UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An investigation of the impact of expanding access to Highly Active Antiretroviral Therapy (HAART) on… Kaida, Angela Kanana 2009

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2010_spring_kaida_angela.pdf [ 4.38MB ]
Metadata
JSON: 24-1.0058273.json
JSON-LD: 24-1.0058273-ld.json
RDF/XML (Pretty): 24-1.0058273-rdf.xml
RDF/JSON: 24-1.0058273-rdf.json
Turtle: 24-1.0058273-turtle.txt
N-Triples: 24-1.0058273-rdf-ntriples.txt
Original Record: 24-1.0058273-source.json
Full Text
24-1.0058273-fulltext.txt
Citation
24-1.0058273.ris

Full Text

AN INVESTIGATION OF THE IMPACT OF EXPANDING ACCESS TO HIGHLY ACTIVE ANTIRETROVIRAL THERAPY (HAART) ON SEXUAL AND REPRODUCTIVE DECISION-MAKING AND BEHAVIOURS OF WOMEN IN HIGH HIV PREVALENCE SETTINGS IN SUB-SAHARAN AFRICA  by ANGELA KANANA KAIDA BSc, University of Waterloo, Canada, 1997 MSc, University of Alberta, Canada, 2000  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Health Care and Epidemiology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  April 2010 © Angela Kaida, 2010  ABSTRACT Background: Given a paucity of information regarding the impact of expanding access to highly active antiretroviral therapy (HAART) on sexual and reproductive decision-making and behaviours of women in high HIV prevalence settings in subSaharan Africa, the objectives of thesis were: To assess whether use and duration of HAART was associated with (1) recent sexual activity among HIV-positive women across three high HIV prevalence settings; (2) fertility intentions and (3) contraceptive use and method mix patterns among women in Soweto, South Africa. And, (4) to develop a reliable HAART optimism scale for use among HIV-positive women and to test the scale’s validity against measures of sexual and reproductive decision-making and behaviours among women in Mbarara, Uganda. Methods: Quantitative data were drawn from surveys and medical record reviews conducted among 751 women attending the Perinatal HIV Research Unit in Soweto, South Africa (253 HAART-experienced, 249 HAART-naïve, and 249 HIV-negative) and 540 HIV-positive women (half of whom were receiving HAART) attending Mbarara University’s HIV clinic in Uganda. Surveys assessed socio-demographics, HIV status and HAART history, sexual and reproductive health decision-making and behaviours, HIV-related clinical assessments, and HAART optimism. Results: The analyses revealed that HIV-positive women receiving HAART are more likely to use contraception overall and dual protection in particular, with minimal differences in fertility intentions or sexual activity relative to their HAARTnaïve counterparts. Moreover, optimism about the effects of HAART, rather than actual use or non-use, may be a more important predictor of fertility intentions and sexual activity of HIV-positive women. Overall, HIV-positive women are less likely to report fertility intentions and more likely to use contraception (and condoms in particular) relative to HIV-negative women from the same community. Conclusions: The findings highlight the potential great value and urgent need for improved integration between HIV prevention, testing, and HAART services and sexual and reproductive health (SRH) programming to address the diverse SRH needs of HIV-infected and –affected women in HIV endemic settings. Such integration is essential to better support the rights of all women to be sexually active and safely achieve their reproductive goals, while minimizing HIV transmission risks.  ii  TABLE OF CONTENTS ABSTRACT .........................................................................................................................................................ii TABLE OF CONTENTS .................................................................................................................................. iii LIST OF TABLES ............................................................................................................................................ vii LIST OF FIGURES ......................................................................................................................................... viii ACKNOWLEDGEMENTS .............................................................................................................................. ix DEDICATION ................................................................................................................................................... xi CO-AUTHORSHIP STATEMENT ...............................................................................................................xii CHAPTER 1: BACKGROUND, STUDY JUSTIFICATION, AND OBJECTIVES.................................. 1 1.1 OVERVIEW OF HIV, HAART, AND SEXUAL AND REPRODUCTIVE HEALTH IN SUB-SAHARAN AFRICA ... 1 1.2 SITUATIONAL ANALYSIS OF HIV INFECTION AND SEXUAL AND REPRODUCTIVE HEALTH IN THE CONTEXT OF EXPANDING ACCESS TO HAART IN SUB-SAHARAN AFRICA ..................................................... 3  1.3 BRIEF SITUATIONAL ANALYSIS OF HIV, HAART SCALE-UP, AND SEXUAL AND REPRODUCTIVE HEALTH IN SOUTH AFRICA AND UGANDA .................................................................................................................... 8  1.3.1 South Africa ......................................................................................................................................... 9 1.3.2 Uganda ............................................................................................................................................... 11 1.4 OUTSTANDING QUESTIONS AND STUDY JUSTIFICATION ......................................................................... 13 1.5 PURPOSE, OBJECTIVES, AND HYPOTHESES ............................................................................................... 15 1.6 OVERVIEW OF THESIS ................................................................................................................................ 19 1.7 REFERENCES .............................................................................................................................................. 22 CHAPTER 2: THE POTENTIAL IMPACT OF ANTIRETROVIRAL THERAPY ON FERTILITY IN SUB-SAHARAN AFRICA .............................................................................................................................. 32 2.1 INTRODUCTION ................................................................................................................................... 32 2.2 HIV AND FERTILITY IN SUB-SAHARAN AFRICA ......................................................................... 34 2.3 ARV THERAPY AND FERTILITY IN SUB-SAHARAN AFRICA .................................................... 35 2.4 ARV THERAPY AND FERTILITY IN DEVELOPED COUNTRIES ................................................. 36 2.5 BONGAARTS’ PROXIMATE DETERMINANTS OF FERTILITY FRAMEWORK: UNDERSTANDING THE MECHANISMS THROUGH WHICH ARV THERAPY MAY INFLUENCE FERTILITY ............................................................................................................................. 39 2.5.1 Marriage/Sexual debut ....................................................................................................................... 40 2.5.2 Onset of permanent sterility .............................................................................................................. 41 2.5.3 Post-partum infecundability .............................................................................................................. 43 2.5.4 Natural fecundability or frequency of intercourse ............................................................................. 45 2.5.5 Use and effectiveness of contraception ............................................................................................... 46 2.5.6 Spontaneous intra-uterine mortality ................................................................................................. 47  iii  2.5.7 Induced abortion ................................................................................................................................ 48 2.6 CONCEPTUAL FRAMEWORK FOR ANALYSING THE POTENTIAL NET IMPACT OF ARV THERAPY ON FERTILITY .......................................................................................................................... 49 2.7 CONCLUSION ........................................................................................................................................ 51 2.8 REFERENCES .......................................................................................................................................... 59 CHAPTER 3: THE RELATIONSHIP BETWEEN HAART USE AND SEXUAL ACTIVITY AMONG HIV-POSITIVE WOMEN OF REPRODUCTIVE AGE IN BRAZIL, SOUTH AFRICA, AND UGANDA .......................................................................................................................................................... 71 3.1 INTRODUCTION ................................................................................................................................... 71 3.2 METHODS ............................................................................................................................................... 73 3.2.1 Study design ...................................................................................................................................... 73 3.2.2 Study setting and sample population ................................................................................................. 73 3.2.3 Data collection ................................................................................................................................... 74 3.2.4 Data analysis...................................................................................................................................... 75 3.2.5 Ethical considerations ........................................................................................................................ 76 3.3 RESULTS .................................................................................................................................................. 77 3.3.1 Baseline characteristics ...................................................................................................................... 77 3.3.2 Univariate analyses ............................................................................................................................ 78 3.3.3 Adjusted analyses............................................................................................................................... 78 3.3.4 Contraceptive use and protected sexual activity ................................................................................ 79 3.4 DISCUSSION ........................................................................................................................................... 79 3.5 REFERENCES .......................................................................................................................................... 84 CHAPTER 4: FERTILITY INTENTIONS OF HIV-POSITIVE WOMEN OF REPRODUCTIVE AGE IN SOWETO, SOUTH AFRICA: THE INFLUENCE OF EXPANDING ACCESS TO HAART IN AN HIV HYPER-ENDEMIC SETTING .............................................................................................................. 88 4.1 INTRODUCTION ................................................................................................................................... 88 4.2 METHODS ............................................................................................................................................... 91 4.2.1 Study design ...................................................................................................................................... 91 4.2.2 Study setting ...................................................................................................................................... 92 4.2.3 Eligibility criteria ............................................................................................................................... 92 4.2.4 Study sample ...................................................................................................................................... 93 4.2.5 Data collection ................................................................................................................................... 94 4.2.6 Data collection instruments ............................................................................................................... 95 4.2.7 Measures ............................................................................................................................................ 96 4.2.8 Statistical analysis ............................................................................................................................. 97 4.2.9 Sub-analyses....................................................................................................................................... 98 4.3 RESULTS .................................................................................................................................................. 98 4.3.1 Study sample ...................................................................................................................................... 98 4.3.2 Baseline characteristics ...................................................................................................................... 99 4.3.3 Prevalence of fertility intentions ...................................................................................................... 100  iv  4.3.4 Factors associated with fertility intentions in the overall sample .................................................... 100 4.3.5 Factors associated with fertility intentions among HIV-positive women ........................................ 101 4.3.6 Sub-analyses..................................................................................................................................... 102 4.4 DISCUSSION ......................................................................................................................................... 103 4.5 CONCLUSION ...................................................................................................................................... 108 4.6 REFERENCES ........................................................................................................................................ 117 CHAPTER 5: CONTRACEPTIVE USE AND METHOD MIX PATTERNS AMONG HIVPOSITIVE AND HIV-NEGATIVE WOMEN IN SOWETO, SOUTH AFRICA IN THE CONTEXT OF EXPANDING ACCESS TO HIGHLY ACTIVE ANTIRETROVIRAL THERAPY ...................... 126 5.1 INTRODUCTION ................................................................................................................................. 126 5.2 METHODS ............................................................................................................................................. 129 5.2.1 Study setting .................................................................................................................................... 129 5.2.2 Study design .................................................................................................................................... 130 5.2.3 Eligibility criteria ............................................................................................................................. 130 5.2.4 Study sample .................................................................................................................................... 131 5.2.5 Data collection ................................................................................................................................. 132 5.2.6 Data collection instruments ............................................................................................................. 133 5.2.7 Measures .......................................................................................................................................... 134 5.2.8 Statistical analysis ........................................................................................................................... 135 5.2.9 Sub-analyses..................................................................................................................................... 136 5.2.10 Ethical considerations .................................................................................................................... 137 5.3 RESULTS ................................................................................................................................................ 137 5.3.1 Baseline characteristics .................................................................................................................... 138 5.3.2 Prevalence of contraceptive use ........................................................................................................ 139 5.3.3 Types of contraceptive methods used ............................................................................................... 139 5.3.4 Univariate and adjusted analysis of contraceptive use: HAART users, HAART non-users, and HIVnegative women......................................................................................................................................... 140 5.3.5 Univariate and adjusted analysis: HIV-positive women .................................................................. 141 5.3.6 Contraceptive use by duration of HAART use ................................................................................ 142 5.3.7 Sub-analyses..................................................................................................................................... 142 5.4 DISCUSSION ......................................................................................................................................... 143 5.5 REFERENCES ........................................................................................................................................ 160 CHAPTER 6: THE WHOMEN’S SCALE (WOMEN’S HAART OPTIMISM MONITORING AND EVALUATION SCALE V.1) AND THE ASSOCIATION WITH FERTILITY INTENTIONS AND SEXUAL BEHAVIOURS AMONG HIV-POSITIVE WOMEN IN UGANDA ................................... 167 6.1 INTRODUCTION ................................................................................................................................. 167 6.2 METHODS ............................................................................................................................................. 170 6.2.1 Study design .................................................................................................................................... 170 6.2.2 Study setting .................................................................................................................................... 170 6.2.3 Eligibility criteria ............................................................................................................................. 171  v  6.2.4 Data collection ................................................................................................................................. 171 6.2.5 Instrument Development ................................................................................................................. 172 6.2.6 Data analysis.................................................................................................................................... 174 6.2.7 Ethical considerations ...................................................................................................................... 178 6.3 RESULTS ................................................................................................................................................ 178 6.3.1 Recruitment ..................................................................................................................................... 178 6.3.2 Baseline characteristics .................................................................................................................... 178 6.3.3 Factor analysis ................................................................................................................................. 179 6.3.4 Association between the derived HAART optimism scale and fertility intentions .......................... 181 6.3.5 Association between the derived HAART optimism scale and sexual activity in the previous three months....................................................................................................................................................... 181 6.3.6 Association between the derived HAART optimism scale and unprotected sexual activity in the previous three months ............................................................................................................................... 181 6.4 DISCUSSION ......................................................................................................................................... 182 6.5 REFERENCES ........................................................................................................................................ 192 CHAPTER 7: SUMMARY, CONTRIBUTIONS, RECOMMENDATIONS, FUTURE RESEARCH, AND CONCLUSION .................................................................................................................................... 199 7.1 SUMMARY OF OBJECTIVES ....................................................................................................................... 199 7.2 SUMMARY OF FINDINGS .......................................................................................................................... 200 7.3 UNIQUE CONTRIBUTIONS ....................................................................................................................... 209 7.4 RECOMMENDATIONS .............................................................................................................................. 215 7.5 FUTURE RESEARCH ................................................................................................................................. 218 7.6 CONCLUSION........................................................................................................................................... 222 7.7 REFERENCES ............................................................................................................................................ 223 APPENDIX 1: HUMAN ETHICS APPROVAL CERTIFICATES .......................................................... 231 APPENDIX 2: DATA COLLECTION TOOL FOR THE SOWETO, SOUTH AFRICA STUDY ...... 241  vi  LIST OF TABLES Table 2.1: Use of Bongaarts’ Proximate Determinants of Fertility framework to describe the potential impact of antiretroviral (ARV) therapy on fertility in sub-Saharan Africa ......... 53 Table 3.1: Baseline characteristics of study population: HIV-infected women of reproductive age in Brazil, South Africa, and Uganda (n=179) ...................................................... 82 Table 3.2: Unadjusted and adjusted Odds Ratios and 95% Confidence Intervals (95% CI) of variables associated with recent sexual intercourse (in the previous month) among HIV-infected women in Brazil, South Africa, and Uganda (n=179) .............................................. 83 Table 4.1: Baseline characteristics of HAART users, non-HAART users, and HIV-negative women (aged 18-44 years and non-sterilized) in Soweto, South Africa (n=674) ....................... 109 Table 4.2: Univariate and Adjusted analyses of variables associated with fertility intentions among non-sterilized women aged 18-44 years in Soweto, South Africa (n=674).. ................................................................................................................................................... 112 Table 4.3: Univariate and adjusted analyses of variables associated with fertility intentions among non-sterilized HIV-positive women aged 18-44 years in Soweto, South Africa (n=432) ......................................................................................................................................... 114 Table 5.1: Baseline characteristics of HAART users, non-HAART users, and HIV-negative women (aged 18-44 years, currently sexually active and non-pregnant) in Soweto, South Africa (n=563) ......................................................................................................................................... 149 Table 5.2: Types of contraceptive methods used by HIV-positive (HAART users and nonHAART users) and HIV-negative women (aged 18-44 years, currently sexually active and non-pregnant) in Soweto, South Africa ............................................................................................ 152 Table 5.3: Univariate and adjusted analyses of variables associated with contraceptive use among women (aged 18-44 years, currently sexually active and non-pregnant) in Soweto, South Africa (n=563) ............................................................................................................................. 154 Table 5.4: Univariate and adjusted analyses of variables associated with contraceptive use among HIV-positive women (aged 18-44 years, currently sexually active and nonpregnant) in Soweto, South Africa (n=349) ...................................................................................... 156 Table 6.1: Characteristics of study population of HIV-positive women in Mbarara, Uganda (n=540) ...................................................................................................................................... 186 Table 6.2: Principal component Factor analysis of 23 statements about HAART .................... 187 Table 6.3: Items with Principal Components varimax factor loadings ....................................... 190 Table 6.4: Statements constituting the derived HAART optimism scale for use among HIV-positive women of reproductive age living in developing countries................................ 190  vii  LIST OF FIGURES Figure 2.1 Bongaarts’ proximate determinants of fertility framework over the reproductive lifespan ............................................................................................................................. 57 Figure 2.2. Conceptual framework for the potential impact of antiretroviral (ARV) therapy on fertility in sub-Saharan Africa.. ...................................................................................................... 58 Figure 4.1: Percent of women (aged 18-44 years and non-sterilized) who intend to have (more) children by length of time on HAART among HIV-positive women in Soweto, South Africa (n=426) ............................................................................................................................ 116 Figure 5.1: Percent of HIV-positive women (aged 18-44 years, currently sexually active and non-pregnant) using contraception by duration of HAART use in Soweto, South Africa (n=349) ......................................................................................................................................... 159 Figure 6.1: Results of the scree test .................................................................................................... 