Sex and young people in urban slums: exploring the material deprivation and sexual risk nexus in Malawi and South Africa

• Main Author: Kamandaya, Mphatso
• Format: Others
• Language: en
• Published: 2016
• Online Access: http://hdl.handle.net/10539/21527

Thesis submitted for the degree of Doctor of Philosophy School of Public Health Faculty of Health Sciences University of the Witwatersrand, Johannesburg 26th April, 2016 === Rationale Young people aged 15-24 years account for nearly half (41 percent) of new HIV (Human Immunodeficiency Virus) infections in people over 15 years globally and 52 percent of this number occurs in the SEA region (UNICEF, 2011). Unprecedented urban growth in cities is occurring in this region and approximately 75 percent of the urban population are young people (UN-HABITAT, 2008). This urban growth has been accompanied by urbanisation of poverty in the growing urban slums or urban informal settlements1, where approximately 70 percent of the urban residents live (UN-HABITAT, 2010b). The UN Human Settlements Programme has identified five characteristics that define an urban slum, namely poor structural quality of housing, overcrowding, insecure residential status, inadequate access to safe water, inadequate access to sanitation and infrastructure (UN-HABITAT, 2010b). Research from some countries in the SEA region has shown that the HIV prevalence and incidence in urban slum populations is double (or more) that in the non-urban slum population of the same city (see, for example, Thomas et al., 2011; Madise et al., 2012; Rehle et al., 2007; Kyobutungi et al., 2008). This difference might be partly due to HIV-related sexual risk behaviour of young people living in urban slums (Mmari and Astone, 2013; Unger, 2013). Given that urban slums are home to a growing number of young people in the SEA region, there is the need for more innovative research on context-specific factors associated with HIV-related sexual risk behaviour to inform a specific combination prevention framework that will impact strongly on the HIV epidemic. 1 In this thesis, I will focus on the term urban slum rather than urban informal settlement as per definition by the United Nations that I have given. Malawi and South Africa are helpful examples of the currently urbanising and the most urbanised countries in the SEA region with some similarities between urban slum contexts, but also with large variations due to the historical, political and economic situations of the respective countries. For example, existing studies in Malawi (Mkandawire, 2011; Mkandawire et al., 2011a; Madise et al., 2007) and South Africa (Tenkorang et al., 2011; Burns and Snow, 2012) have highlighted that young people living in urban slum settings are at high risk of HIV acquisition. In addition, these studies highlight similar challenges faced by the growing numbers of young people living in urban slum settings, which include unemployment, poor access to education, health care, food, and continue to experience high levels of sexual abuse (Mkandawire, 2011; Tenkorang et al., 2011). Although these studies have provided important information, there are significant gaps in knowledge regarding context-specific (or local) indicators of disadvantage associated with sexual risk behaviour. Specifically, existing studies have focused on income-related measures of poverty (for example, low levels of income, unemployment, and education) to predict sexual risk2. One of the strongest critiques emerging of income-related measures of poverty is that they do not reflect access to basic needs and services (Thomas et al., 2011; Vearey et al., 2010). Such a critique has led some researchers to call for moving away from income-related measures of poverty to measures capturing critical aspects of poverty – that is, capturing material deprivation – that best represent the conditions and realities of living experiences in urban slums to predict sexual risk (Greif, 2012; Kunnuji, 2014; Mberu et al., 2013). 2 In this research, I define HIV-related sexual risk (or sexual risk for short) as those sexual practices that have been found to be associated with high risk of HIV acquisition. These sexual practices considered in this research are: (1) non-use of condom at last sex; (2) multiple (two or more) sexual partners; (3) transactional sex; and (4) coercive sex. I use the term sexual risk as an umbrella term of these sexual practices in this thesis. Specific definitions of these sexual practices used in this research are provided in the published papers. To this end, three measures of disadvantage – housing and food insecurity, and poor access to health care – have been highlighted as critical living conditions associated with sexual risk in urban slums in the SEA region (Greif, 2012; Mberu et al., 2013). Housing, food and health care access are multidimensional concepts that encompass aspects of housing quality, instability and overcrowding (Aidala et al., 2005); availability, accessibility and utilisation of food (Ivers and Cullen, 2011) and affordability, acceptability and availability of health care (Peters et al., 2008; Thiede and McIntyre, 2008). My review of the literature confirms that there is no consistency on what constitutes and how to measure these critical living conditions in relation to HIV sexual risk in urban slums of the SEA region. Another limitation in existing studies is that