The work is focussed on the HIV epidemic among key populations in Latin America, a region that has received limited international attention due to the concentrated nature of its epidemic. It faces important challenges in prevention and risks missing the opportunity to control the epidemic. Despite the evidence available showing the disproportionate burden of infection among key populations, principally men who have sex with men (MSM) and to a lesser extent injecting drug users (IDU) and female sex workers (FSW), a small percentage of countries’ HIV prevention spending is allocated towards prevention interventions among these groups. In this thesis, we first estimate the distribution of new infections by type of exposure in the Dominican Republic, Mexico and Peru and find that MSM account for a large share of new infections in the three countries (over 50% in Mexico and Peru and 30% in Dominican Republic). This highlights that this population should be prioritised in prevention programmes. We then turn to interventions for MSM, and investigate whether pre-exposure prophylaxis (PrEP) to prevent HIV acquisition among MSM in Lima, Peru represents a cost-effective intervention at the population level. We find PrEP to be a potentially cost-effective intervention although unaffordable at a large scale. We conclude it should be considered as an additional tool within a combination prevention approach among this population. Following these, we seek to understand the dynamics, causes and consequences of risk behaviours and HIV infection in another key population that has rarely been prioritised by public health programmes: prison inmates. We use data from the largest prison in Peru and find that HIV prevalence is 4 times higher than national prevalence and inmates engage in riskier sexual behaviours than in the general population. Nevertheless, there are indications of risk behaviours and HIV prevalence reducing in recent years. To help direct prevention efforts in this setting, we characterise the risk profile of inmates who engaged in sex with FSW, other men and in unprotected sex with casual partners and identify inmates who have sex with men as being at heightened risk of infection (due to higher levels of drug use and unprotected sex). We use those data to develop a mathematical model that represents the HIV epidemic in the prison. Contrary to expectations, we find that incidence is likely to be less than 1% per year within the prison and that the dynamics of HIV in prisons are principally driven by inmates infected before their incarceration. We further estimate that since the beginning of the epidemic, at least 5% of HIV cases identified in Lima have passed through the prison, suggesting that the routine screening system in place at prisons could make a substantial contribution to controlling the HIV epidemic across the city. Finally, we formulate a conceptual framework of HIV risk among prison inmates in Latin America focussing on the social and underlying determinants of risk and develop a structured questionnaire to test the hypotheses proposed through the framework and obtain information relevant to the design of interventions within prisons and at the community level. Together this work draws on social and quantitative science to provide new insights into key populations in Latin America that should contribute to stronger and responsive HIV prevention programmes, to the benefit of all those that remain at risk of HIV in the region.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:572229 |
| Date | January 2013 |
| Creators | Bórquez, Annick |
| Contributors | Hallett, Timothy ; Garnett, Geoffrey |
| Publisher | Imperial College London |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/10044/1/11051 |
Page generated in 0.01 seconds