191  viii  ACKNOWLEDGEMENTS  I would like to thank my co-supervisors Drs. Bob Hogg and Patti Janssen, and my committee members, Drs. Steffanie Strathdee and Deborah Money, for their unyielding support, guidance, and enthusiasm for this research endeavour. In particular, I would like to thank Bob for his unwavering and extraordinary commitment to me and this important area of research. I am eternally grateful to my research and supervisory team in South Africa. To Dr. Glenda Gray, my field supervisor and mentor, thank you for being nothing short of inspirational. To Dr. Fatima Laher, an exceptional HIV clinician in Soweto, I am thankful for your friendship, your collegiality, and your big brain. To my professional and dedicated research colleagues and staff at the Perinatal HIV Research Unit: Sister Nokuzola, Hanifah, Mpume, Nthabiseng, Janan, and Jonathan. I thank you for your friendship, commitment, and contributions to this project. To my friends in Jozi and Soweto, thank you for being such wonderful and entertaining hosts! I’d also like to thank Dr. Irene Andia and the incredible clinical and research staff at the Mbarara ISS Clinic and the Mbarara University of Science and Technology in Uganda. A special word of acknowledgement to my mentors in global health and HIV/AIDS research, Drs. David Bangsberg, Francisco Bastos, Walter Kipp, and Jean Lebel, who have guided and supported my work for many years. I’d like to thank the Michael Smith Foundation for Health Research (MSFHR) and the Canadian Institutes of Health Research (CIHR) for supporting me through Doctoral Research Awards. This study would not have been possible without the generous financial support of the Canadian Foundation for AIDS Research (CANFAR), the Social Sciences and Humanities Research Council (SSHRC), the  ix  UCSD Center for AIDS Research and the B.C. Centre for Excellence in HIV/AIDS. I’d also like to thank the Canadian Association for HIV Research (CAHR) for granting me the 2009 New Investigator Award (Epidemiology and Public Health) on the basis of a portion of my thesis work. Thank you to my friends and family who have expressed endless confidence in me and the value of this work. To my fellow graduate students and colleagues at the B.C. Centre for Excellence in HIV/AIDS, thank you for your friendship and for being constant sources of inspiration: Elisa Lloyd-Smith, Viviane Dias Lima, Karissa Johnston, Brandon Marshall, MJ Milloy, Will Small, Kora de Beck, Kate Shannon, Sarah Fielden, Danya Fast, Melanie Rusch, Beth Rachlis, and Aranka Anema. As most of the chapters included in this dissertation are based on peer-reviewed published manuscripts, I’d like to thank the anonymous peer-reviewers who provided critical feedback for much of this work. To my parents and family who, despite their concerns about an academic career, have been behind me from day one. I remain forever grateful to my best friend and academic inspiration, the amazing Patti Lenard. To my husband and number one fan, Steven, and our son, Nikhil (our own little fertility project within this fertility project), your love, support, encouragement, and sleeping schedules made it possible for me to complete this journey. Finally, I would like to express my most sincere thanks to the women and families who participated in this study, for sharing this personal and important information with us.  x  DEDICATION  I would like to dedicate this work to the women of sub-Saharan Africa. Wanawake ni msingi wa maendeleo. And, to two guys: Steven and Nikhil. Nyinyi ni furaha ya moyo wangu. Ninawapenda sana.  xi  CO-AUTHORSHIP STATEMENT This is to certify that the work presented in this thesis was conceived, instrumented, written, and disseminated by the PhD candidate. The co-authors of the manuscripts that constitute part of this thesis made contributions only as is commensurate with committee or collegial duties. The co-authors reviewed each manuscript prior to submission for publication and offered critical evaluations; however, the candidate was responsible for conducting data analyses and preparing the initial drafts of all manuscripts. In addition, the candidate was responsible for revising the manuscripts based on the suggestions of the co-authors, submitting manuscripts for publication and preparing final revisions based on the comments of the journal editors and external peer reviewers.  xii  CHAPTER 1: BACKGROUND, STUDY JUSTIFICATION, AND OBJECTIVES  1.1 Overview of HIV, HAART, and sexual and reproductive health in sub-Saharan Africa  With women accounting for over half of the world’s HIV-infected population and an increasing proportion of incident infections each year, there is clear evidence of the feminization of the global HIV pandemic [1]. The vast majority live in subSaharan Africa where over 12 million women are currently living with HIV, accounting for 61% of the region’s HIV-infected adult population [1]. While HIVpositive women face many of the same sexual and reproductive health burdens and challenges as HIV-negative women, HIV-infection among women of reproductive age raises additional personal, clinical, and public health concerns about the incidence of pregnancy. Such concerns relate to the health and survival of the mother and risks of HIV transmission to sero-discordant sexual partners and to infants during pregnancy, childbirth, and breastfeeding [2].  A sizable body of literature has examined the impact of HIV infection on the predictors, incidence, and outcomes of pregnancy [3-5]; however, there remains a paucity of research that addresses these relationships in the context of sustained use  1  of antiretroviral therapy. There is now unequivocal evidence that receipt of highly active antiretroviral therapy (HAART) among HIV-infected individuals dramatically improves viral suppression, reduces morbidity and mortality at the individual level, and increases life expectancy at the population level [6-9]. Recent global efforts to expand access to HAART and clinical care are delivering these positive outcomes to an increasing number of HIV-infected individuals living in high HIV prevalence, limited-resource settings [10]. The scale-up process has been dramatic; by the end of 2008 an estimated 2.9 million people in sub-Saharan Africa were receiving antiretroviral therapy, a 30-fold increase since the end of 2003 [10].  In addition to the clinical and survival benefits, appropriate use of HAART also dramatically reduces the risks of vertical [11] and horizontal transmission [12, 13] of HIV. Accordingly, increased HAART availability is anticipated to influence sexual and reproductive decision-making and behaviours of HIV-infected and – affected women. There remains, however, limited information concerning these associations given both the recency of HAART scale-up in settings with generalized HIV epidemics as well as the prevailing under-emphasis of reproductive health within HIV programming.  Using an approach that recognizes and supports the rights of all women, including women living with HIV, to safely achieve their reproductive goals [14, 15],  2  this thesis aims to investigate the impact of HAART on sexual and reproductive decision-making and behaviours among HIV-infected and –affected women in two high HIV prevalence settings in sub-Saharan Africa (Soweto, South Africa and Mbarara, Uganda) during a time of expanding HAART access.  1.2 Situational analysis of HIV infection and sexual and reproductive health in the context of expanding access to HAART in sub-Saharan Africa In sub-Saharan Africa, HIV is predominantly transmitted through heterosexual intercourse [1], which further entangles the impact of the HIV epidemic on women’s sexual and reproductive health. The common transmission pathway suggests that both the risk of HIV infection and pregnancy share similar risk factors and distribution in the female population. At the population level, however, substantial evidence indicates that in the absence of antiretroviral treatment, HIVinfected women in sub-Saharan Africa have between 25-40% lower fertility rates than non-infected women [3]. The decrease in fertility is more pronounced among women with advanced HIV disease and lower CD4 cell counts [16-18]. This fertility differential is largely explained by premature mortality among HIV-infected women and/or their sexual partners in addition to HIV-associated morbidity contributing to reduced fecundity [3].  3  Despite a noted relative lower fertility, each year the absolute number of pregnancies among women living with HIV in sub-Saharan Africa exceeds 1.4 million [10], of which an estimated 50-84% are unintended [19-21]. Many of these pregnancies contribute to distressing adverse outcomes for women, children, and their families. Every year nearly 350,000 infants are infected with HIV via mother-tochild transmission (MTCT) [1]. Maternal mortality, the world’s worst health inequity, is exacerbated in the context of HIV with reports indicating that HIVrelated maternal deaths have increased dramatically in regions of high HIV prevalence [22]. In addition, across sub-Saharan Africa, there are an estimated 8.9 million maternal orphans due to HIV/AIDS-associated mortality [23].  In 2002, the World Health Organization (WHO) highlighted HIV infection among infants and young children as a major global public health priority and recommended a four-pronged approach for preventing mother-to-child transmission of HIV (PMTCT): (i) primary prevention of HIV infection among women of childbearing age; (ii) preventing unintended pregnancies among HIVinfected women; (iii) preventing HIV transmission from women living with HIV to their infants and; (iv) providing appropriate treatment, care, and support to mothers living with HIV and their children and families [24]. While the recommendations support scaling-up this comprehensive range of interventions across the prepregnancy to postpartum continuum, most of the focus has centred on antenatal  4  HIV testing and the provision of antiretrovirals to prevent MTCT [10]. Indeed, others have argued that the prevailing view of PMTCT as a pediatric rather than maternal health issue has hampered efforts towards full realization of the WHO’s recommended strategy [25]. Particularly lacking has been an adequate consideration of the intersecting goals of supporting women’s overall sexual and reproductive health needs within HIV prevention, care, and treatment programming and vice versa.  The seven years since the WHO’s recommendations have witnessed an unprecedented global commitment and roll-out of antiretroviral therapy to people living with HIV in sub-Saharan Africa and other resource-limited settings. In 2008, over 1.7 million HIV-infected women in the region were receiving antiretroviral therapy, accounting for an estimated 46% of those in need of treatment [10]. In addition, nearly 45% of pregnant women living with HIV in sub-Saharan Africa received antiretroviral therapy to prevent HIV transmission to their infants, of whom nearly 10% were receiving HAART for their own health [10]. While there remains a high level of unmet need, this level of antiretroviral coverage represents a substantial increase from previous years. Between 2007 and 2008 alone, nearly one million more HIV-infected individuals in need of treatment in sub-Saharan Africa commenced antiretroviral therapy [10].  5  Through improvements in health and survival and reductions in the incidence of MTCT and sexual transmission of HIV, this era of expanding access to HAART has the potential to decrease risks and barriers to reproduction among women living with HIV. This emerging reality of HIV as a manageable chronic disease has corresponded with increasing calls for a rights- and evidenced-based approach to support the sexual and reproductive health needs of HIV-infected women [14, 15, 26, 27]. However, little is known about whether or how access to HIV treatment services may influence the sexual and reproductive health decisionmaking, behaviours, or outcomes of women in high HIV prevalence settings.  Available information about the role of HAART access in resource-rich settings (primarily among men who have sex with men (MSM)) have shown minimal impact of individual receipt of HAART on sexual risk behaviours [28]. However, these studies do report a significant relationship between beliefs about the role of HAART in reducing health and transmission risks (regardless of actual HAART use and HIV status) and the prevalence of unprotected sex [28]. If women in sub-Saharan Africa exhibit HAART-associated behaviour changes similar to those reported from high-income countries, changes in sexual behaviour may yield an increased risk of HIV transmission and pregnancy, especially in regions with both high HIV prevalence and fertility rates.  6  However, the generalizability of studies from high resource settings where HIV is concentrated within particular sub-populations to HIV-positive women in hyper-endemic sub-Saharan African settings remains uncertain. There remain great differences between resource-rich settings and sub-Saharan African countries with respect to HIV prevalence, the level of HAART accessibility, prevailing fertility rates and trends, access to HIV and reproductive and sexual health services, and sociocultural environments, which are likely to influence the direction and magnitude of the effect of HAART access on sexuality and reproduction. Moreover, while the potential health risks may dampen the reproductive goals of some HIV-positive women, stigma associated with childlessness in many sub-Saharan African societies [29] and the strong personal desires and community expectations for biological parenthood [30] remain potent drivers of pregnancy, irrespective of HIV status. In many such cultural contexts, remaining childless can be a societal norm violation more stigmatizing than HIV infection itself [30, 31].  Understanding the role that increasing access to HAART may have on sexual and reproductive decision-making and behaviours has implications for the longterm sustainability and effects of HAART scale-up initiatives. Recent reports have raised concern regarding potential shortages of HAART funding at global levels [32] concomitant with the largely stagnant resourcing for family planning and reproductive health programming in low-income countries [33]. Now is the time to  7  document the ways in which HAART may be impacting the sexual and reproductive decision-making and behaviours of HIV-affected women in feminized epidemic settings. In addition, this understanding will contribute critical information towards efforts aimed at improved integration between policies, programs, and services that intersect the sexual, reproductive, and HIV needs of women in high HIV prevalence settings.  Below, the context of HIV, HAART accessibility, and sexual and reproductive health in the two study settings (Soweto, South Africa and Mbarara, Uganda) is briefly described.  1.3 Brief situational analysis of HIV, HAART scale-up, and sexual and reproductive health in South Africa and Uganda  South Africa and Uganda represent important and appropriate countries within which to investigate the impact of HAART on sexual and reproductive decision-making and behaviours given the unique epidemiology and trends of the HIV epidemic in each country, the different levels of HAART coverage, and the diverse fertility rates and trends. In addition, each country has high quality clinical and research institutions committed to investigating the impact of HAART on their populations.  8  1.3.1 South Africa  With 5.5 million HIV-infected residents, South Africa has the world’s largest absolute number of people living with HIV. The national HIV prevalence among South African adults is 18.8%, a prevalence that has remained relatively stable over several years [1]. The epidemic is highly feminized as women currently account for nearly 60% of all HIV-infected adults [1] and among 15-24 year-olds, women account for 90% of incident HIV infections [34]. HIV prevalence in the study site Soweto, an urban township of Johannesburg, is among the highest in the country. An estimated 30% of women attending antenatal clinics in Soweto test HIV-positive [35].  Each year an estimated 220,000 women living with HIV in South Africa become pregnant [10]. Although the coverage of antiretroviral therapy to prevent MTCT increased from 15% in 2004 to 73% in 2008 [10], each year over 64,000 infants are infected with HIV via MTCT [36].  In 2006, the South African government released the ‘HIV & AIDS and STI Strategic Plan for South Africa, 2007-2011’ [37], a long overdue commitment to providing antiretroviral therapy (among other HIV treatment, prevention, and control strategies) to medically eligible South Africans. Although government and multinational efforts are improving access to antiretroviral therapy, there remains a 9  high level of unmet need. At the end of 2007, an estimated 700,000 adults were receiving ARV, accounting for approximately a third of those in need of treatment [10]. Despite a high level of unmet need, South Africa is operating the largest HIV treatment program in the world with an average of 16,000 people starting HAART each month.  The total fertility rate (TFR) of South Africa is 2.0, which is the lowest fertility rate in sub-Saharan Africa. Use of modern contraceptive methods is estimated at 65% among sexually active women [38]. Available estimates for Gauteng province (in which Soweto is located) estimate a TFR of 2.3 and a contraceptive prevalence rate of 63%, although there is substantial variation by population sub-group [38]. Nearly two-thirds of South African adults reported using a condom at last sex [39]. Family planning services in South Africa are widely and freely available and termination of pregnancy services are legal and available free-of-charge at designated primary health care facilities.  The South African portion of this thesis was conducted at the Perinatal HIV Research Unit (PHRU), which is housed in Chris Hani Baragwanath Hospital in Soweto. The PHRU, established in 1996, is one of Africa’s largest HIV/AIDS research and clinical service centres. Over the last decade, the PHRU’s focus has extended beyond prevention of MTCT services to include HIV treatment, prevention, and  10  psychosocial research and services for adults and children. The PHRU clinics see over 5,000 adult patient visits monthly and provide antiretroviral therapy and clinical care free-of-charge to medically eligible HIV-positive individuals.  1.3.2 Uganda  In Uganda, 46% of the estimated 1 million HIV-infected adults are women of reproductive age. National adult HIV prevalence is 6.7% and is significantly higher among women (7.5%) than among men (5%) [1]. Between the early 1990s to the early 2000s, Uganda was the first country in sub-Saharan Africa to show a decrease in HIV prevalence [40]. The drop in prevalence is largely attributed to changes in preventive sexual behaviors including an increase in primary and secondary sexual abstinence, a reduction in multiple concurrent partnerships, and an increase in condom use [41]. Since then, however, studies have reported a concerning recent rise in HIV incidence in some population sub-groups as well as an increase in the overall population prevalence [42].  The well-known ‘ABC’ approach (abstinence, be faithful, and condom use) has been the foundation of HIV prevention activities in the country. More recently, the ABC strategy has expanded to include Voluntary Testing and Counselling, PMTCT, and improved access to antiretroviral therapy and HIV care and support 11  services. An estimated 82,000 HIV-positive Ugandan women become pregnant each year and in 2008, only half received antiretrovirals for preventing MTCT [10].  In June 2004, Uganda began to offer free antiretroviral treatment to people living with HIV through the Ugandan Antiretroviral Drug Policy. However, only more recently through concerted global efforts have no-cost HAART services been made more widely available. In 2000 an estimated 1,000 Ugandans were receiving HAART compared with an estimated 140,000 at the end of 2008. The HAART coverage rate is 33% meaning that one-third of all those who need treatment are receiving it [10].  At 6.7 births per woman, Uganda has one of the total highest fertility rates in the world. Although levels vary by region, overall use of family planning by married women is low and estimated at 15%; with a sizable proportion of those reporting use of traditional methods [43]. In the absence of HAART, the Ugandan HIV epidemic has had a dramatic impact of population fertility. The combined effect of premature mortality and reduced fertility among HIV-positive women of reproductive age resulted in an estimated 700,000 fewer births than expected between 1980 and 2000. By the year 2000, this estimate represented 6% of all births that would have occurred in Uganda that year [3].  12  The Ugandan portion of this thesis emerges from studies conducted at the Mbarara Regional Referral Hospital located in Mbarara District in southwest Uganda and led by investigators at Harvard Medical School/Massachusetts General Hospital in Boston and the Mbarara University of Science and Technology. Mbarara is 265 km southwest of the capital city, Kampala. The hospital serves a population of approximately 1.1 million people and HIV prevalence in the region is estimated at 10% [44]. Housed within the Mbarara Regional Referral Hospital is the Mbarara University HIV clinic, the Immune Suppression Syndrome (ISS) Clinic. The ISS clinic currently serves more than 13,000 patients, 65% of whom are women. The clinic offers comprehensive HIV care services, including provision of HAART, free of charge. Approximately 35% of the clinic’s population is currently receiving HAART.  1.4 Outstanding Questions and Study Justification Unprotected sexual activity required for conception carries a risk of HIV transmission to uninfected sexual partners [45] and, among HIV-positive women who do become pregnant, carries a risk of mother-to-child-transmission of HIV [46, 47]. In addition, HIV-positive women have a lower life expectancy than negative women [48], increasing the risk of maternal orphanhood [49]. In light of these concerns, early reproductive guidelines for people living with HIV were dissuasive  13  [50] and HIV-positive women who express fertility desires continue to encounter community and healthcare worker disapproval [30, 51]. Despite these concerns, a plethora of evidence indicates that many women living with HIV continue to desire children [30, 52-55], become pregnant [19, 52, 53], and give birth [52, 53, 56] after knowing their HIV-positive status. Evidence also suggests that pregnancy is an important risk factor for acquiring HIV infection [57]. Through measured reductions in morbidity and mortality of HIV-infected individuals and reductions in the risk of vertical and horizontal HIV transmission, expanding access to HIV treatment services in high prevalence settings in subSaharan Africa is anticipated to impact sexual and reproductive decision-making and behaviours. To date, there is a paucity of empirical data that describes how increased access and receipt of HAART influences the underlying determinants of pregnancy including sexual behaviours, fertility decision-making, and contraceptive use. In addition, it remains unknown the degree to which use of HAART itself, enrolment in HAART treatment clinics and associated increased access to health care services, and/or the effect of HAART optimism is the primary pathway through which HAART may impact these aspects of sexuality and reproduction. Moreover, the few recently available studies neglect to consider duration of HAART use and have tended to conduct analyses only among HIV- positive women, without including comparisons with HIV-negative women from the same community [52-  14  54]. The lack of an HIV-negative comparison group in previous studies precludes the opportunity to assess whether HAART users begin to resemble HIV-negative women in their sexual and reproductive decision-making and behaviours, particularly as HIV is increasingly recognized as a manageable chronic disease. Moreover, neglecting to assess the role of duration of HAART precludes the opportunity to investigate how decisions and behaviours may change over increasing lengths of time on therapy.  In settings experiencing generalized HIV epidemics with a higher proportion of infections among women of reproductive age, there is a desperate need to understand the impact of expanding access to HAART on sexual and reproductive health. This information is critically required to help develop sexual and reproductive programs, services, and policies that support the rights of all women to safely achieve their reproductive goals, while minimizing risks to maternal, child, and partner health.  1.5 Purpose, Objectives, and Hypotheses  Given the paucity of available information, the overall purpose of this thesis is to examine whether sexual and reproductive decision-making and behaviours vary by HIV status and receipt of HAART among women in two HIV-endemic settings in 15  sub-Saharan Africa.  The study purpose is supported by the following five objectives: (1) To review the literature on the known and potential impact of antiretroviral therapy on the fertility of HIV-positive women in subSaharan Africa and to develop a conceptual framework to guide investigations aimed at understanding how widespread access to and use of antiretroviral therapy may impact fertility and its proximate and underlying determinants.  (2) To estimate the association between current HAART use and recent sexual activity among HAART-naïve and HAART-experienced HIV-positive women of reproductive age in three global settings with high HIV prevalence (Rio de Janeiro, Brazil, Soweto, South Africa, and Mbarara, Uganda). A secondary objective was to explore the association between current HAART use and sexual behaviours, including use of condoms and contraception. It was hypothesized that through improvements in health and well-being, receipt of HAART would be associated with an increased prevalence of recent sexual activity.  16  (3) To assess the prevalence of fertility intentions and to determine whether fertility intentions vary according to HIV status and use and duration of HAART among women of reproductive age in Soweto, South Africa. Based on the review and conceptual framework presented in Chapter 2, it was hypothesized that HIV-positive women would have lower fertility intentions compared with HIV-negative women. In addition, it was hypothesized that HIV-positive women receiving HAART would have higher fertility intentions than HAART-naïve women, with increasing duration of HAART treatment associated with incrementally higher fertility intentions. Overall, it was hypothesized that HAART use would be associated with narrower differences in fertility intentions between HIV-positive and HIV-negative women.  (4) To assess the prevalence of contraceptive use and to determine whether contraceptive use varies according to HIV status and use and duration of HAART among sexually active women in Soweto, South Africa. A secondary objective was to determine the types of contraceptive methods used and whether contraceptive method profiles vary by HIV status and HAART use. Based on the review and conceptual framework presented in Chapter 2, it was hypothesized that HIV-positive women receiving HAART  17  would be less likely to use contraception compared with HIV-positive women not receiving HAART, with increasing duration of HAART use associated with incrementally lower contraceptive use. It was also hypothesized that contraceptive prevalence among HIV-positive women receiving HAART and HIV-negative women would be similar.  (5) To develop a reliable HAART optimism scale for HIV-positive women in high HIV prevalence settings and to test the scale’s validity against a measure of fertility intentions and two measures of sexual behaviour (i.e., sexual abstinence and unprotected sex) among HAART-naïve and HAARTexperienced women in Mbarara, Uganda. Upon development of a reliable HAART optimism scale, it was hypothesized that HIV-positive women with higher HAART optimism would be more likely to report intentions to have children and would be more likely to be sexually active and practice risky sexual behaviours.  18  1.6 Overview of thesis  This manuscript-based thesis is comprised of seven separate chapters that represent a progression of investigations that address the objectives listed above. This first chapter provides background regarding the feminization of the HIV pandemic in sub-Saharan Africa, intersections between HIV and sexual and reproductive health, an overview of global HAART scale-up efforts and potential impact on sexual and reproductive decision-making and behaviours, and related background to the study settings in Soweto and Mbarara. In addition, this chapter lists the overall purpose and five associated objectives of this thesis.  Chapter 2, published in Current HIV/AIDS Reports in 2006 [58], provides a literature review of the known and potential impacts of highly active antiretroviral therapy on fertility of HIV-positive women in high HIV prevalence settings of subSaharan Africa. The review was conducted in 2005 and is therefore limited to data available at that time. Using Bongaarts’ proximate determinants of fertility framework (adapted for conditions of generalized HIV epidemics) [59], the review paper proposes a conceptual model to examine how use of HAART may mediate the impact of HIV infection on fertility through biological and behavioral determinants. The conceptual framework highlights related research priorities and guided the development of subsequent epidemiological thesis research presented in Chapters 3  19  to 6. Each of these chapters provides an updated review of the literature specific to the determinant of fertility under investigation.  Chapter 3, published in AIDS Care in January 2008 [60], presents the results of a 2005 pilot study that investigated the impact of HAART on recent sexual activity among HIV-positive women in three global settings: Rio de Janiero, Brazil; Soweto, South Africa; and Mbarara, Uganda. In addition to the findings and interpretations presented, the pilot study informed the subsequent development of larger HIV, HAART, and sexual and reproductive health-related epidemiological studies in each of the three sites. Data from the Brazil study are not further included in this thesis.  Chapters 4 (in press at the American Journal of Public Health [61]), 5 (under review), and 6 (published in AIDS and Behavior in 2009 [62]) provide detailed epidemiologic investigations of the relationships between HIV, HAART, and specific sexual and reproductive decision-making and behaviours. The analyses in Chapters 4 and 5 relate to the impact of HAART on fertility intentions and contraceptive use, respectively. These findings emerged from a study that the candidate developed and led in collaboration with the Perinatal HIV Research Unit (PHRU) in Soweto, South Africa. The analysis in Chapter 6 relates to the development and testing of a HAART optimism scale. These findings emerged from a study led by investigators at Harvard Medical School/Massachusetts General  20  Hospital and the Mbarara University of Science and Technology in Mbarara, Uganda. Chapter 7 provides a summary of the key findings, unique contributions of this dissertation, recommendations, and priorities for future research.  