unfortunately for the 40 million people currently infected, the 20+ million who have died, and their families and communities, prevention has not worked. We need to understand how AIDS and its impact will work its way through society. This is the theme of the following chapters.