Understanding the impact of an HIV intervention package for adolescents

Date
2013-12
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: Adolescents are regarded as a high risk group in South Africa with the highest human immunodeficiency virus (HIV) incidence occurring in this group. Prevention among adolescents is therefore a key in decreasing the HIV burden. This thesis aims to assist in the design of trials by simulating the potential outcomes of a combination prevention trial in adolescents. We develop a stochastic individual-based model stratified by sex and age. We then use this model to determine the impact of various prevention packages on HIV incidence among adolescents participating in a hypothetical trial over a three year period. The trial that is simulated involves an intervention arm, in which adolescents are offered a choice of a prevention methods (including medical male circumcision (MMC), oral pre-exposure prophylaxis (PrEP) and antiretroviral-based vaginal microbicides (ARV-VM)), and a control arm. We predict that the impact of a full prevention package on HIV incidence would be a 46% per personyear( PPY) (95% CI 45–47%) risk reduction. The combination of MMC and PrEP has a substantial impact on HIV incidence in males, with a 51% PPY (95% CI 49–53%) relative risk of HIV infection. Offering women the choice of PrEP, a microbicide gel or a microbicide in the form of a vaginal ring would be less effective, with a 57% PPY (95% CI 56–58%) relative risk of HIV acquisition. This is not substantially different from the relative risk estimated when the vaginal ring alone is offered, as the ring is assumed to be the most accept able of the three prevention methods. We determine a sample size requirement of approximately 1013 in each arm of a trial would achieve 80% power to detect a statistically significant reduction in HIV risk. We find that the relative risk is sensitive to the assumed degree of correlation between condom use and the acceptability of the prevention method. We also find that the most efficient trial design may be to offer both MMC and PrEP to males but to offer only a microbicide ring to females. Further work is required to better understand the processes by which adolescent prevention method choices are made.
AFRIKAANSE OPSOMMING: Adolessente word beskou as ‘n hoe risiko groep in Suid Afrika, met die hoogste menslike immuniteitsgebrekvirus (MIV) insidensie in hierdie groep. Voorkoming van MIV onder adolessente is daarom noodsaaklik om die MIV las te verminder. Die doel van hierdie tesis is om te help met die ontwerp van studies deur die moontlike uitkomste van ‘n kombinasie-voorkoming studie in adolessente te simuleer. Ons het ‘n stogastiese individu-gebaseerde model, gestratifiseer met betrekking tot seks en ouderdom, ontwikkel. Ons het toe die model gebruik om die impak van ‘n verskeinheid van voorkomingspakette op MIV insidensie onder adolessente wat deelneem aan ‘n hipotetiese proef oor ‘n drie jaar periode, te bepaal. Die proef wat gesimuleer word behels a intervensie groep, waarin die jong volwassenes ‘n keuse van voorbehoedings metodes (insluitende mediese manlike besnydenis (MMB), pre-blootstelling profilakse (PrBP) en anti-retrovirale vaginale mikrobisiedes (ARV-VM)) aangebied word, en ‘n kontrole groep. Ons voorspel dat die impak van ‘n volle voorkomingspaket op MIV insidensie ‘n 46% per persoon-jaar (PPJ) (95% VI 47–47%) risiko vermindering sal wees. Die kombinasie van MMB en PrBP het ‘n substansiele impak op MIV insidensie onder mans, met ‘n relatiewe risiko van MIV infeksie van 51% PPJ (95% VI 49–53%). Om die keuse van PrBP, ‘n mikrobisiede gel of ‘n mikrobisiede in die vorm van ‘n vaginale ring aan vrouens te bied, is minder effektief, met ‘n relatiewe risiko van MIV infeksie van 57% PPJ (95% VI 56%–58%). Hierdie verskil nie substansieel van die beraamde relatiewe risiko in die geval waar slegs die vaginale ring gebied word nie, aangesien daar aanvaar word dat die ring die mees aanvaarde van die drie voorkomingsmetodes is. Ons het bepaal dat ‘n steekproef van ongeveer 1013 individue in elke arm van die proef nodig is om ‘n 80% kans te he om ‘n statisties betekenisvolle afname in MIV-risiko te bespeur. Ons vind dat die relatiewe risiko sensitief is tot die aanvaarde graad van die korrelasies tussen kondoom-gebruik en die aanvaarding van die voorkomings metodes. Ons het ook gevind dat dit mag wees dat die mees doeltreffende proef ontwerp is om beide MMB en PrBP vir mans en slegs ‘n mikrobisiede ring vir vrouens te bied. Verdere werk word benodig om die prosesse waarby jong volwassenes keuses maak oor voorkomingsmetodes te verstaan.
Description
Thesis (MSc)--Stellenbosch University, 2013.
Keywords
HIV infections -- South Africa -- Prevention, Stochastic models, Randomized controlled trials, Dissertations -- Mathematical sciences, Theses -- Mathematical sciences, Dissertations -- Mathematics, Theses -- Mathematics, AIDS (Disease) in adolescence -- South Africa -- Prevention
Citation