21  1.7 References  1.  UNAIDS. AIDS epidemic update: December 2007. Geneva: Joint United Nations Programme on HIV/AIDS (UNAIDS) and World Health Organization (WHO). 2007.  2.  Allen S. Why is fertility an issue for HIV-infected and at-risk women? J Acquir Immune Defic Syndr 2005; 38 Suppl 1:S1-3.  3.  Lewis JJ, Ronsmans C, Ezeh A, Gregson S. The population impact of HIV on fertility in sub-Saharan Africa. AIDS 2004; 18 Suppl 2:S35-43.  4.  Zaba B, Gregson S. Measuring the impact of HIV on fertility in Africa. AIDS 1998; 12 Suppl 1:S41-50.  5.  World Health Organization (WHO). HIV in pregnancy: A review. Geneva: World Health Organization (WHO); 1998.  6.  Hogg RS, Heath KV, Yip B, Craib KJ, O'Shaughnessy MV, Schechter MT, et al. Improved survival among HIV-infected individuals following initiation of antiretroviral therapy. JAMA 1998; 279:450-4.  7.  Palella FJ, Jr., Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, et al. Declining morbidity and mortality among patients with advanced  22  human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998; 338:853-60.  8.  Seyler C, Anglaret X, Dakoury-Dogbo N, Messou E, Toure S, Danel C, et al. Medium-term survival, morbidity and immunovirological evolution in HIVinfected adults receiving antiretroviral therapy, Abidjan, Cote d'Ivoire. Antivir Ther 2003; 8:385-93.  9.  Wools-Kaloustian K, Kimaiyo S, Diero L, Siika A, Sidle J, Yiannoutsos CT, et al. Viability and effectiveness of large-scale HIV treatment initiatives in subSaharan Africa: experience from western Kenya. AIDS 2006; 20:41-8.  10.  World Health Organization (WHO). Towards Universal Access: Scaling up priority HIV//AIDS interventions in the health sector - Progress Report 2009. Geneva: WHO; 2009.  11.  Cooper ER, Charurat M, Mofenson L, Hanson IC, Pitt J, Diaz C, et al. Combination antiretroviral strategies for the treatment of pregnant HIV-1infected women and prevention of perinatal HIV-1 transmission. J Acquir Immune Defic Syndr 2002; 29:484-94.  12.  Castilla J, Del Romero J, Hernando V, Marincovich B, Garcia S, Rodriguez C. Effectiveness of highly active antiretroviral therapy in reducing heterosexual transmission of HIV. J Acquir Immune Defic Syndr 2005; 40:96-101.  23  13.  Fang CT, Hsu HM, Twu SJ, Chen MY, Chang YY, Hwang JS, et al. Decreased HIV transmission after a policy of providing free access to highly active antiretroviral therapy in Taiwan. J Infect Dis 2004; 190:879-85.  14.  The Global Network of People Living with HIV/AIDS (GNP+). Advancing the sexual and reproductive health and human rights of people living with HIV: A guidance package; 2009. Available from: http://www.who.int/reproductivehealth/publications/linkages/9789490241018 /en/index.html. Accessed: November 15th, 2009.  15.  Gruskin S, Ferguson L, O'Malley J. Ensuring sexual and reproductive health for people living with HIV: an overview of key human rights, policy and health systems issues. Reprod Health Matters 2007; 15:4-26.  16.  Ross A, Van der Paal L, Lubega R, Mayanja BN, Shafer LA, Whitworth J. HIV-1 disease progression and fertility: the incidence of recognized pregnancy and pregnancy outcome in Uganda. AIDS 2004; 18:799-804.  17.  Loko MA, Toure S, Dakoury-Dogbo N, Gabillard D, Leroy V, Anglaret X. Decreasing incidence of pregnancy by decreasing CD4 cell count in HIVinfected women in Cote d'Ivoire: a 7-year cohort study. AIDS 2005; 19:443-5.  18.  Sedgh G, Larsen U, Spiegelman D, Msamanga G, Fawzi WW. HIV-1 disease progression and fertility in Dar es Salaam, Tanzania. J Acquir Immune Defic Syndr 2005; 39:439-45.  24  19.  Desgrees-Du-Lou A, Msellati P, Viho I, Yao A, Yapi D, Kassi P, et al. Contraceptive use, protected sexual intercourse and incidence of pregnancies among African HIV-infected women. DITRAME ANRS 049 Project, Abidjan 1995-2000. Int J STD AIDS 2002; 13:462-8.  20.  Laher F, Todd CS, Stibich MA, Phofa R, Behane X, Mohapi L, et al. A Qualitative Assessment of Decisions Affecting Contraceptive Utilization and Fertility Intentions among HIV-Positive Women in Soweto, South Africa. AIDS Behav 2009.  21.  Rochat TJ, Richter LM, Doll HA, Buthelezi NP, Tomkins A, Stein A. Depression among pregnant rural South African women undergoing HIV testing. JAMA 2006; 295:1376-8.  22.  National Committee on Confidential Enquiries into Maternal Deaths. Saving Mothers 1999-2001. Pretoria: Department of Health, South Africa; 2003.  23.  UNAIDS. Children on the Brink 2004: A joint report of new orphan estimates and a framework for action; 2004. Available from: http://www.unicef.org/publications/files/cob_layout6-013.pdf. Accessed: November 10th, 2008.  24.  World Health Organization (WHO). Prevention of HIV in infants and young children: review of evidence and WHO's activities. Geneva: World Health Organization (WHO); 2002.  25  25.  McIntyre JA. Sex, pregnancy, hormones, and HIV. Lancet 2005; 366:1141-2.  26.  World Health Organization (WHO). Sexual and reproductive health of women living with HIV/AIDS: guidelines on care, treatment, and support for women living with HIV/AIDS and their children in resource-constrained settings. Geneva: World Health Organization; 2006.  27.  Berer M. HIV/AIDS, sexual and reproductive health: intersections and implications for national programmes. Health Policy Plan 2004; 19 Suppl 1:i62i70.  28.  Crepaz N, Hart TA, Marks G. Highly active antiretroviral therapy and sexual risk behavior: a meta-analytic review. JAMA 2004; 292:224-36.  29.  Dyer SJ. The value of children in African countries: insights from studies on infertility. J Psychosom Obstet Gynaecol 2007; 28:69-77.  30.  Cooper D, Harries J, Myer L, Orner P, Bracken H, Zweigenthal V. "Life is still going on": reproductive intentions among HIV-positive women and men in South Africa. Soc Sci Med 2007; 65:274-83.  31.  Dyer SJ, Abrahams N, Hoffman M, van der Spuy ZM. 'Men leave me as I cannot have children': women's experiences with involuntary childlessness. Hum Reprod 2002; 17:1663-8.  26  32.  Ahmed K, Smith J, Whiteside A, Global funding of the AIDS response, in Newsletter of the International AIDS Society. 2009. p. 8-9.  33.  UN News Centre, UN and World Bank voice concerns about funding for family planning, in UN News Centre. 2009: New York.  34.  Rehle T, Shisana O, Pillay V, Zuma K, Puren A, Parker W. National HIV incidence measures--new insights into the South African epidemic. S Afr Med J 2007; 97:194-9.  35.  South African Department of Health. The national HIV and syphilis prevalence survey, South Africa 2007. Pretoria: South African Department of Health; 2008.  36.  Barker PM, Venter F. Setting district-based annual targets for HAART and PMTCT - a first step in planning effective intervention for the HIV/AIDS epidemic. S Afr Med J 2007; 97:916-7.  37.  South Africa Department of Health. HIV & AIDS and STI Strategic Plan for South Africa, 2007-2011. Pretoria: South Africa Department of Health; 2006.  38.  SADHS. South Africa Demographic and Health Survey 2003. Pretoria: Department of Health; 2004.  39.  Shisana O, Rehle T, Simbayi L, Zuma K, Jooste S, Pillay-van-Wyk V, et al. South African national HIV prevalence, incidence, behaviour and communication  27  survey 2008: A turning tide among teenagers? Cape Town: Human Sciences Research Council (HSRC); 2009.  40.  Asamoah-Odei E, Garcia Calleja JM, Boerma JT. HIV prevalence and trends in sub-Saharan Africa: no decline and large subregional differences. Lancet 2004; 364:35-40.  41.  Kirungi WL, Musinguzi J, Madraa E, Mulumba N, Callejja T, Ghys P, et al. Trends in antenatal HIV prevalence in urban Uganda associated with uptake of preventive sexual behaviour. Sex Transm Infect 2006; 82 Suppl 1:i36-41.  42.  Shafer LA, Biraro S, Nakiyingi-Miiro J, Kamali A, Ssematimba D, Ouma J, et al. HIV prevalence and incidence are no longer falling in southwest Uganda: evidence from a rural population cohort 1989-2005. AIDS 2008; 22:1641-9.  43.  Uganda Bureau of Statistics and Macro International Inc. Uganda Demographic and Health Survey 2006. Calverton: Uganda Bureau of Statistics and Macro International Inc.; 2007.  44.  Ugandan Ministry of Health. Uganda HIV/AIDS sero-behavioral survey, 20042005. Calverton: Ministry of Health/ORC Macro; 2006.  45.  Wawer MJ, Gray RH, Sewankambo NK, Serwadda D, Li X, Laeyendecker O, et al. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis 2005; 191:1403-9.  28  46.  European Collaborative Study. HIV-infected pregnant women and vertical transmission in Europe since 1986. European collaborative study. AIDS 2001; 15:761-70.  47.  Mofenson LM. Mother-child HIV-1 transmission: Timing and determinants. Obstet Gynecol Clin North Am 1997; 24:759-84.  48.  Nunn AJ, Mulder DW, Kamali A, Ruberantwari A, Kengeya-Kayondo JF, Whitworth J. Mortality associated with HIV-1 infection over five years in a rural Ugandan population: cohort study. BMJ 1997; 315:767-71.  49.  Makumbi FE, Gray RH, Serwadda D, Nalugoda F, Kiddugavu M, Sewankambo NK, et al. The incidence and prevalence of orphanhood associated with parental HIV infection: a population-based study in Rakai, Uganda. AIDS 2005; 19:1669-76.  50.  CDC. Recommendations for assisting in the prevention of perinatal transmission of human T-lymphotropic virus type III/lymphadenopathyassociated virus and acquired immunodeficiency syndrome. MMWR Morb Mortal Wkly Rep 1985; 34:721-6, 731-2.  51.  Myer L, Morroni C, Cooper D. Community attitudes towards sexual activity and childbearing by HIV-positive people in South Africa. AIDS Care 2006; 18:772-6.  29  52.  Homsy J, Bunnell R, Moore D, King R, Malamba S, Nakityo R, et al. Reproductive intentions and outcomes among women on antiretroviral therapy in rural Uganda: a prospective cohort study. PLoS ONE 2009; 4:e4149.  53.  Maier M, Andia I, Emenyonu N, Guzman D, Kaida A, Pepper L, et al. Antiretroviral therapy is associated with increased fertility desire, but not pregnancy or live birth, among HIV+ women in an early HIV treatment program in rural Uganda. AIDS Behav 2009; 13 Suppl 1:28-37.  54.  Myer L, Morroni C, Rebe K. Prevalence and determinants of fertility intentions of HIV-infected women and men receiving antiretroviral therapy in South Africa. AIDS Patient Care STDS 2007; 21:278-85.  55.  Nakayiwa S, Abang B, Packel L, Lifshay J, Purcell DW, King R, et al. Desire for children and pregnancy risk behavior among HIV-infected men and women in Uganda. AIDS Behav 2006; 10:S95-104.  56.  Gray GE, McIntyre JA. HIV and pregnancy. BMJ 2007; 334:950-3.  57.  Pettifor AE, Measham DM, Rees HV, Padian NS. Sexual power and HIV risk, South Africa. Emerg Infect Dis 2004; 10:1996-2004.  58.  Kaida A, Andia I, Maier M, Strathdee SA, Bangsberg DR, Spiegel J, et al. The potential impact of antiretroviral therapy on fertility in sub-Saharan Africa. Curr HIV/AIDS Rep 2006; 3:187-94.  30  59.  Bongaarts J, Potter RG. Fertility, biology, and behavior: An analysis of the proximate determinants. New York: Academic Press; 1983.  60.  Kaida A, Gray G, Bastos FI, Andia I, Maier M, McIntyre J, et al. The relationship between HAART use and sexual activity among HIV-positive women of reproductive age in Brazil, South Africa, and Uganda. AIDS Care 2008; 20:21-5.  61.  Kaida A, Laher F, Strathdee SA, Janssen PA, Money D, Hogg RS, et al. Fertility intentions of HIV-positive women of reproductive age in Soweto, South Africa: The influence of expanding access to HAART in an HIV hyperendemic setting. Am J Public Health In press.  62.  Kaida A, Lima VD, Andia I, Kabakyenga J, Mbabazi P, Emenyonu N, et al. The WHOMEN's scale (Women's HAART Optimism Monitoring and EvaluatioN Scale v.1) and the association with fertility intentions and sexual behaviours among HIV-positive women in Uganda. AIDS Behav 2009; 13 Suppl 1:72-81.  31  CHAPTER 21: THE POTENTIAL IMPACT OF ANTIRETROVIRAL THERAPY ON FERTILITY IN SUB-SAHARAN AFRICA  2.1 INTRODUCTION  Women of reproductive age comprise 46% of the world’s HIV-infected adult population. The majority live in sub-Saharan Africa (SSA), where approximately 14 million women of child-bearing age are currently living with HIV/AIDS [1]. HIV infection in these women raises important individual and public health concerns about the incidence of pregnancy. Such concerns relate to the survival and health of the mother and the health of her baby, including risk of vertical disease transmission, orphanhood, and other adverse pregnancy outcomes [2].  In SSA, HIV is predominantly transmitted through heterosexual intercourse [1], which further entangles the impact of the HIV epidemic on fertility. The common pathway suggests that both the risk of HIV infection and pregnancy may share similar risk factors and distribution in the population. Yet, substantial evidence indicates that HIV-infected women in SSA have between 25-40% lower 1  A version of this chapter has been published. Kaida A, Andia I, Maier M, Strathdee SA, Bangsberg DR, Spiegel J, Bastos FI, Gray G, Hogg R. The potential impact of antiretroviral therapy on fertility in sub-Saharan Africa. Current HIV/AIDS Reports 2006; 3:177-184.  32  fertility than non-infected women [3,4]. It should be noted that the term fertility is used here in the demographic sense, as in the actual level of reproduction based on the number of live births that occur, rather than the biologic capability to reproduce (termed fecundity).  The evidence for reduced fertility arises from settings with limited (or no) access to HIV/AIDS treatment. Recently, however, access to antiretroviral (ARV) treatment in SSA is improving considerably [5,6]. The potential impact of ARV therapy on the survival and quality of life of individuals with HIV infection has been demonstrated in numerous studies from developed and developing countries [7-10]. As ARV therapy becomes increasingly accessible in SSA, it is important to understand whether and how the associated clinical improvements correspond with changes in the incidence of pregnancy and fertility among women with HIV infection.  Thus, the purpose of this article is to review the literature regarding the potential impact of ARV therapy on fertility in SSA. First, evidence for lower fertility among women with HIV infection (not on therapy) compared with uninfected women is briefly reviewed. Then, the literature on ARV therapy and fertility in SSA is examined. Owing to limited empirical evidence from this setting, the evidence from developed countries is reviewed and implications for developing regions are  33  discussed. Next, Bongaart’s proximate determinants of fertility framework (adapted for conditions of a generalized HIV epidemic) [11,12] is used to examine the underlying mechanisms through which use of ARV therapy may impact the fertility of HIV-infected women in SSA. Finally, a conceptual framework is proposed to guide future research aimed at understanding how widespread use of ARV therapy may impact fertility in sub-Saharan Africa.  2.2 HIV AND FERTILITY IN SUB-SAHARAN AFRICA  In societies with low contraceptive use, women with HIV infection have lower fertility than non-infected women. This finding is explained by increases in HIV-associated mortality and both the direct biological effects of HIV on the fecundity of infected women and the indirect impact of the virus on the behavioural determinants of fertility [3,4].  Two review articles have examined the relationship between HIV and fertility in SSA [3,4]. In 1998, Zaba and Gregson [3] reviewed six studies from SSA and concluded that HIV-infected women have between 25-40% lower fertility than noninfected women for most age groups. Only in the youngest age group (15-19 years) was HIV-infection associated with higher fertility, presumably due to selection for early sexual activity. A more recent review of an additional 13 studies similarly 34  concluded that HIV-infected women in SSA experienced lower fertility than their non-infected counterparts [4]. The authors estimated that each 1% increase in population prevalence of HIV in SSA countries results in a population-attributable decline in fertility of 0.37% (95% CI: 0.30%, 0.44%) [4]. The investigators used mathematical modelling to demonstrate the substantial impact of such a reduction in population fertility. They estimated that the combined effect of increased mortality among women with HIV infection and an assumed 20% reduction in fertility resulted in 700,000 fewer births in Uganda between 1980 and 2000.  These findings have been further substantiated by recent studies from SSA that report similar decreases in fertility among HIV-infected women compared with uninfected women [13,14]. Recent studies have also shown that fertility decreases dramatically by HIV disease progression [14,15] and decreasing CD4 cell counts [1416].  2.3 ARV THERAPY AND FERTILITY IN SUB-SAHARAN AFRICA  Between June and December 2004, the number of people on ARV therapy in SSA doubled from 150,000 to 310,000 [6]. Although coverage varies dramatically by country, in Botswana, Namibia, and Uganda greater than 25% of those who need  35  treatment are receiving it; 13 other countries in the region have greater than 10% coverage [6].  As ARV therapy becomes increasing accessible in SSA, the associated improvements in health, quality of life, and survival are anticipated to influence both the biological and behavioural fertility determinants of infected women. There remains, however, little empirical evidence to support this claim due, in part, to the recency of ARV treatment programs in the region.  2.4 ARV THERAPY AND FERTILITY IN DEVELOPED COUNTRIES  Research in developed countries suggests that HIV is associated with a decline in fertility and that ARV therapy reverses this decline. Studies conducted before the widespread availability of ARV therapy have shown that HIV-infected women in developed countries are less likely to become pregnant [17-21] and give birth [21,22] compared with uninfected women. Studies also report that the incidence of pregnancy and livebirth declines as AIDS develops [23] and that women with HIV infection suffer higher rates of adverse outcomes (including voluntary and spontaneous abortions) than uninfected women [19-21].  36  Current hypotheses suggest that the observed decline in fertility will be largely reversed with the introduction of highly active antiretroviral therapy (HAART), however, available studies have yielded somewhat differing results.  Blair et al [24] found that HIV-infected women in the United States were 20% more likely to become pregnant in the “HAART era” (1997-2001) than in previous years (1992-1996) (adjusted Relative Risk (ARR): 1.2; 95% CI: 1.1, 1.4). The higher pregnancy rate during the HAART era was thought to be due to both increased survival times and delayed progression to AIDS, which resulted in more opportunities to become pregnant.  Another American cohort study of HIV-infected women followed between 1994 and 2002 revealed somewhat different results. Massad et al [25] reported that use of ARVs at baseline was associated with a decreased incidence of pregnancy compared with HIV-infected women not on therapy (Odds Ratio (OR): 0.34; 95% CI: 0.49-0.98 for mono- or combination therapy; OR: 0.34; 95% CI: 0.03-4.28 for HAART). The study also reported that induced abortion became less common in the later HAART era (1999-2002) compared with earlier periods. The net impact of these findings (i.e., decreased incidence of pregnancy and increased incidence of livebirth) on fertility in the HAART era remains uncertain.  37  Finally, a European prospective cohort study of HIV-infected women followed between 1985 and 1998 found no statistically significant increased trend in the age-adjusted incidence of pregnancy after HIV diagnosis over time [17]. In particular, the rate of pregnancy did not change in the HAART era (1997-1998) compared with earlier periods. However, because the study included only the first two years of widespread HAART use, it may be too early to conclude that HAART did not influence reproductive decision-making.  Delineating the impact of HAART on fertility is complicated by the lack of information about changes in other determinants of fertility over the follow-up period of these studies. It is unclear, for example, whether women altered contraceptive or sexual practices or experienced changes in other variables important to understanding fertility differentials. Indeed, an observed increase in fertility may be a reaction to the availability of treatment to prevent mother-to-childtransmission (MTCT), rather than to the impact of HAART. Furthermore, the overall low fertility rate and HIV prevalence in these settings, compared with rates in SSA, may minimize the observed differential impact of ARVs on fertility. There may also be important cultural differences that influence fertility decisions in various contexts, including societal expectations to bear children [26]. Thus, while informative, the prevailing evidence from developed countries may not be fully applicable to settings in SSA.  38  2.5 BONGAARTS’ PROXIMATE DETERMINANTS OF FERTILITY FRAMEWORK: UNDERSTANDING THE MECHANISMS THROUGH WHICH ARV THERAPY MAY INFLUENCE FERTILITY We used Bongaarts’ proximate determinants of fertility framework (adapted for conditions of a generalised HIV epidemic) to explore the underlying mechanisms through which ARV therapy may influence fertility of HIV-infected women in SSA [11,12]. An overview of the framework is shown in Figure 2.1.  The proximate determinants of fertility are the biological and behavioural factors that directly influence fertility over a woman’s reproductive life span. It is through these proximate determinants that broader social, economic, and cultural variables act to affect fertility. As shown in Figure 2.1, two proximate determinants, marriage (or stable sexual union) and onset of permanent sterility, determine the duration of a woman’s reproductive period. The other five proximate determinants influence the rate of childbearing and the duration of birth intervals. These include post-partum infecundability, natural fecundability or frequency of intercourse, use and effectiveness of contraception, spontaneous intrauterine mortality, and induced abortion. A thorough description of these seven proximate determinants has been provided elsewhere [11].  Other investigators have adapted Bongaarts’ framework to describe how HIV influences the proximate determinants to impact fertility in SSA [27,28]. The  39  following section outlines how the use of ARV therapy may mediate the influence of HIV on fertility through the proximate determinants. This discussion is summarized in Table 2.1.  2.5.1 Marriage/Sexual debut  Bongaarts and Potter [11] denoted marriage, a proxy for sexual debut and sexual activity, as the time on the reproductive life span when a woman is most “at risk” for pregnancy. Since then, however, data on sexual debut and activity have become increasingly available and investigators have advised using these variables instead [29]. This may be particularly pertinent in some regions of SSA where premarital fertility is common [30,31].  Whichever the measure, it is clear that the HIV epidemic is altering many of the basic demographic features that comprise this proximate determinant of fertility in SSA. In many countries, the epidemic is increasing both the age of sexual debut and age at first marriage [32-34], which shortens the reproductive life span, thereby lowering fertility [11,27,28].  For already married couples, reports suggest that the HIV epidemic is shortening the duration of marriage through increased incidence of divorce and  40  widowhood [35-37]. Since birth rates are lower outside of marriage [27], high rates of divorce and widowhood decrease fertility. Reports from Uganda suggest that widows and divorced women find it more difficult to remarry due to fears of infecting new partners [28,36]. Fears of infection may also be reducing pre- and extra-marital sexual relations and polygamous unions [28]. This increase in monogamous relationships may lead to higher fertility [28,38].  How ARVs mitigate the relationship between HIV and age at marriage/sexual debut remains unknown. It is particularly difficult to hypothesize the direction of ARV influence because these are population-level rather than individual-level determinants. As such, even as ARV therapy becomes more widely available to HIV-infected individuals in SSA, it may be unlikely to drastically alter populationlevel trends for increasing age at sexual debut and marriage in the short term of the epidemic. However, by increasing the survival time and health of infected individuals, ARVs are hypothesized to prolong the duration of marriage and extend the reproductive period, thereby potentially increasing fertility.  2.5.2 Onset of permanent sterility  The onset of permanent sterility demarcates the end of a woman’s reproductive period [11]. Untreated sexually transmitted infections (STIs) can lead to 41  earlier onset of permanent sterility [39-41]. Studies from SSA reveal that HIVinfected women have a higher rate of prior syphilis infection [42] and current STIs [43] than HIV-negative women, which can cause sterility, especially if left untreated. Earlier onset of permanent sterility reduces fertility [11].  The STI treatment campaigns that have accompanied HIV prevention efforts in Tanzania and Uganda have reduced the population-level incidence of STIs [44,45]. In the absence of increased rates of contraceptive use, such reductions in untreated STIs are anticipated to increase fertility [27,28,46].  The impact of widespread use of ARVs on this proximate determinant is currently unknown. It is anticipated, however, that increased access to ARVs will increase patient contact with the healthcare system thereby increasing opportunities for STI education, testing, and treatment. Collectively, a decrease in the incidence of untreated STIs may increase fertility. In the absence of STI treatment services, however, the availability of ARV therapy will not impact sub-fertility associated with STIs.  42  2.5.3 Post-partum infecundability  Both breastfeeding and post-partum abstinence from sexual intercourse contribute to post-partum infecundability [11] and are important determinants of birth interval length [47,48].  In an attempt to minimize the risk of MTCT of HIV, there is evidence that some infected women decide to reduce or forego breastfeeding [34]. Such reductions in breastfeeding may shorten a women’s short-term infecundability, leading to earlier pregnancy and higher fertility [27,28].  Women in SSA have historically practiced periods of post-partum abstinence. Recent reports indicate that the mean duration of postpartum abstinence varies substantially by country, ranging from a low of four months in Zimbabwe to a high of 19 months in Burkina Faso [49]. Reports from Zimbabwe and Benin suggest that husbands may be more likely to engage in extra-marital sex during this period [34,50]. Anecdotal evidence suggests that in regions with high HIV prevalence, women may shorten the duration of post-partum abstinence to decrease their chances of acquiring HIV through the extra-marital activities of her husband [27,28]. Shorter periods of post-partum abstinence are hypothesized to increase fertility [3,28,51].  43  The impact of widespread ARV therapy on breastfeeding practises is largely unknown. Recent evidence suggests that provision of ARVs during late pregnancy and lactation lowers the viral load in breast milk and substantially reduces the risk of transmission by breastfeeding [52]. If future studies corroborate these results, women on ARV therapy in SSA may be encouraged to initiate and prolong breastfeeding at lower risk to their infants. Increased rates of breastfeeding can be expected to result in decreased fertility [47].  The impact of ARV therapy on duration of post-partum abstinence is currently unknown. There may, however, be a relationship between general abstinence and ARV therapy. A study from Cote d’Ivoire reported that HIV-infected women on ARVs were less likely to be sexually abstinent (for 6 or more months) compared with women not on therapy, although the relationship was only marginally significant (53% vs. 63%, respectively; p=0.08) [53]. A similar pattern was reported for HIV-infected men. If additional evidence supports the finding that ARV therapy reduces rates of sexual abstinence, fertility may be expected to increase, in the absence of contraceptive use.  44  2.5.4 Natural fecundability or frequency of intercourse  The components of this proximate determinant of fertility include coital frequency, incidence of amenorrhea, and natural fecundability (including spermatogenesis), all three of which are impacted by HIV.  In the absence of treatment, HIV-infected women have lower levels of coital frequency owing to higher morbidity in the women [54] or their male partners [4]. Studies from Uganda [54,55] and North America [56] have shown that HIV significantly increases the incidence of amenorrhea. Other studies have shown that HIV reduces the fecundability of women with infected partners due to decreased sperm production [57,58]. In sum, available evidence suggests that HIV functions to decrease coital frequency, increase incidence of amenorrhea, and decrease sperm production of male sexual partners, and thus decreases fertility.  