the association between material deprivation and sexual risk is most frequently examined through the inclusion of either one dimension of deprivation or a single deprivation-related item (Greif, 2012). The lack of data on multiple-item measurement of the components of material deprivation has always been cited as a reason for considering one dimension of deprivation or a single deprivation-related item. It is likely that additional nuance underlying this association is not yet examined given that single-item measures of deprivations overestimate or underestimate the statistical significance leading to biased results (Noble et al., 2010). My research contributes in this direction by using multiple measures of housing and food insecurity and poor access to health care to conceptualise material deprivation in relation to sexual risk. Aim Based on the findings from two sub studies, this PhD aims to better understand relationships between local measures of material deprivation, economic deprivation, and dimensions of sexual risk that will inform the development of an urban slum-specific combination HIV prevention framework to reduce the HIV risk of young people in Malawi and South Africa. Methods Two sub studies were undertaken: one in South Africa (Sub study I) and another one in Malawi (Sub study II). Sub study I was a secondary analysis of a cross-sectional survey of young people (n = 530) living in urban slum households extracted from the 2011 loveLife survey with young people aged 18-23 years old in South African four of nine provinces – KwaZulu Natal, Mpumalanga, Eastern Cape, and Gauteng. Young people in the age cohort of 18-23 years old are important given that they transition rapidly from low HIV vulnerability when they are 10-19 years old to high HIV vulnerability (UNICEF, 2011). Sub study II included five focus group discussions and 12 in-depth interviews, undertaken with a total of 60 young people aged 18-23 years old, exploring living experiences and sexual risk practices in the urban slums of Blantyre, Malawi. This exploratory qualitative study informed a cross-sectional survey that explored material deprivation and sexual practices among young people (n = 1,071) in the urban slums of Blantyre, Malawi. The focal research variables in Sub study I were ‘material deprivation’, ‘financial difficulty’ and ‘sexual risk-taking’. Material deprivation was assessed by summing the single-item measures of housing and food insecurity, and poor access to health care to create a dichotomous measure that indicated the presence of one or more hardships. Financial difficulty was defined by assessing items that asked respondents if they received any income from any source last month or if anyone in the household receive a grant. The dependent variable, sexual risk-taking, was defined from a series of items about condom use, multiple (two or more) sexual partners and transactional sex. Variables from these measures were dichotomous, indicating whether a respondent used condom(s) at last sex or not, had two or more sexual partners or not, and exchanged sex for money or goods in the last 12 months or not. The variable on transactional sex involved the giving of cash or goods for males and receiving for females. Given this information, ‘high sexual risk-taking’ was defined by those who reported that they had not used condoms at last sex or exchanged sex for money or goods in the last 12 months irrespective of the number of sexual partners they had or those who reported use of condoms at last sexual intercourse, but had more than one sexual partner. In Sub study II, material deprivation was defined by a total of 20 indicators of insufficient housing, food insecurity, and poor access to health care (see Paper II, Table 1). From these indicators, all households in the study sites were assigned a deprivation sum score based on 17 items (11 on food insecurity, five on housing and one on poor access to health care) available from the survey instrument (see Paper III). Households with the highest scores were defined as ‘deprived households’ (Dodoo et al., 2007; Kuipers et al., 2013). Unemployment was determined by asking the young people their primary occupation – the work from which they earn most of their income (Dzator, 2013). Six occupational categories (construction, trading, services, industry, agriculture, and unemployed) were created from the responses and used in the analysis. Coercive sex was measured asking whether a respondent had been forced by a partner or non-partner to have sex when he or she did not want to. Through synthesis of the findings from the two sub studies, four central themes were identified: (1) patterns of sexual risk; (2) underlying reasons for decisions to engage in transactional sex; (3) measures of disadvantage associated with sexual risk; and (4) study design and methods. These four themes assist me in making recommendations for a new urban slum-specific combination HIV prevention model for young people in the SEA region. Key findings Prevalence of no condom use at last sex was significantly higher in both young men and young women living in urban slums in Malawi than their counterparts in South Africa (p<0.001, respectively). With regards to multiple sexual partners, young people in Malawi were less likely to report having multiple partners than their counterparts in South Africa (49.3% vs. 64.1% (p=0.002) for young