The health improvements associated with the use of ARV therapy are anticipated to increase coital frequency [59]. The impact on amenorrhea and spermatogenesis remains unknown but is expected to result in improvements due to reduced morbidity [60]. Thus, the net effect of ARV therapy on this proximate determinant may be to increase fertility.  45  2.5.5 Use and effectiveness of contraception  Use of effective contraception is a main determinant of fertility [11]. Survey data from SSA countries with high HIV prevalence reveal increased rates of condom use to protect against STIs and HIV [28,61,62]. If this increased prevalence of condom use is among previous non-contraceptive users, then it can be expected to decrease fertility. If however, increased condom use occurs at the expense of more effective contraceptive methods (such as oral contraceptives), this may result in an increase in fertility [28,34].  Women with HIV infection may be concerned about MTCT and the potential for leaving orphans and thus may be more likely to use hormonal contraceptives to prevent pregnancy [63-65]. These actions would serve to decrease fertility [28]. On the other hand, owing to high infant mortality associated with HIV/AIDS, women may want more children to maximize the chances that at least a few will survive and/or to replace children who have previously died [27]. Such actions will serve to increase fertility.  The impact of ARV therapy on condom use is unknown. It is hypothesized, however, that the improved health associated with ARVs will increase desire for children since a woman may feel more confident that she and her children will survive [66]. This may result in decreased contraceptive use and thus increased  46  fertility. Similarly, it is hypothesized that women on ARV therapy will be less likely to use contraception and undergo fewer abortions due to reduced fears of infant infection and orphanhood, thereby increasing fertility [27,28].  2.5.6 Spontaneous intra-uterine mortality  Incidence of spontaneous abortion and stillbirths are important determinants of the rate of childbearing and the birth interval length [11]. The limited available evidence supports a possible increased risk of both spontaneous abortion and stillbirth among HIV-infected women in developing countries. A systematic review reported that HIV-infected women in developing countries had a statistically significant increased risk of stillbirth [summary OR: 4.15; 95% CI: 2.81-6.15] and spontaneous abortion [summary OR: 4.05; 95% CI: 2.75-5.96] compared with noninfected women [67]. Higher incidence of spontaneous abortion and stillbirth translates to decreased fertility [11].  The evidence regarding the impact of ARVs on spontaneous intra-uterine mortality remains inconclusive. For instance, while there are reports that HIVinfected women treated with HAART prior to pregnancy have a significantly higher risk of experiencing fetal death [68], and miscarriage [20], other studies have not found an increased incidence of stillbirths [69]. In spite of inconclusive findings,  47  investigators have concluded that the overall risks of adverse pregnancy outcomes that are attributable to ARV therapy are low and are likely to be outweighed by the known benefits of therapy during pregnancy [70,71]. In the absence of conclusive results on the impact of ARVs on spontaneous intra-uterine mortality, the subsequent impact on fertility remains unknown.  2.5.7 Induced abortion  Incidence of induced abortion is the final proximate determinant of fertility [11]. Studies from Uganda [54], Cote d’Ivoire [72], Cameroon and Kenya [62] report higher rates of induced abortion among HIV-infected women compared with HIVnegative women, which decreases fertility [28]. Induced abortions may be more likely among HIV-infected women due to concerns about vertical transmission and fear of leaving orphans [27,28].  The impact of ARVs on rates of induced abortion in SSA is largely unknown. Results from an American study revealed that abortion became less common after the introduction of HAART [25]. This may reflect women’s improved outlook for their own health and survival and reduced concerns about the risks of MTCT. It is therefore hypothesized that the widespread availability of ARV therapy in SSA will similarly result in a reduced rate of induced abortion, thereby increasing fertility.  48  2.6 CONCEPTUAL FRAMEWORK FOR ANALYSING THE POTENTIAL NET IMPACT OF ARV THERAPY ON FERTILITY  The previous section provided a synthesis of the real and hypothesized impacts of ARVs on the behavioural and biological proximate determinants of fertility in the context of high HIV prevalence. As shown, ARV therapy acts to orient some determinants towards increased fertility, whereas others offset this effect. For yet other proximate determinants, the impact of ARVs remains unknown. Of course, the net impact on fertility is influenced by social, economic, cultural, and genderrelated factors beyond simply HIV status and ARV use. Thus, the following framework is proposed to conceptualize the hierarchy of factors that determine fertility and to identify where and how ARVs may operate within this hierarchy.  In Figure 2.2, the proximate determinants have been separated into their biological and behavioural components and are oriented in terms of their influence on fertility in the presence of a generalized HIV epidemic. As shown, the proximate determinants directly influence fertility at varying stages within a woman’s reproductive life span, either affecting the duration of the reproductive period or the rate of childbearing and birth intervals.  The proximate determinants are influenced by broader social, economic, cultural, and gender-related determinants of fertility (e.g., economic development,  49  urbanization, and maternal education). As depicted in the framework, these broader determinants operate at three levels, the individual (which includes fertility goals and desires), the household/family (which includes considerations of household gender equity and family wealth), and the community (which includes the political economy and cultural forces).  ARV therapy has been placed in the middle of the framework to indicate its potential to influence the proximate determinants through improvements in health, quality of life and survival of women with HIV infection, and “ARV optimism”. The placement of ARVs in the framework also indicates that the provision of therapy is similarly influenced by broader social, economic, cultural, and gender-related determinants.  As shown in Figure 2.2, ARV therapy is hypothesized to increase fertility of women with HIV infection by reversing trends of reduced rates of coital frequency and spermatogenesis (among infected men) and by reversing trends of increased rates of amenorrhea and induced abortion. In contrast, ARV therapy is anticipated to decrease fertility by reversing trends of reduced rates of breastfeeding. The impact of ARV therapy on the remaining proximate determinants is largely unknown.  50  This framework is, of course, highly simplified since it does not show the complex interactions that may occur between the proximate determinants nor does it reveal the numerous possible feedback loops. Rather, by providing an overview of the hierarchical relationships between determinants of fertility, it is hoped that this framework can inform future research aimed at understanding how widespread use of ARV therapy may impact fertility in SSA.  2.7 CONCLUSION  The reductions in morbidity and mortality associated with ARV therapy have the potential to reverse some of the trends towards decreased fertility of women living with HIV in SSA. As ARVs become increasingly available in the region, issues related to the sexual and reproductive health and fertility of HIV-infected and – affected women must be addressed. Future research should focus on:  1. Estimating the impact of ARV therapy on sexual behaviour and contraceptive practices. 2. Assessing whether the improved health status associated with the use of ARV therapy impacts fertility intentions of HIV-infected women, men, and couples.  51  3. Estimating the impact of ARV therapy on maternal survival and fertility, including consideration of the impact on rates of MTCT. 4. Estimating the impact of ARV therapy on the ability of women to conceive and bear (more) children. 5. Understanding the gender dynamics of fertility decision-making in the context of ARV therapy, including consideration of the role of male sexual partners.  52  Table 2.1: Use of Bongaarts’ Proximate Determinants of Fertility framework to describe the potential impact of antiretroviral (ARV) therapy on fertility in sub-Saharan Africa Proximate Determinant of Fertility (P.D.)  Impact of HIV on P.D.  [Ref]  Possible Impact on Fertility  [Reference]  Impact of ARVs on P.D.  [Ref]  Reason  Hypothesized Impact of ARVs on Fertility  Improved survival and health prolongs marriage, which leads to increased reproductive opportunities  Increase fertility  1. Marriage/Stable sexual union 1.1 Age at sexual debut  Increase  [32-34]  Decrease fertility  [27,28]  Unknown  1.2 Age at marriage  Increase  [32-34]  Decrease fertility  [27,28]  Unknown  1.3 Incidence of pre- and extramarital sexual relations  Reduce  [28]  Decrease fertility  [27,28]  Unknown  1.4 Incidence of polygamy  Reduce  [28]  Increase fertility  [27,28,38]  Unknown  1.5 Incidence of divorce  Increase  [34,36,3 7]  Decrease fertility  [27,28]  Decrease  1.6 Incidence of remarriage  Reduce  [34,36]  Decrease fertility  27,28,36]  Unknown  53  Table 2.2 (continued): Use of Bongaarts’ Proximate Determinants of Fertility framework to describe the potential impact of antiretroviral (ARV) therapy on fertility in sub-Saharan Africa Proximate Determinant of Fertility (P.D.) 1.7 Incidence of widowhood  Impact of HIV on P.D.  [Ref]  Possible Impact on Fertility  [Reference]  Increase  [34,36,3 7]  Decrease fertility  [27,28]  Impact of ARVs on P.D.  [Ref]  Decrease  Reason  Hypothesized Impact of ARVs on Fertility  Improved survival and health prolongs marriage, which leads to increased reproductive opportunities  Increase fertility  Women encouraged to prolong breastfeeding at lower risk to infant  Decrease fertility  2. Onset of permanent sterility (induced or menopause) 2.1 Incidence of STIs (and associated subfertility)  Reduce  [44,45]  Increase fertility  [27,28,46]  Unknown, but no difference in the absence of STI treatment  3. Post-partum infecundability 3.1 Rate of Breastfeeding  Reduce  [34]  Increase fertility  [27,28]  Increase  3.2 Rate of postpartum abstinence  Reduce  [27,28]  Increase fertility  [3,28,51]  Unknown  [52]  Table 2.3 (continued): Use of Bongaarts’ Proximate Determinants of Fertility framework to describe the potential impact of antiretroviral (ARV) therapy on fertility in sub-Saharan Africa 54  Proximate Determinant of Fertility (P.D.)  Impact of HIV on P.D.  [Ref]  Possible Impact on Fertility  [Reference]  Impact of ARVs on P.D.  [Ref]  Reason  Hypothesized Impact of ARVs on Fertility  4. Natural fecundability or frequency of intercourse 4.1 Coital Frequency  Reduce  [4,54]  Decrease fertility  [3,28,54]  Increase  [59]  Decreased morbidity may increase coital frequency  Increase fertility  4.2 Incidence of amenorrhea  Increase  [54-56]  Decrease fertility  [3,28,55]  Decrease  [60]  Improved health may reduce incidence of amenorrhea  Increase fertility  4.3 Natural fecundability (spermatogenesis)  Decrease  [57,58]  Decrease fertility  [3,28,57,58]  Increase  [28]  Improved health may increase spermatogenesis of male partner  Increase fertility  5. Use and effectiveness of contraception 5.1 Switching contraceptive method from the pill to condoms  Increase  [28,34]  Increase fertility  [27,28]  Unknown  5.2 Condom use among previous non-contraceptive users  Increase  [28,60, 61]  Decrease fertility  [27,28]  Unknown  Table 2.4 (continued): Use of Bongaarts’ Proximate Determinants of Fertility framework to describe the potential impact of antiretroviral (ARV) therapy on fertility in sub-Saharan Africa Proximate Determinant of  Impact of  [Ref]  Possible Impact on  [Reference]  Impact of  [Ref]  Reason  Hypothesized Impact 55  Fertility (P.D.)  HIV on P.D.  Fertility  ARVs on P.D.  of ARVs on Fertility  5.3 Contraceptive use for child “insurance and replacement” desires  Reduce  [27]  Increase fertility  [27,28]  Further reduce  [66]  Better health may increase desire to insure and/or replace children  Increase fertility  5.4 Contraceptive use & abortion to avoid infant infection & orphanhood  Increase  [28,6264]  Decrease fertility  [3,27,28]  Decrease  [27,28]  Less contraceptive use due to reduced fears of infant infection and orphanhood  Increase fertility  6. Spontaneous intra-uterine mortality 6.1 Incidence of spontaneous abortion  Increase  [67]  Decrease fertility  [27,28]  Inconclusive  [20,68]  Mixed empirical findings  Inconclusive  6.2 Incidence of stillbirth  Increase  [67]  Decrease fertility  [27,28]  Inconclusive  [69]  Mixed empirical findings  Inconclusive  Increase  [54,62,7 2]  Decrease fertility  [27,28]  Decrease  [25]  Reduced concerns about MTCT and therefore reduced incidence of induced abortion  Increase fertility  7. Induced abortion 7.1 Incidence of induced abortion  Notes: Ref = Reference  56  Figure 2.1 Bongaarts’ proximate determinants of fertility framework over the reproductive lifespan (adapted with permission from Bongaarts and Potter [11])  57  Figure 2.2 Conceptual framework for the potential impact of antiretroviral (ARV) therapy on fertility in sub-Saharan Africa. .Notes:  (i) (ii) (iii)  The proximate determinates are oriented for the impact of HIV in the absence of ARV therapy The arrows indicate the potential impact of ARV therapy. STIs—sexually transmitted infections.  58  2.8 REFERENCES  1. UNAIDS. AIDS epidemic update: December 2005. Geneva, Switzerland: UNAIDS/WHO; 2005.  2. Allen S. Why is fertility an issue for HIV-infected and at-risk women? J Acquir Immune Defic Syndr 2005; 38 (Suppl 1): S1-S3.  3. Zaba B, Gregson S. Measuring the impact of HIV on fertility in Africa. AIDS 1998; 12(Suppl 1): S41-S45.  4. Lewis JJ, Ronsmans C, Ezeh A, Gregson S. The population impact of HIV on fertility in sub-Saharan Africa. AIDS 2004; 18 (Suppl 2): S35-S43.  5. Editorial. WHO 2003-08: A programme of quiet thunder takes shape. Lancet 2003; 362: 179.  6. World Health Organization (WHO). “3 by 5” progress report. Geneva, Switzerland: UNAIDS/WHO; 2004.  7. Palella FJ Jr, Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N Engl J Med 1998; 338: 853-860.  59  8. Hogg RS, Heath KV, Yip B, Craib KJ, O’Shaughnessy MV, Schechter MT, et al. Improved survival among HIV-infected individuals following initiation of antiretroviral therapy. JAMA 1998; 279: 450-454.  9. Seyler C, Anglaret X, Dakoury-Dogbo N, Messou E, Toure S, Danel C, et al. Medium-term survival, morbidity and immunovirological evolution in HIVinfected adults receiving antiretroviral therapy, Abidjan, Cote d'Ivoire. Antivir Ther 2003; 8: 385-393.  10. Wools-Kaloustian K, Kimaiyo S, Diero L, Siika A, Sidle J, Yiannoutsos CT, et al. Viability and effectiveness of large-scale HIV treatment initiatives in sub-Saharan Africa: experience from western Kenya. AIDS 2006; 20: 41-8.  11. Bongaarts J, Potter RG. Fertility, biology, and behavior: An analysis of the proximate determinants. New York: Academic Press; 1983.  12. Boerma JT, Weir SS. Integrating demographic and epidemiological approaches to research on HIV/AIDS: the proximate-determinants framework. J Infect Dis 2005; 191 (Suppl 1): S61-S67.  13. Fabiani M, Nattabi B, Ayella EO, Ogwang M, Declich S. Differences in fertility by HIV serostatus and adjusted HIV prevalence data from an antenatal clinic in northern Uganda. Trop Med Int Health 2006; 11: 182-187.  60  14. Ross A, Van der Paal L, Lubega R. HIV-1 disease progression and fertility: the incidence of recognized pregnancy and pregnancy outcome in Uganda. AIDS 2004; 18: 799-804.  15. Sedgh G, Larsen U, Spiegelman D, Msamanga G, Fawzi WW. HIV-1 disease progression and fertility in Dar es Salaam, Tanzania. J Acquir Immune Defic Syndr 2005; 39: 439-445.  16. Loko MA, Toure S, Dakoury-Dogbo N, Gabillard D, Leroy V, Anglaret X. Decreasing incidence of pregnancy by decreasing CD4 cell count in HIV-infected women in Cote d'Ivoire: a 7-year cohort study. AIDS 2005; 19: 443-445.  17. van Benthem BH, De Vincenzi I, Delmas MC, Larsen C, van den Hoek A. Pregnancies before and after HIV diagnosis in a European cohort of HIV-infected women. European Study on the Natural History of HIV Infection in Women. AIDS 2000; 14: 2171-2178.  18. Forsyth BW, Davis JA, Freudigman KA, Katz KH, Zelterman D. Pregnancy and birth rates among HIV-infected women in the United States: the confounding effects of illicit drug use. AIDS 2002; 16: 471-479.  19. De Vincenzi I, Jadand C, Couturier E, Brunet JB, Gallais H, Gastaut, JA, et al. Pregnancy and contraception in a French cohort of HIV-infected women. SEROCO Study Group. AIDS 1997; 11: 333-338.  61  20. D'Ubaldo C, Pezzotti P, Rezza G, Branca M, Hipólito G. Association between HIV-1 infection and miscarriage: a retrospective study. DIANAIDS Collaborative Study Group. AIDS 1998; 12: 1087-1093.  21. Thackway SV, Furner V, Mijch A, Cooper DA, Holland D, Martinez P, et al. Fertility and reproductive choice in women with HIV-1 infection. AIDS 1997; 11: 663–667.  22. Hankins C, Tran T, Lapointe N. Sexual behaviour and pregnancy outcome in HIV-infected women. Canadian Women’s HIV Study Group. J Acquir Immune Defic Syndr Hum Retrovirol 1998; 18: 479–487.  23. Lee LM, Wortley PM, Fleming PL, Eldred LJ, Gray RH. Duration of human immunodeficiency virus infection and likelihood of giving birth in a Medicaid population in Maryland. Am J Epidemiol 2000; 151: 1020–1028.  24. Blair JM, Hanson DL, Jones JL, Dworkin MS. Trends in pregnancy rates among women with human immunodeficiency virus. Obstet Gynecol 2004; 103: 663-668.  25. Massad LS, Springer G, Jacobson L, Watts H, Anastos K, Korn A, et al. Pregnancy rates and predictors of conception, miscarriage and abortion in US women with HIV. AIDS 2004; 18: 281-286.  26. Fapohunda ER, Todaro MP. Family structure, implicit contracts, and the demand for children in southern Nigeria. Popul Dev Rev 1998; 14: 571-594.  62  27. Gregson S, Zaba B, Hunter SC. The impact of HIV-1 on fertility in sub-Saharan Africa: Causes and consequences. Prepared for the Expert Group Meeting on Completing The Fertility Transition, New York, 11-14 March 2002. [United Nations website]; November 2005. Available at: http://www.un.org/esa/population/publications/completingfertility/RevisedZaba paper.PDF. Accessed March 20, 2006.  28. Ntozi JPM. Impact of HIV/AIDS on Fertility in sub-Saharan Africa. Fourth Meeting of the Follow-up Committee on the implementation of the DND and the ICPDPA. Younde, Cameroon 28-31 January 2002. pp 1-15. UNECA (United Nations Economic Commission for Africa) website. Available at: http://www.uneca.org/eca_resources/major_eca_websites/icpd/fourth/fc405.%20doc%20impact%20hiv%20on%20fertility.htm. Accessed April 1, 2006.  29. Stover J. Revising the proximate determinants of fertility framework: What have we learned in the past 20 years? Stud Fam Plann 1998; 29: 255-267.  30. Gage AJ, Meekers D. Sexual activity before marriage in sub-Saharan Africa. Soc Biol 1994; 41: 44-60.  31. Garenne M, Zwang J. Premarital fertility in Namibia: trends, factors and consequences. J Biosoc Sci 2006; 38: 145-67.  32. Asiimwe-Okiror G, Opio AA, Musinguzi J, Madraa E, Tembo G, Carael M. Change in sexual behaviour and decline in HIV infection among young pregnant women in urban Uganda. AIDS 1997; 11: 1757-1763. 63  33. Mukiza-Gapere J, Ntozi JP. Impact of AIDS on marriage patterns, customs and practices in Uganda. Health Transit Rev 1995; 5 (Suppl): 201-208.  34. Gregson S, Anderson RM, Ndlovu J, Zhuwau T, Chandiwana SK. Recent upturn in mortality in rural Zimbabwe: evidence for an early demographic impact of HIV-1 infection? AIDS 1997; 11: 1269-1280.  35. Gregson S, Zhuwau T, Anderson RM, Chimbadzwa T, Chiwandiwa SK. Age and religion selection biases in HIV-1 prevalence data from antenatal clinics in Manicaland, Zimbabwe. Cent Afr J Med 1995; 41: 339-346.  36. Ntozi JPM, Nakanaabi IM, Lubaale YAM. Fertility levels and trends in the face of the AIDS epidemic in Uganda. Health Transit Rev 1997; 7 (Suppl): 145-155.  37. Porter L, Hao L, Bishai D, Serwadda D, Wawer MJ, Lutalo T, et al. HIV status and union dissolution in sub-Saharan Africa: the case of Rakai, Uganda. Demography 2004; 41: 465-482.  38. Pebley A, Mbugua W. Polygyny and fertility in sub-Saharan Africa. In Reproduction and social organization in sub-Saharan Africa. Edited by Lesthaeghe RJ. California: University of California Press; 1989: 338-364.  39. Larsen U. A comparative study of the levels and differentials of fertility in Cameroon, Kenya and Sudan. In Reproduction and social organization in sub-Saharan Africa. Edited by Lesthaeghe RJ. California: University of California Press; 1989: 167-206  64  40. Arya OP, Nsanzumuhire H, Taber SR. Clinical, cultural, and demographic aspects of gonorrhoea in a rural community in Uganda. Bull World Health Organ 1973; 49: 587-595.  41. Larsen U. Sterility in sub-Saharan Africa. Popul Stud 1994; 48: 459-474.  42. Temmerman M, Plummer FA, Mirza NB, Ndinya-Achola JO, Wamola IA, Nagelkerke N, et al. Infection with HIV as a risk factor for adverse obstetrical outcome. AIDS 1990; 4: 1087-1093.  43. Leroy V, Ladner J, Nyiraziraje M, De Clercq A, Bazubagira A, Van de Perre P, et al. Effect of HIV-1 infection on pregnancy outcome in women in Kigali, Rwanda, 1992-1994. Pregnancy and HIV Study Group. AIDS 1998; 12: 643-650.  44. Mayaud P, Mosha F, Todd J, Balira R, Mgara J, West B, et al. Improved treatment services significantly reduce the prevalence of sexually transmitted diseases in rural Tanzania: results of a randomized controlled trial. AIDS 1997; 11: 18731880.  45. Wawer MJ, Sewankambo NK, Serwadda D, Quinn TC, Paxton LA, Kiwanuka N, et al. Control of sexually transmitted diseases for AIDS prevention in Uganda: a randomised community trial. Rakai Project Study Group. Lancet 1999; 353: 525535.  65  46. Garnett GP, Swinton J, Brunham RC, Anderson RM. Gonococcal infection, infertility, and population growth: II. The influence of heterogeneity in sexual behaviour. IMA J Math Appl Med Biol 1992; 9: 127-144.  47. Becker S, Rutstein S, Labbok MH. Estimation of births averted due to breastfeeding and increases in levels of contraception needed to substitute for breastfeeding. J Biosoc Sci 2003; 35: 559-574.  48. Desgrees-du-Lou A, Brou H. Resumption of sexual relations following childbirth: norms, practices, and reproductive health issues in Abidjan, Cote d’Ivoire. Reprod Health Matters 2005; 13: 155-163.  49. Centers for Disease Control and Prevention (CDC). Family planning methods and practice: Africa. 2nd ed. Atlanta, Georgia: United States Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Division of Reproductive Health; 2000.  50. Cleland JG, Ali MM, Capo-Chichi V. Post-partum sexual abstinence in West Africa: implications for AIDS-control and family planning programmes. AIDS 1999; 13: 125-131.  51. Carael M, Cleland J, Deheneffe JC, Ferry B, Ingham R. Sexual behaviour in developing countries: implications for HIV control. AIDS 1995; 9: 1171-1175.  66  52. Gaillard P, Folwer MG, Dabis F. Use of antiretroviral drugs to prevent HIV-1 transmission through breastfeeding: from animal studies to randomized clinical trials. J Acquir Defic Syndr 2004; 35: 178-187.  53. Moatti JP, Prudhomme J, Traore DC, Juillet-Amari A, Akribi HA, Msellati P, et al. Access to antiretroviral treatment and sexual behaviours of HIV-infected patients aware of their serostatus in Cote d’Ivoire. AIDS 2003; 17 (Suppl 3): S69-S77.  54. Gray RH, Wawer MJ, Serwadda D, Sewankambo N, Li C, Wabwire-Mangen F, et al. Population-based study of fertility in women with HIV-1 infection in Uganda. Lancet 1998; 351: 98-103.  55. Widy-Wirski R, Berkley S, Downing R, Okware S, Recine U, Mugerwa R, et al. Evaluation of the WHO clinical case definition for AIDS in Uganda. JAMA 1988; 260: 3286-3289.  56. Chirgwin KD, Feldman J, Muneyyirci-Delale O, Landesman S, Minkoff H. Menstrual function in human immunodeficiency virus-infected women without acquired immunodeficiency syndrome. J Acquir Immune Defic Syndr Hum Retrovirol 1996; 12: 489-494.  57. Krieger JN, Coombs RW, Collier AC, Koehler JK, Ross SO, Chaloupka K, et al. Fertility parameters in men infected with human immunodeficiency virus. J Infect Dis 1991; 164: 464-469.  67  58. Martin PM, Gresenguet G, Herve VM, et al. Decreased number of spermatozoa in HIV-1-infected individuals. AIDS 1992; 6: 130.  59. Gray G, Bastos F, Andia I, Maier M, McIntyre J, Hogg RS, et al. Determinants of HAART use among women of reproductive age in Brazil, Uganda, and South Africa. Abstract accepted to the XVI International AIDS Conference. Toronto, Canada; 2006.  60. Terceira N, Gregson S, Zaba B, Mason PR. The contribution of HIV to fertility decline in rural Zimbabwe, 1985-2000. Popul Stud 2003; 57: 149-164.  61. Lutalo T, Kidugavu M, Wawer MJ, Serwadda D, Zabin LS, Gray RH. Trends and determinants of contraceptive use in Rakai District, Uganda, 1995-98. Stud Fam Plann 2000; 31: 217-227.  62. Okeibunor JC. Threat of AIDS and condom use in a Nigerian urban community: implications for fertility regulation in Nigeria. A study report for Union for African Population Studies (UAPS). Dakar, Senegal; 1999.  63. Glynn JR, Buve A, Carael M, Kahindo M, Macauley IB, Musonda RM, et al. Decreased fertility among HIV-1-infected women attending antenatal clinics in three African cities. J Acquir Immune Defic Syndr 2000; 25: 345-352.  68  64. Rutenberg N, Biddlecom AE, Kaona AD. Reproductive decision-making in the context of HIV and AIDS: a qualitative study in Ndola, Zambia. Int Fam Plan Perspect 2000; 26: 124-130.  65. Baylies C. The impact of HIV on family size preference in Zambia. Reprod Health Matters 2000; 8: 77-86.  66. Panozzo L, Battegay M, Friedl A, Vernazza PL. Swiss Cohort Study: High risk behaviour and fertility desires among heterosexual HIV-positive patients with a serodiscordant partner – two challenging issues. Swiss Med Wkly 2003; 133: 124127.  67. Brocklehurst P, French R. The association between maternal HIV infection and perinatal outcome: a systematic review of the literature and meta-analysis. Br J Obstet Gynaecol 1998; 105: 836-848.  68. Suy A, Martinez E, Coll O, Lonca M, Palacio M, de Lazzari E, et al. Increased risk of pre-eclampsia and fetal death in HIV-infected pregnant women receiving highly active antiretroviral therapy. AIDS 2006; 20: 59-66.  69. Tuomala RE, Shapiro DE, Mofenson LM, Bryson Y, Culnane M, Hughes MD, et al. Antiretroviral therapy during pregnancy and the risk of an adverse outcome. N Engl J Med 2002; 346: 1863-1870.  69  70. Tuomala RE, Watts DH, Li D, Vajaranant M, Pitt J, Hammill H, et al. Improved obstetric outcomes and few maternal toxicities are associated with antiretroviral therapy, including highly active antiretroviral therapy during pregnancy. J Acquir Immune defic Syndr 2005; 38: 449-473.  71. Thorne C, Newell ML. The safety of antiretroviral drugs in pregnancy. Expert Opin Drug Saf 2005; 4: 323-335.  72. Desgrees-du-Lou A, Msellati P, Ramon R, Noba V, Viho I, Dabis F, et al. HIV-1 infection and reproductive history: a retrospective study among pregnant women, Abidjan, Cote d'Ivoire, 1995-1996. Ditrame Project. Int J STD AIDS 1998; 9: 452-456.  70  CHAPTER 32: THE RELATIONSHIP BETWEEN HAART USE AND SEXUAL ACTIVITY AMONG HIV-POSITIVE WOMEN OF REPRODUCTIVE AGE IN BRAZIL, SOUTH AFRICA, AND UGANDA  3.1 INTRODUCTION  The vast majority of the world’s 14 million HIV-infected women live in developing and transitional countries where sexual contact is the primary mode of HIV transmission [1]. A growing body of evidence from such countries reveals that antiretroviral therapy has dramatically improved the survival and quality of life for HIV-infected individuals [2, 3, 4, 5]. Recent global efforts to improve access to highly active antiretroviral therapy (HAART) and clinical care should serve to deliver these positive outcomes to more HIV-infected women in high prevalence and lowresource settings [6].  In addition to the anticipated clinical effects, HAART use may influence the sexual behaviour of HIV-infected women. Earlier research demonstrated that in the absence of treatment, HIV-infected women were less sexually active than HIV2  A version of this chapter has been published. Kaida A, Gray G, Bastos FI, Andia I, Maier M, McIntyre J, Grinsztejn B, Strathdee SA, Bangsberg DR, Hogg R. The relationship between HAART use and sexual activity among HIV-positive women of reproductive age in Brazil, South Africa, and Uganda. AIDS Care 2008;20: 21-25.  71  negative women owing, in part, to higher morbidity [7, 8]. The health improvements associated with the use of HAART are anticipated to increase sexual activity due to improved health status [9] and perceptions of reduced infectivity [10], however, related empirical evidence is lacking.  Given the close relationship between sexual transmission of HIV and the incidence of pregnancy, information about the sexual behaviours of HIV-positive women on HAART constitute a critical component of comprehensive initiatives aimed at improving the quality of life of HIV-infected women, including sexual and reproductive decision-making.  The purpose of this study was to determine whether current HAART use is associated with recent sexual intercourse among HIV-infected women of reproductive age (18-49 years) from Brazil, South Africa, and Uganda. A secondary purpose was to estimate differences in protected sex and contraceptive use among sexually active HAART users and non-users.  72  3.2 METHODS  3.2.1 Study design  This analysis is based on data collected from a cross-sectional survey conducted in Brazil, South Africa, and Uganda. The primary outcome was recent sexual activity defined as vaginal intercourse in the previous month. The secondary outcomes were protected vaginal intercourse (defined as “always used a condom”) and contraceptive use in the previous six months. The key explanatory variable was current HAART use. The study was conceived as a pilot project to determine the feasibility of conducting more comprehensive sexual and reproductive health studies among a larger sample of HIV-positive women in Brazil, South Africa, and Uganda.  3.2.2 Study setting and sample population  A total of 179 HIV-positive women in patient care were randomly selected for participation through the Mbarara Hospital HIV Clinic in Mbarara, Uganda (n=85); the Perinatal HIV Research Unit in Soweto, South Africa (n=50); and the IPECFiocruz cohort in Rio de Janeiro, Brazil (n=44) [11].  73  Eligible women had to have documented HIV status, be of reproductive age (defined as 18-49 years), and be seeking regular care at one of the three study centres.  3.2.3 Data collection  A consecutive sample of HIV-positive women accessing services at each study site was selected to participate. A study representative explained the study purpose, procedures, risks, and benefits as part of the informed consent process. Participation rates exceeded 90% in each of the study sites.  Upon identification and consent, study participants responded to an interviewer-administered questionnaire in either English or the local language. Measures included: socio-demographics; HIV/AIDS history, diagnosis, and treatment; self-reported health; sexual and contraceptive behaviour; fertility desires; and HAART optimism (a 13-item scale yielding a range of scores from 13-52, with higher scores indicating greater optimism) [12].  The interviewers at each study site were women from the local community. They were trained in survey research, including instruction on how to conduct an  74  interview and how to obtain informed consent. Data collection took place between March and June 2005.  3.2.4 Data analysis  We compared the prevalence of recent sexual intercourse across groups (i.e., sexually active non-HAART users vs. sexually active HAART users). Unadjusted odds ratios (OR) and 95% confidence intervals (95% CI) were computed to determine the magnitude of the association between independent variables and recent sexual intercourse.  Multivariate logistic regression was used to model the association between HAART use and the likelihood of recent sexual intercourse, while controlling for covariates. All independent variables were tested for colinearity. It was suspected a priori that the variables ‘marital Status’ and ‘HIV-infected partner/spouse’ would be highly associated and this was empirically confirmed. Therefore, only ‘marital status’ was considered for entry into the multivariate model. To yield adjusted odds ratios (AOR) and 95% CIs, a stepwise procedure was used to enter HAART use and all variables significant in the univariate analysis into a logistic regression model [13]. The variables ‘country’ and ‘age’ were forced into the model to obtain adjusted estimates of effect. 75  Prevalence of Contraceptive Use and Protected Sexual Activity  Among women reporting recent sexual intercourse, we compared the proportion that report using condoms (i.e., “protected sex”) and contraceptives between HAART users and non-users. Crude odds ratios and 95% CIs were computed to determine the magnitude of the association between HAART use and protected sex and contraceptive use. Given that this was a secondary objective of the analysis and owing to the small sample size, multivariate analysis was not conducted for these two outcome variables.  All statistical tests are two-sided and considered significant at α = 0.05. Statistical analyses were done using SAS for Windows (version 9.1) [14].  3.2.5 Ethical considerations  Ethics approval for this study was obtained from the Human Research Ethics Committee (Medical) of the University of the Witwatersrand in Johannesburg, South Africa, the Faculty of Medicine Research and Ethics Committee and the Institutional Ethics Review Board of Mbarara University, and the local and national Institutional Review Boards (IRBs) in Brazil. Study investigators only had access to de-identified data.  76  3.3 RESULTS  Overall, 46% of the participants reported sexual intercourse in the previous month (83/179). By study site, 33% of women in Uganda, 55% of women in Brazil, and 62% of women in South Africa reported recent sexual intercourse.  3.3.1 Baseline characteristics  Distributions of baseline covariates are presented in Table 3.1. Nearly twothirds (65%) of women reported currently using HAART (56% in Uganda, 64% in Brazil, and 80% in South Africa). The mean age of women was 34.5 years [SD=6.9]. Only 22% were employed full-time and the mean years of formal schooling was 7.2 [SD=4.0]. Thirty-four percent (34%) were currently married and nearly one-third (32%) reported knowing that their primary partner/spouse was HIV-infected. Sixtyeight percent (68%) reported having been diagnosed with AIDS, however, 66% reported their current health status as “Excellent, very good, or good”, and the mean HAART optimism score was 30 [SD=6.7]. The majority of women did not want any (more) children (65%).  77  3.3.2 Univariate analyses  Women on HAART were just as likely as women not on HAART to report recent sexual intercourse (OR: 0.93; 95% CI: 0.50-1.71) (Table 3.2). Women who reported recent sexual intercourse were, however, more likely to have higher HAART optimism scores, have more education, be currently married, have an HIVpositive primary partner/spouse, desire more children, and not have been diagnosed with AIDS.  3.3.3 Adjusted analyses  After adjusting for potential confounders, there remained no association between HAART use and recent sexual activity (adjusted odds ratio (AOR): 0.76; 95% CI: 0.34, 1.72; see Table 3.2). Being currently married, desiring more children, and having a higher HAART optimism score remained most strongly associated with recent sexual intercourse.  78  3.3.4 Contraceptive use and protected sexual activity  Of the 83 women reporting recent sexual intercourse, 76% were using contraceptive methods and 63% were engaging in protected sex (i.e., using condoms). Women using HAART were significantly more likely to practice protected sex (crude OR: 3.64; 95% CI: 1.41-9.38) and non-significantly more likely to use contraceptive methods (crude OR: 2.15; 95% CI: 0.77-5.99) than non-users.  3.4 DISCUSSION  We found that, among HIV-infected women in Brazil, South Africa, and Uganda, HAART users and non-users were equally likely to report recent sexual intercourse. This finding is consistent with other reports of HIV-infected men and women in Uganda [15, 16] and Côte d’Ivoire [17].  In sub-analyses, we investigated the association between HAART use and recent sexual activity for each country. While the effect of socio-demographic covariates varied, none of the country-specific models showed a significant relationship between HAART use and recent sexual activity.  Recent sexual activity among HIV-positive women in our study appears to be most strongly influenced by factors unrelated to HAART use, including being  79  currently married and wanting more children. This finding was expected based on the literature on predictors of sexual activity and coital frequency among women in general [18, 19]. Interestingly, HAART optimism was also associated with recent sexual activity suggesting that being optimistic about the benefits of HAART is a better predictor of recent sexual intercourse than HAART use itself. It is difficult to compare these findings since the vast majority of literature on the impact of HAART optimism comes from MSM populations in resource-rich areas [20, 21].  Of note, while HAART use wasn’t associated with recent sexual intercourse overall, it was positively associated with other sexual behaviours, including protected sex and contraceptive use. Recently, others have similarly reported that HAART users in developing countries are more likely to practice protected sex [15, 16, 17], however, there is limited comparative literature on the impact of HAART on contraceptive use.  The relatively small sample size and the diversity of the countries involved may constitute important limitations to this study. These features of the study design are a consequence of conducting a pilot study. Presently, however, larger, separate studies are being conducted in each of the three study settings to further verify the results presented here.  80  In conclusion, this study supports an absence of association between HAART use and recent sexual intercourse. Sexually active HAART users may, however, be more likely to practice protected sex and use contraceptives. This information is important for the development of comprehensive sexual and reproductive health services for HIV-infected women.  81  Table 3.1: Baseline characteristics of study population: HIV-infected women of reproductive age in Brazil, South Africa, and Uganda (n=179) Characteristic Country South Africa Uganda Brazil  Frequency (n)  Percent (%)  50 85 44  28% 48% 25%  Socio-demographic variables Mean Age [SD] Mean Years of Schooling [SD] Employed Fulltime Yes No Marital Status Currently married Not currently married  34.5 [6.9] 7.2 [4.0] 39 139  22% 78%  61 118  34% 66%  62 117  35% 65%  116 61  66% 35%  57 122  32% 68%  122 57  68% 32%  116 63  65% 35%  Fertility desire & Health Status variables Want more children? Yes No Self-reported health status Excellent, Very good, or Good Fair or Poor HIV-related variables HIV-infected primary partner/spouse Yes No or unknown AIDS diagnosis Yes No or unknown Current HAART Use Yes No Mean HAART optimism score [SD]  30 [6.7]  82  Table 3.2: Unadjusted and adjusted Odds Ratios and 95% Confidence Intervals (95% CI) of variables associated with recent sexual intercourse (in the previous month) among HIV-infected women in Brazil, South Africa, and Uganda (n=179) Characteristic  Country Brazil Uganda South Africa Age (Mean [SD]) Years of Schooling (Mean [SD]) Employed Fulltime Yes No Currently Married Yes No Desires more children? Yes No Self-reported health status Excellent, very good, or good Fair or Poor HIV-infected spouse/partner Yes No or Unknown AIDS diagnosis Yes No or Unknown HAART Use Yes No HAART optimism score (Mean [SD])  Sexual intercourse in the past month Yes (n=83) No (n=96) % % 29% 37% 34% 33.8 [6.4] 8.2 [3.5]  21% 20% 59% 35.1 [7.2] 6.4 [4.3]  27% 73%  18% 82%  49% 51% 51% 49%  Adjusted OR* [95% CI]  Reference 1.36 [0.60-3.10] 0.41 [0.19-0.86] 0.97 [0.93-1.02] 1.12 [1.04-1.21]  --  1.70 [0.83-3.49]  --  3.71 [1.93-7.14]  6.68 [2.79-16.0]  3.89 [2.02-7.49]  4.09 [1.71-9.81]  1.49 [0.80-2.80]  --  4.32 [2.19-8.52]  --  0.37 [0.19-0.70]  --  0.93 [0.50-1.71]  0.76 [0.34-1.72]  1.09 [1.03-1.14]  1.11 [1.04-1.17]  ---  21% 79% 21% 79%  70%  61%  30%  39%  48% 52%  18% 82%  57% 43%  78% 22%  64% 36% 31.3 [6.1]  Unadjusted OR [95% CI]  66% 34% 27.8 [6.8]  Notes: * Adjusted odds ratios obtained from a stepwise procedure to choose a logistic regression model with forced entry of the variables ‘Age’ and ‘Country’.  83  3.5 REFERENCES  1. UNAIDS. AIDS epidemic update: December 2005. Geneva, Switzerland: UNAIDS/WHO; 2006.  2. Marins JR., Jamal LF, Chen SY, Barros MB, Hudes ES, Barbosa AA, et al. Dramatic improvement in survival among adult Brazilian AIDS patients. AIDS 2003; 17: 1675-1682  3. Seyler C, Anglaret X, Dakoury-Dogbo N, Messou E, Toure S, Danel C, et al. Medium-term survival, morbidity and immunovirological evolution in HIVinfected adults receiving antiretroviral therapy, Abidjan, Cote d'Ivoire. Antivir Ther 2003; 8: 385-393.  4. Spacek LA, Shihab HM, Kamya MR, Mwesigire D, Ronald A, Mayanja H, et al. Response to antiretroviral therapy in HIV-infected patients attending a public, urban clinic in Kampala, Uganda. Clin Infect Dis 2006; 42: 252-259.  5. Wools-Kaloustian K, Kimaiyo S, Diero L, Siika A, Sidle J, Yiannoutsos CT, et al. Viability and effectiveness of large-scale HIV treatment initiatives in subSaharan Africa: Experience from western Kenya. AIDS 2006; 20: 41-48.  6. World Health Organization (WHO). “3 by 5” progress report. Geneva, Switzerland: UNAIDS/WHO; 2004.  84  7. Gray RH, Wawer MJ, Serwadda D, Sewankambo N, Li C, Wabwire-Mangen F, et al. Population-based study of fertility in women with HIV-1 infection in Uganda. Lancet 1998; 351: 98-103.  8. Wilson TE, Massad LS, Riester KA, Barkan S, Richardson J, Young M, et al. Sexual, contraceptive, and drug use behaviors of women with HIV and those at high risk for infection: Results from the women's interagency HIV study. AIDS 1999; 13: 591-598.  9. Kaida A, Andia I, Maier M, Strathdee SA, Bangsberg DR, Spiegel J, et al. The potential impact of antiretroviral therapy on fertility in sub-Saharan Africa. Curr HIV/AIDS Rep 2006; 3: 177-184.  10. Kalichman SC, Rompa D, Austin J, Luke W, DiFonzo K. Viral load, perceived infectivity, and unprotected intercourse. J Acquir Immune Deficiency Syndr (1999) 2001; 28: 303-305.  11. Grinsztejn B, Bastos FI, Veloso VG, Friedman RK, Pilotto JH, Schechter M, et al. Assessing sexually transmitted infections in a cohort of women living with HIV/AIDS, in Rio de Janeiro, Brazil. Int J STD AIDS 2006; 17: 473-478.  12. Van de Ven P, Crawford J, Kippax S, Knox S, Prestage G. A scale of optimismscepticism in the context of HIV treatments. AIDS Care 2000; 12: 171-176.  85  13. Hosmer DW, Lemeshow S. Applied logistic regression. New York, NY: Wiley; 2000.  14. The SAS Institute. SAS Version 9.1. Cary, NC: The SAS Institute; 2005.  15. Bateganya M, Colfax G, Shafer LA, Kityo C, Mugyenyi P, Serwadda D, et al. Antiretroviral therapy and sexual behavior: A comparative study between antiretroviral- naive and -experienced patients at an urban HIV/AIDS care and research center in Kampala, Uganda. AIDS Patient Care STDS 2005; 19: 760-768.  16. Bunnell R, Ekwaru JP, Solberg P, Wamai N, Bikaako-Kajura W, Were W, et al. Changes in sexual behavior and risk of HIV transmission after antiretroviral therapy and prevention interventions in rural Uganda. AIDS 2006; 20: 85-92.  17. Moatti JP, Prudhomme J, Traore DC, Juillet-Amari A, Akribi HA, Msellati P, et al. Access to antiretroviral treatment and sexual behaviours of HIV-infected patients aware of their serostatus in Cote d'Ivoire. AIDS 2003; (Suppl 3)17: S69-77.  18. Bongaarts J, Potter RG. Fertility, biology, and behaviour: An analysis of the proximate determinants. New York, NY: Academic Press; 1983.  19. Rao KV, Demaris A. Coital frequency among married and cohabiting couples in the United States. J Biosoc Sci 1995; 27: 135-150.  86  20. Abelson J, Rawstorne P, Crawford J, Mao L, Prestage G, Kippax S. HIV optimism does not explain increases in high-risk sexual behaviour among gay men of positive or negative HIV status in Sydney, Australia. AIDS 2006; 20: 1215-1216.  21. Sullivan PS, Drake AJ, Sanchez TH. Prevalence of treatment optimism-related risk behavior and associated factors among men who have sex with men in 11 states, 2000-2001. AIDS Behav e-Pub ahead of print; 2006.  87  CHAPTER 43: FERTILITY INTENTIONS OF HIV-POSITIVE WOMEN OF REPRODUCTIVE AGE IN SOWETO, SOUTH AFRICA: THE INFLUENCE OF EXPANDING ACCESS TO HAART IN AN HIV HYPER-ENDEMIC SETTING  4.1 INTRODUCTION  In sub-Saharan Africa (SSA), women of childbearing age comprise 61% of people living with HIV, accounting for over 12 million women [1]. In many regions, HIV incidence is increasing most dramatically among young women aged 18-30 years [1, 2], which coincides with their peak reproductive years [3]. Globally, a plethora of evidence indicates that many women living with HIV continue to desire children [4-8], become pregnant [5, 6, 9], and give birth [5, 6, 10] after knowing their HIV-positive status.  Irrespective of HIV seropositivity, fertility decision-making can be complex [11]; however, reproduction among HIV-infected women introduces additional 3  A version of this chapter has been accepted for publication. Kaida A, Laher F, Strathdee SA, Janssen PA, Money D, Hogg RS, Gray G. Fertility intentions of HIV-positive women of reproductive age in Soweto, South Africa: The influence of expanding access to HAART in an HIV hyper-endemic setting. American Journal of Public Health (June 2010).  88  personal, public health, and clinical care issues [12]. The vast majority of conceptions occur without the use of reproductive technologies such as sperm washing and artificial insemination [13]. Thus, the unprotected sexual activity required for conception carries a risk of HIV transmission to uninfected sexual partners [14]. Reproduction among HIV-positive women also carries a risk of vertical transmission during pregnancy, labour, and through breastfeeding [15, 16]. Moreover, HIVpositive women have a lower life expectancy than negative women [17], increasing the risk of maternal orphanhood [18]. In light of these concerns, early reproductive guidelines for people living with HIV were dissuasive [19] and HIV-positive women who express fertility desires continue to encounter community and healthcare worker disapproval [4, 20].  Nonetheless, while the potential health risks may have dampened the fertility intentions of some HIV-positive women, stigma associated with childlessness in many societies [21] and the strong personal desires for biological parenthood [4] remain potent drivers of fertility intentions, despite an HIV-positive status. Indeed, in some cultural contexts, remaining childless can be a societal norm violation more stigmatizing than the HIV infection itself [4, 22].  Expanding access to highly active antiretroviral therapy (HAART) is changing the landscape of fertility decision-making for people living with HIV [23].  89  HAART increases life expectancy [24-26], decreases morbidity [25, 27], and dramatically reduces the risks of vertical [28] and horizontal transmission [29, 30]. In this era of expanding HAART access, the significant reductions in health risks and barriers to reproduction among people living with HIV has corresponded with increased calls for a rights- and evidenced-based approach to reproduction [31, 32]. Since fertility intentions are the strongest predictor of eventual fertility [33], creating effective and responsive sexual and reproductive health services for HIV-positive women in the context of expanding HAART access requires a clear understanding of expressed fertility intentions.  Existing evidence concerning the influence of expanding HAART access on fertility intentions is largely incomplete. While recent regional studies have shown that HAART use is associated with higher fertility intentions, these studies have neglected to consider the duration of HAART use [6, 8] and tended only to compare fertility intentions of positive women, without conducting a comparison with HIVnegative women from the same community [6-8]. Moreover, the lack of an HIVnegative comparison group precludes the opportunity to assess whether HAART users begin to resemble HIV-negative women in their fertility intentions, particularly as HIV is increasingly recognized as a manageable chronic disease.  90  Given the high HIV prevalence among women of reproductive age in Soweto, South Africa [1], the objective of this study was to assess the prevalence of fertility intentions and to determine whether fertility intentions varied according to HIV status and HAART use among women. We hypothesized that HIV-positive women would have lower fertility intentions compared with HIV-negative women. In addition, we hypothesized that HIV-positive women receiving HAART would have higher fertility intentions than HIV-positive women who were HAART-naive, with increasing duration of HAART treatment associated with incrementally higher fertility intentions. Overall, we hypothesized that HAART use would narrow the measurable differences in fertility intentions between HIV-positive and HIVnegative women [23].  4.2 METHODS  4.2.1 Study design  This analysis is based on cross-sectional survey data of HIV-positive (HAART receiving and HAART-naïve) and HIV-negative women seeking services at the Perinatal HIV Research Unit (PHRU) in Soweto, South Africa. A medical chart  91  review was also conducted to confirm HIV serology and HAART use history of HIV-positive women.  4.2.2 Study setting  The PHRU, one of Africa’s largest HIV research and clinical service centres, is housed within the Chris Hani Baragwanath Hospital in Soweto, an urban South African township located 15km outside of Johannesburg. The PHRU clinic sees over 5,000 patient visits monthly and provides free antiretroviral therapy and clinical care to medically eligible HIV-positive individuals and on-going wellness care for those not medically eligible for antiretroviral treatment. The PHRU also operates a Prevention Studies area that includes a Voluntary Counseling and Testing (VCT) centre.  4.2.3 Eligibility criteria  To be eligible to participate in the study, women were required to be 18-49 years of age, attending a PHRU clinic, residing in Soweto, competent to give informed consent, and willing to allow medical record review for the purposes of confirming HIV status and HAART history. We considered women to be HAART 92  users if they had been taking HAART medications for at least one month. We considered women to be non-HAART users if they had never taken HAART, except for vertical transmission prophylaxis.  4.2.4 Study sample  We enrolled 751 women into the study, including 253 HIV-positive women receiving HAART, 249 HIV-positive but HAART-naïve women, and 249 HIVnegative women. This sampling strategy provided one case group (HAART users) and two comparison groups (HAART-naïve and HIV-negative women).  HAART users were sampled from the PHRU’s PEPFAR Clinic which has provided free antiretroviral therapy to medically eligible patients since July 2004. Currently, the PEPFAR clinic has over 1,000 patients receiving HAART, 75% of whom are female. PEPFAR patients are followed-up every three months and generally receive one of two standard HAART regimens: Regimen 1 is d4T/3TC/EFV or NVP and Regimen 2 is Kaletra/ddI/AZT [34].  HIV-positive, HAART-naïve women were sampled from the PHRU’s Wellness Clinic, initiated in January 2003 with the goal of providing preventive care to HIV-positive individuals. Wellness patients are followed-up approximately every  93  six months. When patients are medically eligible for HAART, they are referred to the PEPFAR clinic or to one of the nearby government ART clinics. There are approximately 3,000 active patients in the Wellness Clinic.  HIV-negative women were sampled from the VCT clinic, which was initiated in mid-2002 and sees approximately 400 people per month. Testing is conducted onsite during visits that last an average of two hours. Approximately 65% of attendees are women and approximately 30% of all attendees are HIV-positive.  For this analysis on fertility intentions, we restricted the study sample to women aged 18-44 years of age who were not sterilized (i.e., did not report hysterectomy or female sterilization). This yielded an analytic sample of 674 women including 217 HAART users, 215 HAART-naïve women, and 242 HIV-negative women.  4.2.5 Data collection  Every female patient attending the PEPFAR Clinic and the VCT clinic was consecutively approached by a research assistant to assess eligibility and interest in participating in the study. Since many more women attend the Wellness Clinic, a list was made of chart numbers of women attending the clinic each day. A random sample of chart numbers (40% of the total number of charts present) was then drawn 94  and the corresponding women were approached to assess eligibility and interest in participating in the study.  After confirming eligibility and seeking informed consent, all participants were asked to complete a 15-25 minute interviewer-administered questionnaire in English. The study interviewers were multilingual and trained to ensure accurate and consistent translation of the questionnaire if required or requested by the participant. Pilot testing of 45 women revealed that women were able to understand and answer the questionnaire. Approximately 12 women were interviewed daily by three trained research assistants between May and December, 2007. Participants were reimbursed 20 Rand for their transportation costs to and from the PHRU (~ $3.00 USD). Research assistants were women from the local community who had previous research experience and were recent Social Sciences’ graduates of a local university.  4.2.6 Data collection instruments  The questionnaire assessed socio-demographic characteristics (e.g., age, education, employment status, marital status, parity); HIV serostatus and date of HIV-positive diagnosis; clinical stage of disease (CD4 cell count and viral load); HAART history; fertility intentions; fertility history (number and timing of 95  pregnancies, abortions, stillbirths, miscarriages, and live births); contraceptive practices; and sexual history.  For all HIV-positive women, medical records were reviewed to confirm HIV status, HAART history, and to obtain clinical data including WHO stage of disease [35] and CD4 cell count. Viral load assessments were only conducted on women from the PEPFAR clinic (i.e., HAART users). The medical record was considered the referent measure for inconsistencies between self-reported and medical record data.  4.2.7 Measures  The primary outcome was self-reported fertility intentions, which was defined by answers to the question: “Are you planning to have (any more) children in the future?”. Women were free to respond “Yes”, “No”, or “Don’t know”. Given recent findings suggesting that the undecided tend to more closely resemble the expressed fertility intentions of the majority of those with stated desires [36], the small proportion of women who responded “Don’t Know” (5%) were included in the “No” category. There was little difference in the proportion of women reporting “Don’t know” by HIV and HAART use status.  96  The primary explanatory variable was current HAART use. The secondary explanatory variable was HIV status. Variables known to be associated with fertility intentions were included in the analysis to provide an adjusted estimate of the association. Covariates included age, education, employment, monthly household income, current sexual partnership status, number of living children, and HIV clinical variables including most recent CD4, nadir CD4, and WHO Stage of Disease.  4.2.8 Statistical analysis  We computed the prevalence of fertility intentions among each of the three groups of women in our study. We conducted two separate models to measure the presence and strength of the association between HAART use and the likelihood of reporting fertility intentions, while controlling for covariates. The first model compared HAART users and non-HAART users to HIV-negative women. The second model compared HAART users to non-HAART users and allowed adjustment for HIV-associated clinical characteristics.  In both models, univariate analyses were used to assess the relationship between fertility intentions and HAART use and covariates. Differences in fertility intentions between groups are reported using Pearson’s chi-squared test (for categorical covariates), ANOVA, or Student’s independent t-test (for continuous 97  variables). The association between fertility intentions and HAART use is reported using a crude odds ratio with a 95% confidence interval (95% CI). After testing for colinearity (using Spearman’s rho (ρ)) [37] and interaction [38], all covariates with significant associations (p<0.10) in the univariate analysis were included in the final multivariate logistic regression model to obtain adjusted odds ratios and 95% confidence intervals. All statistical tests were two-sided and were considered significant at α = 0.05. The data were analyzed using SAS version 9.1 [39].  4.2.9 Sub-analyses  We conducted the same analyses described above but restricted our sample to women aged 18-30 years, the peak childbearing years among women in South Africa [3].  4.3 RESULTS  4.3.1 Study sample  A total of 801 women were approached for participation, of whom 751 consented, completed the questionnaire and underwent a medical record review 98  (response rate = 94%). The analysis of fertility intentions was restricted to women aged 18-44 years who were not sterilized, yielding a study sample of 674 women.  