men, and 18.7% vs. 28.7% (p=0.032) for young women). Prevalence of coercive sex in Malawi was 15.8% and 44.4% in young men and young women respectively. Lastly, transactional sex was significantly higher in young men and young women in Malawi than their counterparts in South Africa (p<0.001, respectively). The qualitative data analysis reveals that housing and food insecurity influenced both young men’s and young women’s motivations for engaging in transactional sex. Poor access to health care and a desire for high-value social goods (such as cellular phones, the latest hair, and clothing styles) influenced the young women’s motivations for transactional sex. Adjusted logistic regression models show that material deprivation was significantly associated with increased odds of high sexual risk-taking for young men (adjusted odds ratio [AOR]=1.20; 95% confidence interval [CI]=1.10, 5.58) and young women (AOR=1.43; 95%CI=1.35, 3.28). However, financial difficulty is the most salient influence on young women’s high sexual risk-taking (AOR=2.11; 95%CI=1.66, 2.70) (Paper I). A multi-level model adjusted for other risk behaviours, age, marital status, duration of residence, household structure, school status, level of education and received money from relatives shows that unemployment was associated with young men to report experiencing coercive sex (AOR=1.77, 95%CI: 1.09, 3.21) while material deprivation (AOR=1.34, 95%CI: 0.75, 2.39) was not. Young women in deprived households were more likely to report experiencing coercive sex (AOR=1.37, 95%CI: 1.07, 2.22) than in less deprived households (Paper III). A further analysis of logged coefficients from the regression models (Model 2 in Paper I, Table 4, and Paper III, Table 5) multiplied by their standard deviations shows that: (1) for young women, financial difficulty (0.16 = (log 2.11)*(0.50)) exerted the strongest effects on sexual risk-taking followed by material deprivation (0.10 = (log 1.43)*(0.66)) (Paper I). In the same Paper I, material deprivation (0.04 = (log 1.20)*(0.50)) shows noteworthy and significant effects on sexual risk-taking for young men; (2) material deprivation (0.08 = (log 1.37)*(0.58)) and unemployment (0.12 = (log 1.77)*(0.50)) were the most influential variables associated with coercive sex among young men and young women respectively. Based on this further analysis, I argue that existing HIV prevention efforts based on research that employed only measures of economic deprivation – and did not explore material deprivation – may not be effectively responding to the specific conditions found to be critical in the urban slum settings in Malawi and South Africa. With regard to the strengths of association between various deprivations and dimensions of sexual risk, this research shows what the gaps are in understanding the differences between South Africa and Malawi and therefore what future research is needed to develop tools to assess the generic and specific settings in different countries in the SEA region regarding HIV risk. Material deprivation has been shown in this study to remain a robust predictor of sexual risk in statistical models (Papers I and III). Furthermore, findings from the qualitative study (Paper II) indicate that a desire for high-value social items (for example, latest clothing styles, hair products, and cellular phones) underlie young women’s motivations for transactional sex in urban slums. Thus, it is argued that responses to material deprivation, to reach full potential, need to engage with a desire for fashionable goods as well. Through the synthesis of the findings from the two studies, I have suggested key recommendations and strategies required for responses that prioritise young people in urban slums in Malawi and South Africa. Implications Based on the findings of this research, an urban-slum specific combination HIV prevention framework is suggested for young people in the SEA region. It is argued that a combination HIV prevention framework targeting young people living in these settings requires understanding of both material disadvantage and desire for high-value social goods, since intervention efforts that focus on a purely income-related measure of poverty of levels of income are likely to fail. This requirement is likely to be useful for addressing the unique challenges faced by young people residing in urban slums. Moreover, the understanding suggested here will enable countries in the SEA region to reduce health risks among young people living in urban slums. This research contributes a new methodological approach that highlights what data are needed in order to better understand context-specific measures of disadvantage associated with sexual risk among young people living in urban slum settings in the SEA region. The study findings suggest that there is need to develop appropriate tools to collect more detailed research data on the generic and context-specific issues in different urban slum settings in the SEA region regarding structural determinants of HIV sexual risk that would inform context-specific responses. Future multi-country study should employ a settings approach to assess HIV-related sexual risks and guide the development of appropriate responses for young people living in urban slums of the SEA region. Keywords: Material deprivation, young people, urban slums, HIV, sexual risk, combination HIV prevention. === MB2016