4.3.2 Baseline characteristics  There were important differences in baseline covariates by HIV and HAART use status (Table 4.1). Mean age was 30 years [SD=6.7], with HIV-negative women being significantly younger than HIV-positive women. Overall, 51% of women had less than a grade 12 education, 61% were unemployed, and 71% had a monthly household income less than 3,000 ZAR ($ 350 USD). A small proportion of women were married (8%) but the vast majority were currently in a sexual relationship (78%), with a mean of 0.90 [SD=0.68] sexual partners in the previous six months. One quarter reported that their primary sexual partner was HIV-positive, one-third reported that he was HIV-negative, and the remaining 42% did not know the HIV status of their sexual partners. The mean number of lifetime sexual partners was 5.3 [SD=5.6]. Overall, mean parity was 1.4 [SD=1.1]. Of women with at least one livebirth, 14% had lost a child. Overall, 41% of women had two or more living children.  Among HIV-positive women, mean time since HIV diagnosis was 59 months [SD=36]; 58% of HAART users had recent CD4 counts ≥ 350 cells/uL compared with 99  45% of HAART-naïve women. Overall, 15% had nadir CD4 counts < 50 cells/uL. Nearly all were in WHO Stage of Disease I or II (97%) and 94% had disclosed their HIV status to someone.  Among HAART users, median duration of HAART use of 31 months [IQR: 28, 33], ranging from 1 month to 89 months. Most (81%) had undetectable viral loads (< 50 copies/ml).  4.3.3 Prevalence of fertility intentions  Overall, 44% of women reported that they intended to have (more) children. This varied significantly by HIV status with 31% of HAART users, 29% of HAARTnaïve women, and 68% of HIV-negative women reporting fertility intentions (p<0.0001).  4.3.4 Factors associated with fertility intentions in the overall sample  In the unadjusted analyses, many of the measured covariates were significantly associated with fertility intentions (Table 4.2). Compared with HIVnegative women, HIV-positive women were significantly less likely to report  100  fertility intentions (HAART users OR: 0.21; 95% CI: 0.14, 0.32; HAART-naïve women OR: 0.20; 95% CI: 0.13, 0.29).  After adjusting for covariates shown in Table 4.1, HIV status remained significantly associated with fertility intentions. Compared with HIV-negative women, HIV-positive women were significantly less likely to report fertility intentions (HAART users AOR: 0.40; 95% CI: 0.23, 0.69; HAART-naive women AOR: 0.35; 95% CI: 0.21, 0.60). Currently being in a sexual relationship and having fewer living children also remained independently associated with fertility intentions.  4.3.5 Factors associated with fertility intentions among HIV-positive women  As seen in Table 4.3, in an analysis restricted to HIV-positive women, HAART users and non-users were equally likely to report fertility intentions (OR: 1.08; 95% CI: 0.71, 1.63). There were no significant differences in fertility intentions by any of the measured clinical characteristics including duration of HIV diagnosis, recent CD4, nadir CD4, WHO Stage of Disease, or disclosure.  After adjusting for covariates shown in Table 4.1, HAART users and nonusers remained equally likely to report fertility intentions (AOR: 1.16; 95% CI: 0.72,  101  1.86). Currently being in a sexual relationship and having fewer living children remained independently associated with fertility intentions.  As shown in Figure 4.1, there was no association between length of time on HAART and reported fertility intentions.  4.3.6 Sub-analyses  Although fertility intentions among young women (18-30 years) are higher than among the total sample, they still varied significantly by HIV status: 38% of HAART users, 34% of HAART-naïve women, and 75% of HIV-negative women aged 18-30 years expressed fertility intentions (p<0.0001). In multivariate analyses we found that the same variables that were associated with fertility intentions in the overall sample were similarly associated with fertility intentions among young women in both direction and magnitude (data not shown). This was the case for both models (i.e., all women and only HIV-positive women).  102  4.4 DISCUSSION  In contrast with our primary hypothesis, we found that fertility intentions of HIV-positive women did not differ by use or duration of HAART. Consistent with our secondary hypothesis, however, HIV-positive women were significantly less likely to report fertility intentions compared with HIV-negative women.  Our findings of nearly one-third of HIV-positive women reporting fertility intentions largely irrespective of HAART use, contrast with findings from other subSaharan African sites, which report threefold higher fertility intentions among HAART users [6] and higher fertility intentions associated with increasing duration of HAART use [7]. A recent American study showed that HAART use was associated with a lower prevalence of fertility desires [40]. Our findings are, however, consistent with those of a Canadian study which reported no association between HAART use and fertility intentions [41].  Potential reasons for our findings may relate to the nature of HAART and HIV care services at the PHRU. Namely, HAART has been available at the PHRU since July 2004, nearly three years longer than at government clinics in South Africa [42] and at least two years longer than most sub-Saharan African settings [43]. The lack of association between HAART use and fertility intentions in our study may reflect the fact that HIV-positive women who are not yet receiving HAART can be 103  confident that treatment is available once they are medically eligible, thereby minimizing differences between groups. Moreover, HAART-naïve women were sampled from the Wellness clinic, which provides regular clinical care to HIVpositive women. Thus, any influence that regular contact with health care providers has on fertility decision-making is likely to be similar in both groups. In this way, the PHRU setting may be more comparable to Canadian study findings [41], where HAART has been available at no charge to medically eligible HIV-infected individuals since 1996 [26, 41]. Consistent with other studies of HIV-positive and negative women around the world, number of living children and current partnership status were also strongly associated with fertility intentions.  While HAART use did not affect fertility intentions among HIV-positive women, we did find that HIV-positive women were 60% less likely to report an intention to have (more) children compared with HIV-negative women. It is difficult to compare these findings to other settings since we could identify no studies that directly measured differences in fertility intentions between HIV-negative and positive women. Moreover, while the expressed fertility intentions of HIV-positive women were lower than for HIV-negative women, they remain substantial. In our study, nearly one-third of HIV-positive women reported fertility intentions, a proportion that increased significantly by younger age and fewer living children.  104  Our findings are at the upper range of fertility intentions for HIV-positive women reported from other settings [6-8, 41, 44].  All women, including women living with HIV, should be supported to achieve their reproductive goals in the healthiest and safest possible manner [32]. Given the noted prevalence of fertility intentions among HIV-positive women, it is critical that factual and non-stigmatizing information and support be incorporated into HIV treatment services to optimize positive outcomes for mother, father, and baby. This includes counseling services regarding HAART and pregnancy [45], safer options to conceive (including HAART as prevention [46]), safer labour options, PMTCT services, antenatal and postnatal care, and infant feeding options. Currently, no clear guidelines are available regarding the ideal time for pregnancy for an HIVpositive woman (with respect to CD4 level, stage of treatment, treatment regimen, viral load, or HIV and health status of her partner) and this information is urgently needed.  Although 30% of HIV-positive women reported intentions to have (more) children in our study, 70% did not. Thus, it is critical to ensure that effective, nonjudgmental family planning services (including access to termination of pregnancy services) are available to all women who need them. This is of particular importance since evidence has shown that despite lower fertility intentions, infected women  105  may be similarly likely to become pregnant [47-49] and terminate a pregnancy [48] as uninfected women. More work must be done on the provision of contraceptive options to close the gap between reported fertility intentions and fertility.  It must be highlighted that the fertility intentions of HIV-negative women in this HIV hyper-endemic setting were very high (68%). In South Africa overall, the prevalence of HIV infection is highest among young women, which corresponds with the peak reproductive years [3]. Conception requires unprotected sexual activity and the HIV status of the sexual partners for many of these women is unknown. Indeed, only 20% of the general adult population of South Africa knows their HIV status [50, 51]. As such, reproduction for HIV-negative women in Soweto must be considered an important risk for HIV acquisition.  The limitations of this study must be acknowledged. First, the cross-sectional nature of this analysis precludes us from determining causality between the explanatory variable and the outcome, even though fertility intentions were assessed as a future event while HIV and HAART use status were assessed in the present. Although reverse causality is considered unlikely, longitudinal studies are needed to investigate a potentially time-sensitive relationship between HIV status, HAART use, and fertility intentions. Second, there is a risk of social desirability bias whereby HIV-positive women may under-report their fertility intentions because of  106  disapproving views expressed by community health workers and community members [4, 20] . If under-reporting was differential, then our effect estimates are likely somewhat inflated. We took precautions against reporting bias by using standardized questions of fertility intentions and employing non-clinic staff to conduct the interviews. Third, recent literature has described the limitations of using dichotomized measures of “fertility intentions”. Moreover, others have emphasized the dynamic nature of fertility intentions, which we were unable to fully capture in this study [4, 52, 53]. Fourth, there were important baseline differences between the HIV-positive and HIV-negative women in our study, which cannot be fully adjusted for in the analyses. In particular, HIV-positive women in our study were significantly older than HIV-negative women and age is both a known and important predictor of fertility intentions and is associated with a number of other covariates (e.g., parity, education status). In an attempt to address this difference in age, we conducted a sub-analysis of fertility intentions restricted to women less than 30 years of age. We found no differences in the variables that predicted fertility intentions nor the magnitude of the associations. The results of the sub-analysis suggest that our overall findings are robust, despite differences in age at baseline. Finally, a quantitative analysis such as this fails to capture the salient influence of cultural dynamics on fertility decision-making. Cultural beliefs and practices have often and consistently been discussed as critical determinants of 107  fertility intentions [54, 55]. HIV status and HAART use alone are unlikely to be the sole or even primary drivers of reproductive decision-making [23]. Indeed, qualitative studies have highlighted that the desire for motherhood, opinions of partners and health care providers, religious values, and the perceived capacity to successfully parent emerge as critical factors influencing fertility decision-making of HIV-positive women [4, 52].  4.5 CONCLUSION  Our findings suggest important associations between fertility intentions and HIV status, largely irrespective of HAART use. The fertility intention profile of HIVpositive women in Soweto demands that integrated HAART, HIV care, and reproductive health services be made available to support the rights of HIV-positive women to safely achieve their fertility goals, while minimizing risks of vertical and horizontal transmission. The substantial fertility intentions of HIV-negative women in this HIV hyper-endemic region are of great importance and demand consideration of targeted reproductive health services to minimize their risks of HIV acquisition through realization of their reproductive goals.  108  Table 4.1: Baseline characteristics of HAART users, non-HAART users, and HIVnegative women (aged 18-44 years and non-sterilized) in Soweto, South Africa (n=674) Variable  HAART users  Non-HAART  HIV-negative  Overall  (n=217)  users (n=215)  (n=242)  (n=674)  n (%)  n (%)  n (%)  n (%)  32.0 [5.7]  25.0 [6.0]  30.0 [6.7]  p-value¥  Socio-demographic Characteristics Mean Age (yrs) [SD]  33.5 [5.0]  <0.0001 <0.0001  Age Group (yrs) 18-24  6 (3%)  22 (10%)  138 (57%)  166 (25%)  25-29  38 (18%)  47 (22%)  47 (19%)  132 (20%)  30-34  83 (38%)  75 (35%)  41 (17%)  199 (30%)  35-39  59 (27%)  45 (21%)  7 (3%)  111 (17%)  40-44  30 (14%)  25 (12%)  9 (4%)  64 (10%)  Education  <0.0001  Less than Grade 12  138 (64%)  128 (60%)  75 (31%)  341 (51%)  Grade 12 or higher  78 (36%)  87 (40%)  167 (69%)  332 (49%)  Employment Status  0.0718  Employed  89 (41%)  91 (43%)  80 (33%)  260 (39%)  Unemployed  127 (59%)  122 (57%)  162 (67%)  411 (61%)  Household income (per  <0.0001  month) Less than 3000 ZAR  182 (84%)  169 (79%)  125 (52%)  476 (71%)  3,000 or more ZAR  21 (10%)  35 (16%)  75 (31%)  131 (19%)  Don’t know/Refused  14 (6%)  11 (5%)  42 (17%)  67 (10%) 0.0004  Currently in a sexual relationship No  62 (29%)  51 (24%)  33 (14%)  146 (22%)  Yes  155 (71%)  164 (76%)  209 (86%)  528 (78%)  0.77 [0.48]  0.82 [0.53]  1.09 [0.88]  0.90 [0.68]  Mean # of sexual partners  <0.0001  in the previous 6 months [SD]  109  Table 4.1 (continued): Baseline characteristics of HAART users, non-HAART users, and HIV-negative women (aged 18-44 years and non-sterilized) in Soweto, South Africa (n=674) Variable  HAART users  Non-HAART  HIV-negative  Overall  (n=217)  users (n=215)  (n=242)  (n=674)  n (%)  n (%)  n (%)  n (%)  p-value¥  <0.0001  HIV status of regular sexual partner/husband* Don’t Know  57 (38%)  72 (49%)  78 (41%)  207 (42%)  HIV-negative  28 (19%)  24 (16%)  109 (58%)  161 (33%)  HIV-positive  66 (44%)  52 (35%)  2 (1%)  120 (25%)  1.8 [1.1]  1.8 [1.1]  0.80 [0.9]  1.4 [1.1]  Mean parity [SD] Ever lost a child**  <0.0001  No  152 (78%)  168 (87%)  132 (99%)  452 (86%)  Yes  44 (22%)  25 (13%)  2 (1%)  71 (14%) <0.0001  Number of living 29 (13%)  24 (11%)  109 (45%)  162 (24%)  0  80 (37%)  77 (36%)  82 (34%)  239 (35%)  1  108 (50%)  114 (53%)  51 (21%)  273 (41%)  49.8 [33.1]  N/A  58.7 [35.6]  children  <0.0001  2 or more HIV history and clinical characteristics Mean # of months since  67.6 [35.8]  <0.0001  HIV diagnosis [SD] Recent CD4 count  0.0088  < 200  37 (17%)  39 (18%)  N/A  76 (18%)  200 to < 350  52 (24%)  79 (37%)  131 (31%)  350 or greater  124 (58%)  95 (45%)  219 (51%)  Nadir CD4 count  <0.0001  < 50  62 (29%)  3 (1%)  N/A  65 (15%)  50 to < 200  139 (65%)  44 (21%)  183 (43%)  200 to < 350  5 (2%)  83 (39%)  88 (21%)  350 or greater  7 (3%)  83 (39%)  90 (21%)  110  Table 4.1 (continued): Baseline characteristics of HAART users, non-HAART users, and HIV-negative women (aged 18-44 years and non-sterilized) in Soweto, South Africa (n=674) Variable  HAART users  Non-HAART  HIV-negative  Overall  (n=217)  users (n=215)  (n=242)  (n=674)  n (%)  n (%)  n (%)  n (%)  WHO Stage of Disease Stage I/II Stage III/IV  p-value¥  0.5868 207 (97%)  205 (96%)  6 (3%)  8 (4%)  N/A  412 (97%) 14 (3%)  Disclosed HIV status to  0.0032  anybody  5 (2%)  19 (9%)  No  211 (98%)  196 (91%)  N/A  24 (6%) 407 (94%)  Yes Notes: ¥ Differences between groups are reported using Pearson’s chi-squared test statistic (for categorical variables) and Student’s independent t-test or ANOVA (for continuous variables). * Restricted to those who reported having a regular sexual partner/husband and who responded to the question about partner’s status (n=488) ** Restricted to those who reported having had a livebirth (n=523) SD = Standard Deviation  111  Table 4.2: Univariate and Adjusted analyses of variables associated with fertility intentions among non-sterilized women aged 18-44 years in Soweto, South Africa (n=674) Variable  Fertility intentions No (%)  Yes (%)  (n=380)  (n=294)  HIV-negative  78 (21%)  HIV-positive, HAART-  Crude OR  Adjusted OR  OR  95% CI  AOR  95% CI  164 (56%)  Ref.  Ref.  Ref.  Ref.  152 (40%)  63 (21%)  0.20  0.13, 0.29  0.35  0.21, 0.60  150 (39%)  67 (23%)  0.21  0.14, 0.32  0.40  0.23, 0.69  18-24  52 (14%)  114 (39%)  Ref.  Ref.  Ref.  Ref.  25-29  68 (18%)  64 (22%)  0.43  0.27, 0.69  1.88  1.00, 3.52  30-34  124 (33%)  75 (26%)  0.28  0.18, 0.43  1.44  0.78, 2.64  35-39  83 (22%)  28 (10%)  0.15  0.09, 0.26  1.29  0.62, 2.71  40-44  52 (14%)  12 (4%)  0.11  0.05, 0.21  0.84  0.34, 2.06  Less than Grade 12  222 (59%)  119 (40%)  Ref.  Ref.  Ref.  Ref.  Grade 12 or higher  157 (41%)  175 (60%)  2.08  1.53, 2.84  0.92  0.61, 1.39  Unemployed  230 (61%)  181 (62%)  Ref.  Ref.  --  --  Employed  147 (39%)  113 (38%)  0.98  0.71, 1.34  Less than 3000 ZAR  305 (80%)  171 (58%)  Ref.  Ref.  Ref.  Ref.  3,000 or more ZAR  50 (13%)  81 (28%)  2.89  1.94, 4.31  1.40  0.84, 2.33  DK/Refused  25 (7%)  42 (14%)  3.00  1.77, 5.09  1.73  0.90, 3.33  HIV & HAART Use  naive HIV-positive, receiving HAART Age Group (yrs)  Education  Employment Status  Household income (per month)  112  Table 4.2 (continued): Univariate and Adjusted analyses of variables associated with fertility intentions among non-sterilized women aged 18-44 years in Soweto, South Africa (n=674) Variable  Fertility intentions No (%)  Yes (%)  (n=380)  (n=294)  No  111 (29%)  Yes  Crude OR  Adjusted OR  OR  95% CI  AOR  95% CI  35 (12%)  Ref.  Ref.  Ref.  Ref.  269 (71%)  259 (88%)  3.05  2.01, 4.63  3.07  1.86, 5.05  0  33 (9%)  129 (44%)  Ref.  Ref.  Ref.  Ref.  1  117 (31%)  122 (42%)  0.27  0.17, 0.42  0.30  0.18, 0.50  2+  230 (61%)  43 (15%)  0.05  0.03, 0.08  0.06  0.03, 0.11  Currently in a sexual relationship  Number of living children  Notes: ‘Ref.’ refers to the ‘Reference category’.  113  Table 4.3: Univariate and adjusted analyses of variables associated with fertility intentions among non-sterilized HIV-positive women aged 18-44 years in Soweto, South Africa (n=432) Variable  Fertility Intentions  Crude OR  Adjusted OR  No  Yes  OR  95% CI  OR  95% CI  HAART-naive  152 (50%)  63 (48%)  Ref.  Ref.  Ref.  Ref.  Receiving HAART  150 (50%)  67 (52%)  1.08  0.71, 1.63  1.16  0.72, 1.86  33.2 [5.6]  31.7 [4.9]  0.95  0.91, 0.99  0.99  0.94, 1.04  Less than Grade 12  110 (37%)  55 (42%)  Ref.  Ref.  --  --  Grade 12 or higher  191 (63%)  75 (58%)  1.27  0.84, 1.94  Unemployed  179 (60%)  70 (54%)  Ref.  Ref.  --  --  Employed  120 (40%)  60 (46%)  1.28  0.84, 1.94  < than 3000 ZAR  256 (85%)  95 (73%)  Ref.  Ref.  Ref.  Ref.  ≥ 3,000 or more  32 (11%)  24 (18%)  2.02  1.13, 3.61  1.64  0.85, 3.17  Don’t Know  14 (5%)  11 (8%)  2.13  0.93, 4.83  1.90  0.75, 4.80  No  94 (31%)  19 (15%)  Ref.  Ref.  Ref.  Ref.  Yes  208 (69%)  111 (85%)  2.64  1.53, 4.55  2.98  1.63, 5.46  0  19 (6%)  34 (26%)  Ref.  Ref.  Ref.  Ref.  1  92 (30%)  65 (50%)  0.40  0.21, 0.75  0.36  0.18, 0.71  2+  191 (63%)  31 (24%)  0.09  0.05, 0.18  0.09  0.04, 0.17  58.2 [35.2]  59.9 [36.3]  1.00  1.00, 1.01  --  --  HAART Use  Age (per increase in year) Education  Employment Status  Household income (per month)  Currently in a sexual relationship  Number of living children  Mean # of months since HIV dx [SD]  114  Table 4.3 (continued): Univariate and adjusted analyses of variables associated with fertility intentions among non-sterilized HIV-positive women aged 18-44 years in Soweto, South Africa (n=432)  Variable  Fertility Intentions  Crude OR  Adjusted OR  No  Yes  OR  95% CI  OR  95% CI  < 200  55 (19%)  21 (16%)  Ref.  Ref.  --  --  200 to < 350  89 (30%)  42 (33%)  1.25  0.66, 2.30  350 or greater  153 (52%)  66 (51%)  1.13  0.63, 2.02  < 50  45 (15%)  20 (16%)  Ref.  Ref.  --  --  50 to < 200  124 (42%)  59 (46%)  1.07  0.58, 1.97  200 to < 350  64 (22%)  24 (19%)  0.84  0.42, 1.71  350 or greater  64 (22%)  26 (20%)  0.91  0.46, 1.83  --  --  --  --  Recent CD4  Nadir CD4  WHO Stage of Disease Stage I/II Stage III/IV  286 (96%)  126 (98%)  Ref.  Ref.  11 (4%)  3 (2%)  0.62  0.17, 2.26  Disclosed HIV status to anybody No  18 (6%)  6 (5%)  Ref.  Ref.  Yes  283 (94%)  124 (95%)  1.31  0.51, 3.39  Notes: ‘Ref.’ refers to the ‘Reference category’.  115  Percent reporting intention to have (more) children  55% 50% 45% 40% 35% 30%  32%  30%  33% 29%  29% 27%  25% 20% 15% 10% 5% 0% 0 months (not on HAART)  < 1 yr  1 - 2 yrs  2 - 3 yrs  3 - 4 yrs  4+ yrs  Duration of HAART use  Figure 4.1: Percent of women (aged 18-44 years and non-sterilized) who intend to have (more) children by length of time on HAART among HIV-positive women in Soweto, South Africa (n=426)  116  4.6 REFERENCES  1.  UNAIDS. Report on the global HIV/AIDS epidemic 2008. Geneva: UNAIDS; 2008.  2.  Department of Health. National HIV and Syphilis Prevalence Survey, South Africa 2006. Pretoria, South Africa: Department of Health; 2007.  3.  Department of Health. South Africa Demographic and Health Survey 2003. Pretoria, South Africa: Department of Health; 2004.  4.  Cooper D, Harries J, Myer L, Orner P, Bracken H, Zweigenthal V. "Life is still going on": reproductive intentions among HIV-positive women and men in South Africa. Soc Sci Med 2007; 65: 274-83.  5.  Homsy J, Bunnell R, Moore D, King R, Malamba S, Nakityo R, et al. Reproductive intentions and outcomes among women on antiretroviral therapy in rural Uganda: a prospective cohort study. PLoS ONE 2009; 4: e4149.  6.  Maier M, Andia I, Emenyonu N, Guzman D, Kaida A, Pepper L, et al. Antiretroviral therapy is associated with increased fertility desire, but not pregnancy or live birth, among HIV+ women in an early HIV treatment program in rural Uganda. AIDS Behav 2009; 13 (Suppl 1): 28-37.  117  7.  Myer L, Morroni C, Rebe K. Prevalence and determinants of fertility intentions of HIV-infected women and men receiving antiretroviral therapy in South Africa. AIDS Patient Care STDS 2007; 21: 278-85.  8.  Nakayiwa S, Abang B, Packel L, Lifshay J, Purcell DW, King R, et al. Desire for children and pregnancy risk behavior among HIV-infected men and women in Uganda. AIDS Behav 2006; 10: S95-104.  9.  Desgrees-Du-Lou A, Msellati P, Viho I, Yao A, Yapi D, Kassi P, et al. Contraceptive use, protected sexual intercourse and incidence of pregnancies among African HIV-infected women. DITRAME ANRS 049 Project, Abidjan 1995-2000. Int J STD AIDS 2002; 13: 462-8.  10.  Gray GE, McIntyre JA. HIV and pregnancy. BMJ 2007; 334: 950-3.  11.  Hollerbach PE. Fertility decision-making processes: a critical essay. New York, New York: Population Council; 1982.  12.  Thornton AC, Romanelli F, Collins JD. Reproduction decision making for couples affected by HIV: a review of the literature. Top HIV Med 2004; 12: 617.  13.  Delvaux T, Nostlinger C. Reproductive choice for women and men living with HIV: contraception, abortion and fertility. Reprod Health Matters 2007; 15: 46-66.  118  14.  Wawer MJ, Gray RH, Sewankambo NK, Serwadda D, Li X, Laeyendecker O, et al. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis 2005; 191: 1403-9.  15.  European Collaborative Study. HIV-infected pregnant women and vertical transmission in Europe since 1986. European collaborative study. AIDS 2001; 15: 761-70.  16.  Mofenson LM. Mother-child HIV-1 transmission: Timing and determinants. Obstet Gynecol Clin North Am 1997; 24: 759-84.  17.  Nunn AJ, Mulder DW, Kamali A, Ruberantwari A, Kengeya-Kayondo JF, Whitworth J. Mortality associated with HIV-1 infection over five years in a rural Ugandan population: cohort study. BMJ 1997; 315: 767-71.  18.  Makumbi FE, Gray RH, Serwadda D, Nalugoda F, Kiddugavu M, Sewankambo NK, et al. The incidence and prevalence of orphanhood associated with parental HIV infection: a population-based study in Rakai, Uganda. AIDS 2005; 19: 1669-76.  19.  CDC. Recommendations for assisting in the prevention of perinatal transmission of human T-lymphotropic virus type III/lymphadenopathyassociated virus and acquired immunodeficiency syndrome. MMWR Morb Mortal Wkly Rep 1985; 34: 721-6, 731-2.  119  20.  Myer L, Morroni C, Cooper D. Community attitudes towards sexual activity and childbearing by HIV-positive people in South Africa. AIDS Care 2006; 18: 772-6.  21.  Dyer SJ. The value of children in African countries: insights from studies on infertility. J Psychosom Obstet Gynaecol 2007; 28: 69-77.  22.  Dyer SJ, Abrahams N, Hoffman M, van der Spuy ZM. 'Men leave me as I cannot have children': women's experiences with involuntary childlessness. Hum Reprod 2002; 17: 1663-8.  23.  Kaida A, Andia I, Maier M, Strathdee SA, Bangsberg DR, Spiegel J, et al. The potential impact of antiretroviral therapy on fertility in sub-Saharan Africa. Curr HIV/AIDS Rep 2006; 3: 187-94.  24.  Hogg RS, O'Shaughnessy MV, Gataric N, Yip B, Craib K, Schechter MT, et al. Decline in deaths from AIDS due to new antiretrovirals. Lancet 1997; 349: 1294.  25.  Palella FJ, Jr., Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998; 338: 853-60.  120  26.  Hogg RS, Heath KV, Yip B, Craib KJ, O'Shaughnessy MV, Schechter MT, et al. Improved survival among HIV-infected individuals following initiation of antiretroviral therapy. JAMA 1998; 279: 450-4.  27.  Hogg RS, Yip B, Chan KJ, Wood E, Craib KJ, O'Shaughnessy MV, et al. Rates of disease progression by baseline CD4 cell count and viral load after initiating triple-drug therapy. JAMA 2001; 286: 2568-77.  28.  Cooper ER, Charurat M, Mofenson L, Hanson IC, Pitt J, Diaz C, et al. Combination antiretroviral strategies for the treatment of pregnant HIV-1infected women and prevention of perinatal HIV-1 transmission. J Acquir Immune Defic Syndr 2002; 29: 484-94.  29.  Castilla J, Del Romero J, Hernando V, Marincovich B, Garcia S, Rodriguez C. Effectiveness of highly active antiretroviral therapy in reducing heterosexual transmission of HIV. J Acquir Immune Defic Syndr 2005; 40: 96-101.  30.  Fang CT, Hsu HM, Twu SJ, Chen MY, Chang YY, Hwang JS, et al. Decreased HIV transmission after a policy of providing free access to highly active antiretroviral therapy in Taiwan. J Infect Dis 2004; 190: 879-85.  31.  Berer M. HIV/AIDS, sexual and reproductive health: intersections and implications for national programmes. Health Policy Plan 2004; 19 (Suppl 1): i62-i70.  121  32.  Gruskin S, Ferguson L, O'Malley J. Ensuring sexual and reproductive health for people living with HIV: an overview of key human rights, policy and health systems issues. Reprod Health Matters 2007; 15: 4-26.  33.  Pritchett LH. Desired fertility and the impact of population policies. Popul Dev Rev 1994; 20: 1-55.  34.  Department of Health. National antiretroviral treatment guidelines. Pretoria, South Africa: Department of Health; 2004.  35.  World Health Organization (WHO). Scaling up antiretroviral therapy in resource-limited settings: treatment guidelines for a public health approach. 2003 Revisions. Geneva, Switzerland: WHO; 2004.  36.  Becker S, Sutradhar SC. Fertility intentions: are the undecided more like those who want more or want no more children? J Biosoc Sci 2007; 39: 137-45.  37.  Pagano M, Gauvreau K. Principles of biostatistics, 2nd ed. Pacific Grove, CA: Duxbury; 2000.  38.  Van Ness PH, Allore HG. Using SAS to investigate effect modification. Paper 105-31. SUGI 31. Statistics and Data Analysis 2007; 31:1-10. Available at: http://www2.sas.com/proceedings/sugi31/195-31.pdf. Accessed October 10th, 2007.  122  39.  The SAS Institute. SAS Version 9.1. Cary, NC: The SAS Institute; 2005.  40.  Stanwood NL, Cohn SE, Heiser JR, Pugliese M. Contraception and fertility plans in a cohort of HIV-positive women in care. Contraception 2007; 75: 294-8.  41.  Ogilvie GS, Palepu A, Remple VP, Maan E, Heath K, MacDonald G, et al. Fertility intentions of women of reproductive age living with HIV in British Columbia, Canada. AIDS 2007; 21 (Suppl 1): S83-8.  42.  Kapp C. South Africa unveils new 5-year HIV/AIDS plan. Lancet 2007; 369: 1589-90.  43.  World Health Organization (WHO). Towards universal access: scaling up priority HIV/AIDS interventions in the health sector: Progress report 2008. Geneva, Switzerland: World Health Organization (WHO); 2008.  44.  Chen JL, Philips KA, Kanouse DE, Collins RL, Miu A. Fertility desires and intentions of HIV-positive men and women. Fam Plann Perspect 2001; 33: 14452, 165.  45.  Ekouevi DK, Coffie PA, Becquet R, Tonwe-Gold B, Horo A, Thiebaut R, et al. Antiretroviral therapy in pregnant women with advanced HIV disease and pregnancy outcomes in Abidjan, Cote d'Ivoire. Aids 2008; 22: 1815-20.  123  46.  Montaner JS, Hogg R, Wood E, Kerr T, Tyndall M, Levy AR, et al. The case for expanding access to highly active antiretroviral therapy to curb the growth of the HIV epidemic. Lancet 2006; 368: 531-6.  47.  Ahluwalia IB, DeVellis RF, Thomas JC. Reproductive decisions of women at risk for acquiring HIV infection. AIDS Educ Prev 1998; 10: 90-7.  48.  Pivnick A. Loss and regeneration: influences on the reproductive decisions of HIV positive, drug-using women. Med Anthropol 1994; 16: 39-62.  49.  Selwyn PA, Schoenbaum EE, Davenny K, Robertson VJ, Feingold AR, Shulman JF, et al. Prospective study of human immunodeficiency virus infection and pregnancy outcomes in intravenous drug users. JAMA 1989; 261: 1289-94.  50.  Kalichman SC, Simbayi LC. HIV testing attitudes, AIDS stigma, and voluntary HIV counselling and testing in a black township in Cape Town, South Africa. Sex Transm Infect 2003; 79: 442-7.  51.  Shinsasa O, Simbayi LC. Nelson Mandela/HSRC study of HIV/AIDS: South African National HIV Prevalence, Behavioral Risks and Mass Media, Household Survey, 2002. Cape Town, South Africa: Human Sciences Research Council; 2002.  124  52.  Kirshenbaum SB, Hirky AE, Correale J, Goldstein RB, Johnson MO, Rotheram-Borus MJ, et al. "Throwing the dice": pregnancy decision-making among HIV-positive women in four U.S. cities. Perspect Sex Reprod Health 2004; 36: 106-13.  53.  Taulo F, Berry M, Tsui A, Makanani B, Kafulafula G, Li Q, et al. Fertility intentions of HIV-1 infected and uninfected women in Malawi: a longitudinal study. AIDS Behav 2009; 13 (Suppl 1): 20-7.  54.  Sonko S. Fertility and culture in sub-Saharan Africa: a review. Int Soc Sci J 1994; 46: 397-411.  55.  Caldwell JC, Caldwell P. The cultural context of high fertility in sub-Saharan Africa. Population and Development Review 1987; 13: 409-437.  125  CHAPTER 54: CONTRACEPTIVE USE AND METHOD MIX PATTERNS AMONG HIV-POSITIVE AND HIV-NEGATIVE WOMEN IN SOWETO, SOUTH AFRICA IN THE CONTEXT OF EXPANDING ACCESS TO HIGHLY ACTIVE ANTIRETROVIRAL THERAPY  5.1 INTRODUCTION  Nearly 80% of the world’s 15.5 million HIV-infected women live in subSaharan Africa, where heterosexual intercourse is the primary mode of HIV transmission [1]. Women living with HIV face many of the same contraceptive and reproductive decision-making challenges as HIV-negative women in addition to the HIV-specific risks to maternal, fetal, and partner health associated with conception and pregnancy.  Increasing use of effective contraception is an important strategy towards preventing unintended pregnancies in all women and a critical factor towards  4  A version of this chapter has been submitted for publication. Kaida A, Laher F, Strathdee SA,  Money D, Janssen PA, Hogg RS, Gray G. Contraceptive use and method mix patterns among HIVpositive and HIV-negative women in Soweto, South Africa in the context of expanding access to highly active antiretroviral therapy.  126  reducing maternal and infant mortality [2]. Among HIV-positive women, provision of contraceptive services to prevent unwanted pregnancies remains the “best kept secret” [3] of the four WHO pillars for preventing mother-to-child-transmission (PMTCT) of the virus. A recent modeling study showed that a PMTCT strategy focused on increasing contraception among HIV-positive women could avert 29% more HIV-positive births than prophylactic nevirapine alone, at the same level of expenditure [4]. However, the prevailing under-emphasis of reproductive health within HIV programming is evident in the numbers: an estimated 50-84% of pregnancies among HIV-infected women are unintended [5-7] and each year in subSaharan Africa nearly 350,000 infants are infected with HIV via MTCT [1].  The little that is known about the prevalence and type of contraceptive use among HIV-infected women originates from studies conducted prior to widespread availability of highly active antiretroviral therapy (HAART) in sub-Saharan Africa [5, 8-10]. This is due, in part, to the recency of the dramatic HAART scale-up effort in the region [11]. By increasing life expectancy [12-14], decreasing morbidity [13, 15], and reducing vertical [16] and sexual [17] transmission risks, expanding access to HAART is dramatically reducing the health risks and barriers to reproduction among HIV-affected individuals and couples. This emerging reality of HIV as a manageable chronic disease, with HIV-infected individuals anticipated to live well into (and past) their peak reproductive years, has highlighted the importance of 127  assessing the potential behavioral and biological impacts of HAART on contraceptive use, safety, and efficacy [18-25].  Given the HIV hyper-endemic context of reproduction in Soweto, South Africa [1, 26], the primary objective of this study was to assess the prevalence of contraceptive use and to determine whether contraceptive use varies according to HIV serostatus and use and duration of HAART among sexually active women aged 18-44 years. A secondary objective was to determine the types of contraceptive methods used and whether contraceptive method profiles vary by HIV and HAART use. This research was conducted within Kaida et al’s conceptual framework of the potential impact of HAART on fertility in sub-Saharan Africa [19], where HAART use is hypothesized to reduce individuals’ perceived risk of HIV transmission and disease progression, ease concerns about the risks of reproduction, and alter contraceptive use patterns. As such, we hypothesized that HIV-positive women receiving HAART would be less likely to use contraception compared with HIVpositive women not receiving HAART, with increasing duration of HAART use associated with incrementally lower contraceptive use. Also, we hypothesized that contraceptive prevalence among HIV-positive women receiving HAART and HIVnegative women would be similar.  128  5.2 METHODS  5.2.1 Study setting  At 5.7 million, South Africa has the largest absolute number of people living with HIV in the world, 60% of whom are women [1]. The adult (aged 15-49 years) prevalence of HIV is 19% [1] and in 2006, 30% of pregnant women tested during antenatal clinic surveillance were HIV-positive. Gauteng province, where Soweto is located, has an antenatal HIV prevalence of 31% [26].  HAART became available in public sector clinics in 2004 and by the end of 2007 an estimated 460 000 patients were receiving HAART, an antiretroviral therapy coverage of 28% [11]. Contraceptive methods, including injectables, oral contraceptives, and male condoms, are available at no-cost in government health centres.  This study was conducted at the Perinatal HIV Research Unit (PHRU), a large clinical and research site housed within Chris Hani Baragwanath Hospital in Soweto. The PHRU sees over 5,000 adult patient visits monthly and provides antiretroviral therapy and clinical care to medically-eligible HIV-positive individuals and ongoing wellness care for those not yet eligible for HAART. The PHRU also operates a Prevention Studies area that includes a Voluntary Counseling and Testing (VCT) centre. All PHRU services are provided free-of-charge. 129  5.2.2 Study design  This analysis is based on cross-sectional survey data of HIV-positive (HAART receiving and HAART-naïve) and HIV-negative women seeking services at the PHRU. A medical chart review was also conducted to confirm HIV serology and HAART use history of HIV-positive women.  5.2.3 Eligibility criteria  To be eligible to participate in the study, women were required to be 18-49 years of age, attending a PHRU clinic, residing in Soweto, competent to give informed consent, and willing to allow medical record review for the purposes of confirming HIV status and HAART history. We considered women to be HAART users if they had been receiving HAART for at least one month. We considered women to be HAART-naïve if they had never taken HAART, except for vertical transmission prophylaxis.  130  5.2.4 Study sample  We enrolled 751 women, including 253 HIV-positive women receiving HAART, 249 HIV-positive but HAART-naïve women, and 249 HIV-negative women. This sampling strategy provided one case group (HAART users) and two comparison groups (HAART-naïve and HIV-negative women).  HAART users were sampled from the PHRU’s PEPFAR Clinic which has provided free antiretroviral therapy to medically eligible patients since July 2004. Currently, the PEPFAR clinic has over 1,000 patients receiving HAART, 75% of whom are female. PEPFAR patients are followed-up every three months and generally receive one of two standard HAART regimens: Regimen 1 is d4T/3TC/EFV or NVP and Regimen 2 is Kaletra/ddI/AZT [27].  HIV-positive HAART-naïve women were sampled from the PHRU’s Wellness Clinic, initiated in January 2003 with the goal of providing preventive care to HIV-positive individuals. Wellness patients are followed-up approximately every six months. When patients are medically eligible for HAART, they are referred to the PEPFAR clinic or to one of the nearby government ART clinics. There are approximately 3,000 active patients in the Wellness Clinic.  HIV-negative women were sampled from the VCT clinic, which was initiated in mid-2002 and sees approximately 400 people per month. Testing is conducted 131  onsite during visits that last an average of two hours. Approximately 65% of attendees are women and 30% of all attendees test HIV-positive.  For this analysis on contraceptive use, we restricted the study sample to women aged 18-44 years who were currently sexually active (i.e., reported sexual activity in the previous six months) and not currently pregnant. This was done to enhance the comparability of findings with other studies investigating reproductive and sexual health among HIV-positive populations. The restriction yielded an analytic sample of 563 women (75% of total sample), including 171 women on HAART, 178 HAART-naïve women, and 214 HIV-negative women.  5.2.5 Data collection  Every female patient attending the PEPFAR Clinic and the VCT clinic was consecutively approached by a research assistant to assess eligibility and interest in participating in the study. Since many more women attend the Wellness Clinic, a list was made of chart numbers of women attending the clinic each day. A random sample of chart numbers (40% of the total number of charts present) was then drawn and the corresponding women were approached to assess eligibility and interest in participating in the study.  132  After confirming eligibility and seeking informed consent, all participants were asked to complete a 15-25 minute interviewer-administered questionnaire in English. The study interviewers were multilingual and trained to ensure accurate and consistent translation of the questionnaire if required or requested by the participant. Pilot testing of 45 women revealed that women were able to understand and answer the questionnaire. Approximately 12 women were interviewed daily by three trained research assistants between May and December 2007. Research assistants were women from the local community who had previous research experience and were recent Social Sciences’ graduates of a local university. Interviewers were supervised by an experienced research nurse. Two research nurses with HIV training conducted the medical record review. Participants were given transport reimbursement as compensation.  5.2.6 Data collection instruments  The questionnaire assessed socio-demographic characteristics; HIV status, diagnosis, and treatment; clinical stage of disease; HAART history; fertility intentions; fertility history; contraceptive practices; and sexual history.  133  We reviewed medical records of HIV-positive women to confirm HIV status and HAART history, and to obtain clinical data including CD4 cell counts and WHO stage of disease. Viral load assessments are only available from women in the PEPFAR Clinic (i.e., HAART users). The medical record was considered the referent measure for inconsistencies between self-reported and medical record data.  5.2.7 Measures  The primary outcome was self-reported contraceptive use in the previous six months. Contraceptive methods queried included male and female condoms (restricted to those reporting “Always” use), injections (depomedroxyprogesterone acetate (DMPA) or norethisterone enantate), oral contraceptive pill, diaphragm, IUD, female sterilization, and hysterectomy. In assessing the contraceptive method profile, dual protection was defined as use of both a barrier contraceptive method (primarily the male condom) and use of a hormonal or permanent contraceptive method [28].  The primary explanatory variable was current HAART use. The secondary explanatory variable was HIV status. Covariates included age, education, employment, household income, current sexual partnership, HIV status of regular  134  sexual partners, parity, number of living children, fertility intentions, and HIV clinical variables.  5.2.8 Statistical analysis  We computed and compared the prevalence of contraceptive use between HIVpositive women and HIV-negative women and then between each of the three groups of women. We conducted two separate models to measure the presence and strength of the association between HAART use and the odds of contraceptive use, controlling for covariates. The first model compared HAART users and HAART-naïve women to HIV-negative women. The second model compared HAART users to HAART-naïve women and allowed adjustment for HIV-associated clinical characteristics.  In both models, univariate analyses were used to assess the relationship between HIV status, receipt of HAART, contraceptive use, and covariates. Differences in contraceptive use between groups are reported using Pearson’s chi-squared test (for categorical variables) and ANOVA, or Student’s independent t-test (for continuous variables). After testing for co-linearity (using Spearman’s rho (ρ)) [29] and interaction [30], all covariates with significant associations in the univariate analysis were included in multivariate logistic regression models to obtain adjusted estimates of the  135  association. Age was forced into both multivariate models regardless of its univariate associations. All statistical tests were two-sided and considered significant at α = 0.05.  Among women who reported using contraception, we analyzed types of methods used by women in each of the three groups and overall. In addition to reporting use of each contraceptive method individually, we collapsed women into four mutually exclusive groups including “Dual protection”, “Consistent condom use only”, “Hormonal/Permanent method only”, and “Not using any contraceptive method” and tested for differences by HIV and HAART use status using Pearson’s chi-squared test.  We assessed the association between duration of HAART use and prevalence of contraceptive use using Pearson’s test for trend. HAART-naïve women were included in the “0 months on HAART” category.  5.2.9 Sub-analyses  We conducted the same analyses described above but restricted our sample to women aged 18-34 years to investigate the potential impact of differences in mean baseline age between HIV-positive and HIV-negative women in our study. In addition, this age group corresponds with the peak childbearing years among women in South Africa [31]. 136  5.2.10 Ethical considerations  All participants provided informed consent and all procedures were approved by the Human Research Ethics Committee of the University of the Witwatersrand, the University of British Columbia Health Research Ethics Board, the Simon Fraser University Office of Research Ethics, and the University of California San Diego Institutional Review Board. Information letters and consent forms were available in English and two local languages (isi-Zulu and Sesotho) to ensure comprehensive understanding of the study objectives, potential risks, and benefits.  5.3 RESULTS  Of 801 women approached for participation, 751 consented, completed the questionnaire, and had their medical records reviewed (participation rate = 94%). This analysis was restricted to 563 sexually active, non-pregnant women aged 18-44 years.  137  5.3.1 Baseline characteristics  As shown in Table 5.1, there were important differences in baseline characteristics by HIV and HAART use status. Mean age was 30 years [SD=6.7], however, HIV-negative women were significantly younger than HIV-positive women. Half of women had less than a grade 12 education, 62% were unemployed, and 71% had a monthly household income less than 3,000 ZAR ($380 USD). Nearly one-quarter (23%) reported that her primary sexual partner was HIV-positive, 29% reported that he was HIV-negative, and 42% did not know her partner’s HIV status. Mean parity was 1.5 [SD=1.2] and 44% of women had two or more living children. Nearly half (45%) reported that they intended to have more children.  Among HIV-positive women (n=349), mean time since first HIV diagnosis was 59.7 months [SD=35.4]. Half had recent CD4 counts ≥ 350 cells/mm3 and 12% had nadir CD4 counts < 50 cells/mm3. Nearly all women were in WHO Stage of Disease I or II (98%) and 95% had disclosed their HIV status to someone.  Among HAART users (n=171), median duration of HAART use was 31 months [IQR: 28, 33], ranging from one to 89 months. Eighty percent of HAART users with recorded viral load measures were virally suppressed (<50 copies/ml).  138  5.3.2 Prevalence of contraceptive use  Overall contraceptive prevalence was 78%. This varied significantly by HIV status with 84% of HIV-positive women (including 86% of HAART users and 82% of HAART-naïve women) and 69% of HIV-negative women reporting current contraceptive use (p<0.0001).  5.3.3 Types of contraceptive methods used  The contraceptive method profiles are shown in Table 5.2. Part (a) of the Table 5.2 shows mutually exclusive groups of contraceptive users. As shown, HIVpositive women overall were significantly more likely to use practice dual protection compared with HIV-negative women (33% and 14%, respectively). Much of this difference was accounted for by HAART users, of whom 40% reported using dual protection compared with 24% of HAART-naïve women and 14% of HIV-negative women. HAART users were also significantly more likely to report using condoms (with or without hormonal/permanent methods) (68%) and hormonal/permanent methods (with or without condoms) (58%) compared with non-HAART users and HIV-negative women (p<0.0001).  139  Of the 411 women reporting contraceptive use (Table 5.2 part (b)), 56% used hormonal contraception, 69% used barrier methods (mainly the male condom), and 7% used permanent methods (i.e., hysterectomy and/or female sterilization) with significant differences by HIV status and HAART use. Across all three groups, hormonal contraceptive users utilised injectables more commonly than oral contraceptives. HAART users were significantly more likely to use condoms (79%), compared with non-HAART users (72%), and HIV-negative women (57%) (p=0.0001). Higher proportions of HIV-positive women had a tubal ligation or hysterectomy compared with HIV-negative women (p=0.014).  5.3.4 Univariate and adjusted analysis of contraceptive use: HAART users, HAART non-users, and HIV-negative women  In the unadjusted analyses, many of the measured baseline covariates were significantly associated with contraceptive use (Table 5.3). Compared with HIVnegative women, HIV positive women were significantly more likely to use contraceptive methods (HAART users OR: 2.73; 95% CI: 1.62, 4.59; and non-HAART users OR: 2.04; 95% CI: 1.26, 3.29).  In adjusted analyses, compared with HIV-negative women, HAART users remained significantly more likely to use contraception (OR: 2.40; 95% CI: 1.25, 4.62)  140  while non-HAART users were marginally more likely to use contraception (OR: 1.59; 95% CI: 0.88, 2.85). Overall, HIV-positive women (combining HAART users and HAART-naïve women) had increased adjusted odds of 1.75 (95% CI: 1.02, 2.99) of using contraception compared with HIV-negative women. Younger age, having two or more living children, and expressing an intention not to have (more) children also remained significantly associated with contraceptive use.  5.3.5 Univariate and adjusted analysis: HIV-positive women  As shown in Table 5.4, HAART users and non-users were equally likely to report contraceptive use in unadjusted analyses (OR: 1.34; 95% CI: 0.75, 2.39). There were no significant differences in contraceptive use by HIV clinical characteristics.  In adjusted analyses, HAART users were marginally more likely than nonusers to report contraceptive use (AOR: 1.55; 95% CI: 0.84, 2.88). Younger age, having two or more living children, and expressing an intention not to have more children remained most strongly associated with contraceptive use.  141  5.3.6 Contraceptive use by duration of HAART use  There was no association between duration of HAART use and prevalence of contraceptive use, although there was an apparent lower level of contraceptive use for women receiving HAART between one and two years (Figure 5.1).  5.3.7 Sub-analyses  Contraceptive prevalence of young women (18-34 yrs, n=420) was similar to the overall sample and still varied significantly by HIV and HAART use status: 88% of HAART users, 82% of HAART non-users, and 70% of HIV-negative young women reported using contraception (p=0.0007). In multivariate analyses we found that the same variables associated with contraceptive use in the overall sample were similarly associated with contraceptive use among young women. Compared with HIV-negative women, HAART users remained significantly more likely to use contraception (AOR: 2.24; 95% CI: 1.05, 4.99) while HAART-naïve women were similarly likely to use contraception (AOR: 1.27, 95% CI: 0.65, 2.48).  142  5.4 DISCUSSION  We found that HIV-positive women overall were significantly more likely to use contraception compared with HIV-negative women. In particular, and in contrast with our hypothesis, women receiving HAART were significantly more likely to report contraceptive use while HIV-positive HAART-naïve women were marginally more likely to use contraception compared with HIV-negative women. Overall, over 80% of HIV-positive women in our study reported contraceptive use, which falls within the upper range reported for HIV-positive women elsewhere in sub-Saharan Africa (46%-85%) [9, 10, 18, 25, 32, 33]. Contraceptive prevalence among HIV-negative women in our study was 69%, which is highly comparable to estimates among women in the general South African population [31].  Among HIV-positive women, our finding indicating marginally higher prevalence of contraceptive use among HAART users compared with HAART-naïve women is broadly consistent with recent findings from Uganda [18]. To the best of our knowledge, this is the first study to show a stable effect of contraceptive use by duration of HAART use. Other factors associated with contraceptive use included younger age, having two or more living children, and a lack of intention to have more children, all of which are widely reported to influence contraceptive decisionmaking [10, 18, 34].  143  The reasons for higher contraceptive prevalence among HIV-positive women in general and HAART users in particular were not directly explored in this study. However, an important possible reason for the observed differences is that women receiving HIV treatment and care have more regular contact with health care professionals as a function of the clinical follow-up required to monitor the health of individuals receiving therapy. During these regular clinic visits, reproductive and sexual health issues are addressed and the opportunity to discuss and commence use of contraception is presented.  Contraceptive method choice may have implications for both HIV transmission and pregnancy risks and we found important differences in the types of contraceptive methods used by women in each of our three groups. While condoms are recommended to prevent HIV transmission to uninfected sexual partners, they are less effective than hormonal contraception and sterilization at preventing pregnancy [28]. Overall, a substantial proportion of women in this setting report relying exclusively on the male condom for preventing pregnancy (29%). An additional 25% of women report using condoms in conjunction with a hormonal/permanent method of contraception, resulting in over half of our sample reporting consistent condom use with or without another method. HAART users reported the highest prevalence of consistent condom use. Indeed much of the difference in contraceptive prevalence between our three groups was accounted for 144  by the significantly lower prevalence of condom use among HIV-negative women. Reported rates of condom use among HIV-negative women in our study are comparable to those reported in South Africa overall [31].  Unlike barrier methods, permanent and hormonal contraceptive methods are highly effective at preventing pregnancy but have no role in the prevention of HIV transmission [28]. Overall, 7% of women used permanent methods (hysterectomy and/or female sterilization) with small differences by HIV status and HAART use. The overall prevalence of sterilization is slightly lower than reported rates from South African women in general [31].  Compared with HAART-naïve women, HAART users were more likely to use hormonal contraception, and uptake of DMPA injectables exceeded oral contraceptive use. Reports from other settings suggest that DMPA use is rising owing to its discretion and convenient three-month dosing which corresponds with the HAART follow-up schedule [35]. Higher uptake of progesterone-only injectables in HAART users may also reflect provider preference based on concerns about possible interactions between HAART and estrogen-containing oral contraceptives [36]. Available guidelines advise that women receiving antiretroviral agents should use alternative or additional methods of contraception, beyond oral contraceptives [36].  145  There is no single method that can reliably assist HIV-positive women who wish to avoid pregnancy and HIV transmission to sero-discordant partners. Moreover, given concerns noted above about potential interaction between hormonal contraception and antiretroviral agents, dual protection is encouraged. In our study, not only were HAART users significantly more likely to use contraceptive methods overall, they were more likely than non-HAART users and HIV-negative women to use dual protection. Low prevalence of dual protection among HAART-naïve women (24%) may reflect a population at risk of transmitting HIV to sero-discordant partners and, if they do become pregnant, are a population most requiring antiretroviral prophylaxis through PMTCT services.  Overall, 14% of HAART users, 18% of HAART-naïve women, and 31% of HIV-negative women were not using any form of contraceptive, suggesting risk for unintended pregnancy. A high proportion of women were also unaware of their partner’s HIV status (42% overall). As such, the risks of conception-related HIV acquisition or transmission between sero-discordant couples are serious and integrated HIV and sexual and reproductive health services must be provided to help HIV-affected couples safely achieve their fertility goals.  Limitations of this study must be acknowledged. First, the cross-sectional nature of this analysis precludes us from determining causality between the  146  explanatory variable and the outcome, particularly since contraceptive use, HIV status, and HAART use were assessed at the same point in time. Although reverse causality is considered unlikely (i.e., contraceptive use leading to HAART use), longitudinal studies are needed to investigate this relationship and would enable examination of duration of contraceptive use. Second, there is a risk of social desirability bias whereby HIV-positive women may over-report their contraceptive use (and condom use, in particular) because of pressure from health workers and community members to practice protected sex [37, 38] . If over-reporting was differential, then our effect estimates are likely somewhat inflated. We took precautions against reporting bias by using standardized questions of contraceptive use and employing non-clinic staff to conduct the interviews. Third, there were important baseline differences between the HIV-positive and HIV-negative women in our study, a potential source of selection bias, which cannot be fully adjusted for in the analyses. In particular, HIV-positive women in our study were significantly older and age is a known predictor of contraceptive use and is associated with a number of other covariates (e.g., parity, education status). In an attempt to address this limitation, we conducted a sub-analysis of contraceptive use restricted to women less than 35 years of age. We found no differences in the variables that predicted contraceptive use nor the magnitude of the associations. The results of the sub-analysis suggest that our overall findings are robust, despite differences in age 147  at baseline. Finally, a quantitative analysis such as this fails to capture the salient influence of cultural and gender dynamics on contraceptive decision-making. Indeed, qualitative studies from this setting have highlighted the importance of considering the real and perceived side effects of contraceptives and partner influence and status as mitigating factors influencing contraceptive decision-making of HIV-positive women [39].  In conclusion, our results demonstrate that HIV-positive women overall and women accessing HAART services in particular, are more likely to use contraception overall, and more likely to use barrier, permanent, and dual protection methods in particular, compared with their HIV-negative and HAART-naïve counterparts. The contraceptive use profile of HIV-positive and HIV-negative women in Soweto demands further integration of HIV treatment and care services with reproductive and sexual health services, including the provision of effective contraception. Through the prevention of unintended pregnancy, integrated services are likely to benefit maternal and child health, increase primary prevention of vertical transmission, and decrease incidence of conception-related horizontal transmission to discordant sexual partners.  148  Table 5.1: Baseline characteristics of HAART users, non-HAART users, and HIVnegative women (aged 18-44 years, currently sexually active and non-pregnant) in Soweto, South Africa (n=563) Variable  HAART users (n=171) n (%)  Non-HAART users (n=178) n (%)  HIV-negative (n=214) n (%)  Overall (n=563) n (%)  p-value¥  32.3 [5.6]  25.3 [6.0]  30.0 [6.7]  <0.0001  Socio-demographic Characteristics Mean Age (yrs) [SD]  33.7 [5.0]  <0.0001  Age Group (yrs) 18-24  4 (2%)  15 (8%)  119 (56%)  138 (25%)  25-29  30 (18%)  41 (23%)  43 (20%)  114 (20%)  30-34  66 (39%)  63 (36%)  37 (17%)  166 (30%)  35-39  43 (25%)  35 (20%)  7 (3%)  85 (15%)  40-44  27 (16%)  23 (13%)  8 (4%)  58 (10%)  Education  <0.0001  Less than Grade 12  114 (67%)  109 (61%)  60 (28%)  283 (50%)  Grade 12 or higher  56 (33%)  69 (39%)  154 (72%)  279 (50%)  Employment Status  0.0962  Employed  71 (42%)  75 (42%)  70 (33%)  216 (38%)  Unemployed  100 (58%)  103 (58%)  144 (67%)  347 (62%) <0.0001  Household income (per month) Less than 3000 ZAR  142 (83%)  146 (82%)  111 (52%)  399 (71%)  3,000 or more ZAR  16 (9%)  24 (13%)  65 (30%)  105 (19%)  Don’t know/Refused  13 (8%)  8 (4%)  38 (18%)  59 (10%)  149  Table 5.1 (continued): Baseline characteristics of HAART users, non-HAART users, and HIV-negative women (aged 18-44 years, currently sexually active and nonpregnant) in Soweto, South Africa (n=563) Variable  HAART users (n=171) n (%)  Non-HAART users (n=178) n (%)  HIV-negative (n=214) n (%)  Overall (n=563) n (%)  0.6337  Currently in a sexual relationship No  12 (7%)  16 (9%)  14 (7%)  42 (7%)  Yes  159 (93%)  162 (91%)  200 (93%)  521 (93%) <0.0001  HIV status of regular sexual partner/husband Don’t Know  64 (37%)  87 (49%)  83 (39%)  234 (42%)  HIV-negative  30 (18%)  22 (12%)  111 (52%)  163 (29%)  HIV-positive  69 (40%)  59 (33%)  2 (1%)  130 (23%)  8 (5%)  10 (6%)  18 (8%)  36 (6%)  1.9 [1.1]  1.9 [1.2]  0.85 [0.9]  1.5 [1.2]  Single Mean parity [SD]  p-value¥  Number of living children  <0.0001 <0.0001  0  19 (11%)  19 (11%)  96 (45%)  134 (24%)  1  58 (34%)  56 (31%)  68 (32%)  182 (32%)  2 or more  94 (55%)  103 (58%)  50 (23%)  247 (44%)  Yes  55 (32%)  55 (31%)  146 (68%)  256 (45%)  No  116 (68%)  123 (69%)  68 (32%)  307 (55%)  Fertility Intentions <0.0001  150  Table 5.1 (continued): Baseline characteristics of HAART users, non-HAART users, and HIV-negative women (aged 18-44 years, currently sexually active and nonpregnant) in Soweto, South Africa (n=563) Variable  HAART users (n=171) n (%)  Non-HAART users (n=178) n (%)  HIV-negative (n=214) n (%)  Overall (n=563) n (%)  p-value¥  50.8 [32.1]  N/A  59.7 [35.4]  <0.0001  HIV history and clinical characteristics Mean # of months since HIV diagnosis [SD]  69.0 [36.3]  Recent CD4 count  0.0003  < 200  28 (17%)  34 (19%)  N/A  62 (18%)  200 to < 350  38 (23%)  72 (40%)  110 (32%)  350 or greater  102 (61%)  72 (41%)  174 (50%) <0.0001  Nadir CD4 count N/A  43 (12%)  < 50  41 (24%)  2 (1%)  50 to < 200  116 (69%)  39 (22%)  155 (45%)  200 to < 350  5 (3%)  73 (41%)  78 (23%)  350 or greater  6 (4%)  64 (36%)  70 (20%) 0.5267  WHO Stage of Disease Stage I/II Stage III/IV Disclosed HIV status to anybody No  165 (98%)  173 (97%)  3 (2%)  5 (3%)  N/A  338 (98%) 8 (2%) 0.0493  4 (2%)  12 (7%)  167 (98%)  166 (93%)  N/A  16 (5%) 333 (95%)  Yes Notes: ¥ Differences between groups are reported using Pearson’s chi-squared test statistic (for categorical variables) and Student’s independent t-test or ANOVA (for continuous variables). SD = Standard Deviation N/A = Not Applicable  151  Table 5.2: Types of contraceptive methods used by HIV-positive (HAART users and non-HAART users) and HIVnegative women (aged 18-44 years, currently sexually active and non-pregnant) in Soweto, South Africa HIV-positive women  Overall Contraceptive Prevalence  HAART users (n=171) n (%) 86%  HAART nonusers (n=178) n (%) 82%  All HIV-positive women (n=349) n (%) 84%  HIV-negative women (n=214)  Overall (n=563)  n (%)  n (%)  69%  78%  14%  25%  30%  24%  25%  29%  31%  22%  p-value§  <0.0001  (a) Mutually exclusive categories of type of contraceptive method used: Dual protection (Hormonal/ permanent method AND  40%  24%  18%  23%  28%  35%  14%  18%  33%  consistent condom use) Hormonal/Permanent method  20%  only Consistent condom use only Not using any contraceptive  31% 16%  method <0.0001  152  Table 5.2 (continued): Types of contraceptive methods used by HIV-positive (HAART users and non-HAART users) and HIV-negative women (aged 18-44 years, currently sexually active and non-pregnant) in Soweto, South Africa HIV-positive women HAART HAART nonusers users (n=147) (n=146) n (%) n (%) (b) Contraceptive method mix among contraceptive users (n=411)*: Hormonal Methods  All HIV-positive women (n=293) n (%)  HIV-negative women (n=148)  Overall (n=441)  n (%)  n (%)  p-value§  88 (60%)  68 (47%)  156 (53%)  89 (60%)  245 (56%)  0.0211  82 (56%)  49 (34%)  131 (45%)  66 (45%)  197 (45%)  0.0007  6 (4%)  19 (13%)  25 (9%)  23 (16%)  48 (11%)  0.0041  116 (79%)  105 (72%)  221 (75%)  85 (57%)  306 (69%)  0.0001  116 (79%)  105 (72%)  221 (75%)  84 (57%)  305 (69%)  0.0001  0 (0%)  0 (0%)  0 (0%)  1 (1%)  1 (<1%)  n/a  Permanent methods  12 (8%)  20 (14%)  32 (11%)  4 (3%)  33 (7%)  0.0140  Hysterectomy  4 (3%)  5 (3%)  9 (3%)  0 (0%)  9 (2%)  0.0897  Female sterilization  8 (5%)  15 (10%)  23 (8%)  4 (3%)  27 (6%)  0.0234  Injections Oral Contraceptive Pill Barrier Methods Consistent male condom use Diaphragm  Notes: § p-value from chi-squared test statistics comparing proportions across three groups: HAART-users, HAART-naïve, and HIV-negative women; * Values may not total 100% because one woman may report using more than one method. 153  Table 5.3: Univariate and adjusted analyses of variables associated with contraceptive use among women (aged 18-44 years, currently sexually active and non-pregnant) in Soweto, South Africa (n=563) Variable  Contraceptive Use No (%) Yes (%) (n=122) (n=441)  Crude OR OR 95% CI  Adjusted OR AOR 95% CI  HIV-negative  66 (54%)  148 (34%)  Ref.  Ref.  Ref.  Ref.  HIV-positive,  32 (26%)  146 (33%)  2.04  1.26, 3.29  1.59  0.88, 2.85  24 (20%)  147 (33%)  2.73  1.62, 4.59  2.40  1.25, 4.62  29.0  30.3  [SD=7.4]  [SD=6.5]  1.03  1.00, 1.06  0.94  0.90, 0.98  Less than Grade 12  45 (37%)  238 (54%)  Ref.  Ref.  Ref.  Ref.  Grade 12 or higher  77 (63%)  202 (46%)  0.50  0.33, 0.75  0.70  0.44, 1.13  Unemployed  67 (55%)  280 (63%)  Ref.  Ref.  --  --  Employed  55 (45%)  161 (37%)  0.70  0.47, 1.05  Less than 3000 ZAR  72 (59%)  327 (74%)  Ref.  Ref.  Ref.  Ref.  3,000 or more ZAR  33 (27%)  72 (16%)  0.48  0.30. 0.78  0.94  0.55, 1.63  DK/Refused  17 (14%)  42 (10%)  0.54  0.29, 1.01  0.73  0.37, 1.43  No  12 (10%)  30 (7%)  Ref.  Ref.  --  --  Yes  110 (90%)  411 (93%)  1.50  0.74, 3.02  HIV and HAART Use Status  HAART-naïve HIV-positive, receiving HAART Age (per increase in year) Education  Employment Status  Household income (per month)  Currently in a sexual relationship  154  Table 5.3: (continued) Univariate and adjusted analyses of variables associated with contraceptive use among women (aged 18-44 years, currently sexually active and non-pregnant) in Soweto, South Africa (n=563) Variable  Contraceptive Use No (%) Yes (%) (n=122) (n=441)  Crude OR OR 95% CI  Adjusted OR AOR 95% CI  HIV status of regular sexual partner/  --  --  husband Don’t Know  50 (41%)  184 (42%)  Ref.  Ref.  HIV-negative  42 (34%)  121 (27%)  0.78  0.49, 1.25  HIV-positive  22 (18%)  108 (25%)  1.33  0.77, 2.32  8 (7%)  28 (6%)  0.95  0.41, 2.22  0  43 (35%)  91 (21%)  Ref.  Ref.  Ref.  Ref.  1  50 (41%)  132 (30%)  1.25  0.77, 2.03  1.01  0.59, 1.73  2+  29 (24%)  218 (49%)  3.55  2.09, 6.04  2.39  1.17, 4.89  Yes  81 (66%)  175 (40%)  Ref.  Ref.  Ref.  Ref.  No  41 (34%)  266 (60%)  3.03  1.96, 4.55  1.96  1.17, 3.29  Single Number of living children  Fertility Intentions  Notes: Ref. = Reference category SD = Standard Deviation  155  Table 5.4: Univariate and adjusted analyses of variables associated with contraceptive use among HIV-positive women (aged 18-44 years, currently sexually active and non-pregnant) in Soweto, South Africa (n=349) Variable  Contraceptive Use No (%) Yes (%) (n=56) (n=293)  Crude OR OR 95% CI  Adjusted OR OR 95% CI  HAART Use HAART-naive Receiving HAART  32 (57%) 24 (43%)  146 (50%) 147 (50%)  Ref. 1.34  Ref. 0.75, 2.39  Ref. 1.55  Ref. 0.84, 2.88  Age (per increase in year)  33.3 [SD=5.6]  32.9 [SD=5.3]  0.99  0.94, 1.04  0.93  0.88 0.99  Less than Grade 12  29 (52%)  194 (66%)  Ref.  Ref.  Ref.  Ref.  Grade 12 or higher  27 (48%)  98 (34%)  0.54  0.30, 0.97  0.62  0.33, 1.17  31 (55%)  172 (59%)  Ref.  Ref. --  --  --  --  --  --  Education  Employment Status Unemployed Employed  25 (45%)  121 (41%)  0.87  0.49, 1.56  Less than 3000 ZAR  44 (79%)  244 (83%)  Ref.  Ref.  3,000 or more ZAR  8 (14%)  32 (11%)  0.72  0.31, 1.67  DK  4 (7%)  17 (6%)  0.77  0.25, 2.39  Household income (per month)  Currently in a sexual relationship No  5 (9%)  23 (8%)  Ref.  Ref.  Yes  51 (91%)  270 (92%)  1.15  0.42, 3.17  156  Table 5.4 (continued): Univariate and adjusted analyses of variables associated with contraceptive use among HIV-positive women (aged 18-44 years, currently sexually active and non-pregnant) in Soweto, South Africa (n=349) Variable  Contraceptive Use No (%) Yes (%) (n=56) (n=293)  Crude OR OR 95% CI  Don’t Know  25 (45%)  126 (43%)  Ref.  Ref.  HIV-negative  8 (14%)  44 (15%)  1.09  0.46, 2.60  HIV-positive  21 (38%)  107 (37%)  1.01  0.54, 1.91  2 (4%)  16 (5%)  1.59  0.34, 7.34  0  11 (20%)  27 (9%)  Ref.  1  27 (48%)  87 (30%)  2+  18 (32%)  Yes No  Adjusted OR OR 95% CI  HIV status of regular sexual partner/husband  --  --  Ref.  Ref.  Ref.  1.31  0.58, 2.99  1.12  0.47, 2.65  179 (61%)  4.05  1.73, 9.50  3.07  1.18, 7.96  30 (54%)  80 (27%)  Ref.  Ref.  Ref.  Ref.  26 (46%)  213 (73%)  3.03  1.72, 5.56  2.22  1.15, 4.35  59.6 [SD=33.1]  59.8 [SD=35.9]  1.00  0.99, 1.01  --  --  < 200  8 (14%)  54 (19%)  Ref.  Ref. --  --  200 to < 350  22 (39%)  88 (30%)  0.59  0.25, 1.43  350 or greater  26 (46%)  148 (51%)  0.84  0.36, 1.97  Single Number of living children  Fertility Intentions  Mean # of months since HIV dx [SD] Recent CD4  157  Table 5.4 (continued): Univariate and Adjusted analyses of variables associated with contraceptive use among HIV-positive women (aged 18-44 years, currently sexually active and non-pregnant) in Soweto, South Africa (n=349) Variable  Contraceptive Use No (%) Yes (%) (n=56) (n=293)  Crude OR OR 95% CI  < 50  7 (13%)  36 (12%)  Ref.  Ref.  50 to < 200  22 (39%)  133 (46%)  1.18  0.47, 2.97  200 to < 350  15 (27%)  63 (22%)  0.82  0.31, 2.19  350 or greater  12 (21%)  58 (20%)  0.94  0.34, 2.61  55 (98%)  283 (98%)  Ref.  Ref.  1 (2%)  7 (2%)  1.36  0.17, 11.3  No  0 (0%)  16 (5%)  N/A  N/A  Yes  56 (100%)  277 (95%)  Adjusted OR OR 95% CI  Nadir CD4  --  --  --  --  --  --  WHO Stage of Disease Stage I/II Stage III/IV Disclosed HIV status to anybody  Notes: Ref. = Reference category SD = Standard Deviation N/A = Not applicable  158  110% 100%  100% 92%  Percent using contraceptive methods  90%  86%  82%  84%  80% 70%  67%  60% 50% 40% 30% 20% 10% 0% 0 months (not on HAART)  < 1 yr  1 - 2 yrs  2 - 3 yrs  3 - 4 yrs  4+ yrs  Duration of HAART use  Figure 5.1: Percent of HIV-positive women (aged 18-44 years, currently sexually active and non-pregnant) using contraception by duration of HAART use in Soweto, South Africa (n=349)  159  5.5 REFERENCES  1.  UNAIDS. Report on the global HIV/AIDS epidemic 2008. Geneva: UNAIDS; 2008.  2.  World Health Organization (WHO). Health benefits of family planning. Geneva: WHO; 1995.  3.  Petruney T, Robinson E, Reynolds H, Wilcher R, Cates W. Contraception is the best kept secret for prevention of mother-to-child HIV transmission. Bull World Health Organ 2008; 86:B.  4.  Reynolds HW, Janowitz B, Homan R, Johnson L. The value of contraception to prevent perinatal HIV transmission. Sex Transm Dis 2006; 33:350-6.  5.  Desgrees-Du-Lou A, Msellati P, Viho I, Yao A, Yapi D, Kassi P, et al. Contraceptive use, protected sexual intercourse and incidence of pregnancies among African HIV-infected women. DITRAME ANRS 049 Project, Abidjan 1995-2000. Int J STD AIDS 2002; 13:462-8.  6.  Rochat TJ, Richter LM, Doll HA, Buthelezi NP, Tomkins A, Stein A. Depression among pregnant rural South African women undergoing HIV testing. JAMA 2006; 295:1376-8.  160  7.  Laher F, Lephoto C, Forrest J, Cheyip M, L. M. The impact of parallel HAART and contraception service provision on elective terminations of pregnancy in Soweto, South Africa. Abstract #75. 4th South African AIDS Conference. 2009. Durban, South Africa.  8.  Allen S, Serufilira A, Gruber V, Kegeles S, Van de Perre P, Carael M, et al. Pregnancy and contraception use among urban Rwandan women after HIV testing and counseling. Am J Public Health 1993; 83:705-10.  9.  Balkus J, Bosire R, John-Stewart G, Mbori-Ngacha D, Schiff MA, Wamalwa D, et al. High uptake of postpartum hormonal contraception among HIV-1seropositive women in Kenya. Sex Transm Dis 2007; 34:25-9.  10.  Hoffman IF, Martinson FE, Powers KA, Chilongozi DA, Msiska ED, Kachipapa EI, et al. The year-long effect of HIV-positive test results on pregnancy intentions, contraceptive use, and pregnancy incidence among Malawian women. J Acquir Immune Defic Syndr 2008; 47:477-83.  11.  WHO. Towards universal access: scaling up priority HIV/AIDS interventions in the health sector: Progress report 2008. Geneva: World Health Organization (WHO); 2008.  12.  Hogg RS, O'Shaughnessy MV, Gataric N, Yip B, Craib K, Schechter MT, et al. Decline in deaths from AIDS due to new antiretrovirals. Lancet 1997; 349:1294.  161  13.  Palella FJ, Jr., Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998; 338:853-60.  14.  Hogg RS, Heath KV, Yip B, Craib KJ, O'Shaughnessy MV, Schechter MT, et al. Improved survival among HIV-infected individuals following initiation of antiretroviral therapy. JAMA 1998; 279:450-4.  15.  Hogg RS, Yip B, Chan KJ, Wood E, Craib KJ, O'Shaughnessy MV, et al. Rates of disease progression by baseline CD4 cell count and viral load after initiating triple-drug therapy. JAMA 2001; 286:2568-77.  16.  Cooper ER, Charurat M, Mofenson L, Hanson IC, Pitt J, Diaz C, et al. Combination antiretroviral strategies for the treatment of pregnant HIV-1infected women and prevention of perinatal HIV-1 transmission. J Acquir Immune Defic Syndr 2002; 29:484-94.  17.  Castilla J, Del Romero J, Hernando V, Marincovich B, Garcia S, Rodriguez C. Effectiveness of highly active antiretroviral therapy in reducing heterosexual transmission of HIV. J Acquir Immune Defic Syndr 2005; 40:96-101.  18.  Andia I, Kaida A, Maier M, Guzman D, Emenyonu N, Pepper L, et al. Highly active antiretroviral therapy and increased use of contraceptives among HIVpositive women during expanding access to antiretroviral therapy in Mbarara, Uganda. Am J Public Health 2009; 99:340-7. 162  19.  Kaida A, Andia I, Maier M, Strathdee SA, Bangsberg DR, Spiegel J, et al. The potential impact of antiretroviral therapy on fertility in sub-Saharan Africa. Curr HIV/AIDS Rep 2006; 3:187-94.  20.  Gruskin S, Ferguson L, O'Malley J. Ensuring sexual and reproductive health for people living with HIV: an overview of key human rights, policy and health systems issues. Reprod Health Matters 2007; 15:4-26.  21.  Baeten JM, Lavreys L, Overbaugh J. The influence of hormonal contraceptive use on HIV-1 transmission and disease progression. Clin Infect Dis 2007; 45:360-9.  22.  Coll O, Lopez M, Hernandez S. Fertility choices and management for HIVpositive women. Curr Opin HIV AIDS 2008; 3:186-92.  23.  WHO. Antiretroviral drugs for treating pregnant women and preventing HIV infection in infants: towards universal access. Recommendations for a public health approach. Geneva: WHO; 2006.  24.  Mitchell HS, Stephens E. Contraception choice for HIV positive women. Sex Transm Infect 2004; 80:167-73.  25.  Homsy J, Bunnell R, Moore D, King R, Malamba S, Nakityo R, et al. Reproductive intentions and outcomes among women on antiretroviral therapy in rural Uganda: a prospective cohort study. PLoS ONE 2009; 4:e4149.  163  26.  Department of Health (SA). National HIV and Syphilis Prevalence Survey, South Africa 2006. Pretoria: Department of Health, South Africa; 2007.  27.  Department of Health (SA). National antiretroviral treatment guidelines. Pretoria: Department of Health; 2004.  28.  United Nations (UN). World Contraceptive Use. New York: Department of Economic and Social Affairs, Population Division; 2007.  29.  Pagano M, Gauvreau K. Principles of biostatistics, 2nd edition. Pacific Grove, CA: Duxbury; 2000.  30.  Van Ness PH, Allore HG. Using SAS to investigate effect modification. Paper 10531. SUGI 31.; 2007. Available from: http://www2.sas.com/proceedings/sugi31/195-31.pdf. Accessed: October 10th, 2007.  31.  SADHS. South Africa Demographic and Health Survey 2003. Pretoria: Department of Health; 2004.  32.  Adair T. Unmet need for contraception among HIV-positive women in Lesotho and implications for mother-to-child transmission. J Biosoc Sci 2009; 41:269-78.  164  33.  Peltzer K, Chao LW, Dana P. Family planning among HIV positive and negative prevention of mother to child transmission (PMTCT) clients in a resource poor setting in South Africa. AIDS Behav 2009; 13:973-9.  34.  Williams HA, Watkins CE, Risby JA. Reproductive decision-making and determinants of contraceptive use in HIV-infected women. Clin Obstet Gynecol 1996; 39:333-43.  35.  Seiber EE, Bertrand JT, Sullivan TM. Changes in contraceptive method mix in developing countries. Int Fam Plan Perspect 2007; 33:117-23.  36.  Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents.: Department of Health and Human Services; 2008.  37.  Cooper D, Harries J, Myer L, Orner P, Bracken H, Zweigenthal V. "Life is still going on": reproductive intentions among HIV-positive women and men in South Africa. Soc Sci Med 2007; 65:274-83.  38.  Myer L, Morroni C, Cooper D. Community attitudes towards sexual activity and childbearing by HIV-positive people in South Africa. AIDS Care 2006; 18:772-6.  39.  Laher F, Todd CS, Stibich MA, Phofa R, Behane X, Mohapi L, et al. A Qualitative Assessment of Decisions Affecting Contraceptive Utilization and  165  Fertility Intentions among HIV-Positive Women in Soweto, South Africa. AIDS Behav 2009.  166  CHAPTER 65: THE WHOMEN’S SCALE (WOMEN’S HAART OPTIMISM MONITORING AND EVALUATION SCALE V.1) AND THE ASSOCIATION WITH FERTILITY INTENTIONS AND SEXUAL BEHAVIOURS AMONG HIV-POSITIVE WOMEN IN UGANDA  6.1 INTRODUCTION  The introduction of highly active antiretroviral therapy (HAART) in 1996 heralded a new era of hope for people living with HIV/AIDS. Since then, a growing body of evidence has revealed that by suppressing HIV viral load, HAART has dramatically increased survival and decreased HIV/AIDS-related morbidity of individuals in both resource-rich [1, 2, 3] and resource-limited settings [4, 5, 6, 7].  Alongside the remarkable clinical benefits, early reports speculated that the availability of HAART would influence perceptions of HIV risk and sexual risk behaviour of both HIV-infected and uninfected individuals, thereby creating a potential for increased transmission of HIV and other sexually transmitted infections  5  A version of this chapter has been published. Kaida A, Lima VD, Andia I, Kabakyenga J, Mbabazi P, Emenyonu N, Patterson TL, Hogg RS, Bangsberg DR. The WHOMEN’s scale (Women’s HAART Optimism Monitoring and EvaluatioN scale v.1) and the association with fertility intentions and sexual behaviours among HIV-positive women in Uganda. AIDS and Behavior 2009; Suppl 1:72-81.  167  [8, 9]. These perceptions, termed variably as ‘HAART optimism’ or ‘HIV treatment optimism’, reflected individuals’ optimism about the use and efficacy of HAART and corresponding beliefs concerning the decreased need to engage in safer sexual behaviours [10]. These beliefs include two different but related aspects; namely, a belief that HIV-infected individuals on HAART are less likely to transmit HIV due to reduced viral loads and a belief that HIV infection is less of a threat to health and survival due to HAART availability [11, 12].  The concept of HAART optimism is important to HIV prevention efforts since higher levels have commonly been associated with lower rates of condom use, primarily among men who have sex with men (MSM) [13, 14, 15, 16, 17, 18]. While the association has not been entirely consistent [19], a meta-analysis reported that individuals with high levels of HAART optimism are significantly more likely to engage in unprotected sexual behavior, irrespective of their HIV status and actual HAART use status [20]. Based on the cross-sectional nature of these studies, however, we cannot conclude that HAART optimism is the causal factor for the risk, and not vice versa as shown elsewhere [21].  An important limitation of the published literature on HAART optimism to date is its near exclusive focus on sexual risk behavior among MSM populations in western country settings. No studies are available regarding the role of HAART  168  optimism in HIV endemic countries, owing in part to the recency of widespread HAART availability in these settings [22]. Limited findings are available concerning the prevalence and role of HAART optimism among women [23, 24], despite the fact that women comprise half of all people living with HIV/AIDS around the world [25]. Moreover, limited research has been conducted on the influence of HAART optimism on behaviours other than sexual risk-taking [10]. In particular, unlike MSM populations, HAART optimism may influence reproductive behaviors of HIVpositive women through changes in concerns about vertical transmission [26].  Uganda is an ideal setting to investigate issues related to women’s HAART optimism since it is a country currently experiencing a generalized HIV epidemic, fueled primarily by heterosexual transmission [25]. HIV prevalence among adults aged 15-49 years in Uganda is estimated to be 6.3% and is higher among women (7.5%) than men (5.0%) [27]. HAART scale-up efforts in Uganda began in earnest in 2004 and recent government statistics estimate that approximately 120,000 Ugandans are currently on treatment, accounting for slightly more than one-third of those who need it [22]. At 6.6, Uganda has the second highest total fertility rate in Africa [28] and recent findings report that HIV-positive women on HAART are three times more likely to report fertility desires, compared with HIV-positive women not on HAART [29]. Previous research has demonstrated that self-reported fertility desires and intentions are strong predictors of future fertility [30, 31, 32, 33].  169  The purpose of this study was to develop a reliable HAART optimism scale for HIV-positive women in high prevalence settings and to test the scale’s validity against a measure of fertility intentions and two measures of sexual behaviour (i.e., sexual abstinence and unprotected sex) among a sample of HAART-naïve and HAART-experienced women in Uganda.  6.2 METHODS  6.2.1 Study design  This analysis is based on data collected from a cross-sectional study of 540 HIV-positive women aged 18-50 years attending an HIV clinic in Mbarara, western Uganda. 6.2.2 Study setting  The study setting has been described in detail elsewhere [34]. Briefly, Mbarara Regional Referral Hospital (MRRH) is located in Mbarara district, in southwestern Uganda. Housed within the MRRH is the Mbarara University HIV treatment centre, the Immune Suppression Syndrome (ISS) clinic. The ISS clinic opened in 1998 and currently serves over 13,000 clients, 65% of whom are women.  170  The clinic offers comprehensive HIV care services, including HAART, free-ofcharge. Approximately 35% of the clinic’s population is currently receiving HAART.  6.2.3 Eligibility criteria  HIV-positive women aged 18-50 years, attending the ISS clinic, and competent to give consent were eligible to participate in the study.  6.2.4 Data collection  Women attending the ISS clinic were consecutively approached by a research assistant to assess eligibility and interest prior to requesting consent for participation in the study. Since non-HAART users attend the clinic less frequently than HAART users, non-HAART users were over-sampled.  Upon determining eligibility and consent, participants received a 20-30 minute interviewer-administered structured questionnaire in either English or Runyankole (the dominant regional language). The Runyankole questionnaire was translated from the original English version into Runyankole, and then backtranslated into English to ensure consistency between the two versions.  The  questionnaire was piloted among 40 women and was administered either before or  171  after the patient’s clinical encounter. Approximately 15 women were interviewed daily by three trained research assistants over eight months from November 2005 to June 2006.  The questionnaire assessed socio-demographic characteristics, HIV/AIDS history, diagnosis, and treatment, reproductive and sexual decision-making and behaviours, and HAART optimism. A medical record review was conducted to confirm HAART history and to obtain clinical data including WHO stage of disease [35] and CD4 cell count. The medical record was considered the referent measure for inconsistencies between self-reported and medical record data.  6.2.5 Instrument Development  The study questionnaire included 23 statements related to HIV and treatment with HAART, concern about horizontal and vertical HIV transmission, and concern about severity of HIV disease. Eleven statements were taken from Van de Ven et al’s scale (2000), an existing and widely used HAART optimism scale validated for use among MSM populations [36]. While the Van de Ven scale is comprised of 12 statements, we excluded one item from our questionnaire because it was not relevant to our setting (i.e., “The availability of treatment (PEP) immediately after unsafe sex makes safe sex less important”). The remaining 11 statements were 172  subjected to focus group testing with ten Ugandan women to assess the interpretability of the scale statements in this setting. The statements were translated and back-translated from English to Runyankole, and were checked for cultural appropriateness and understanding by women of varying levels of formal education. After focus group testing, the language used for each statement from Van de Ven’s scale was changed slightly to render it more understandable by women with lower levels of education. For instance, the term “undetectable viral load” was replaced with “a very small amount of virus”. An additional 12 statements were developed from a literature review and from the results of related qualitative interviews with Ugandan women and three clinic staff members. These interviews were conducted to assess content and face validity.  The final list included 23 statements and incorporated a four-point response format. Respondents were asked to indicate how much they agreed or disagreed with each of the 23 statements. Responses to each statement were scored from 1 (Strongly Disagree) to 4 (Strongly Agree), with some questions requiring inversion. The complete list of statements is shown in Table 6.3.  173  Measuring fertility intentions and sexual behaviour  In order to assess the potential influence of HAART optimism on fertility intentions and sexual behaviours, the following three measures were included in the questionnaire.  The first outcome of interest was fertility intention. Fertility intention was assessed by answers to the question “Are you planning to have any (more) children in the future?’’  The second outcome of interest was sexual activity/sexual abstinence in the previous three months. This variable was specific to vaginal intercourse.  The third outcome of interest was unprotected sexual activity (i.e., vaginal intercourse without a condom) in the previous three months. This analysis was restricted to women who reported being sexually active in the previous three months.  6.2.6 Data analysis  Univariate analyses of the baseline characteristics of the study population was conducted, including socio-demographic and clinical variables. Categorical  174  variables are presented as frequencies and percents while continuous variables are presented as medians and interquartile ranges [IQRs].  Factor analysis  The entire dataset was subjected to a principal components and factor analysis to identify which of the 23 statements best expressed the construct of HAART optimism. The principal components analysis served as a data reduction method (i.e., to reduce the number of variables in the scale) [37].  The principal components were extracted via variance maximizing (i.e., varimax) whereby the lines on which the variance was maximal was identified, with each line representing a factor. Because each consecutive factor was defined to maximize the variability not captured by the preceding factor, consecutive factors were independent of each other. However, each consecutive factor accounted for less and less variability. Extraction of factors was stopped when there was only very little random variability left. The assumptions of varimax rotation were assessed using a non-orthogonal rotation (i.e., oblimin) [37]. Items with loadings above 0.40, with no or few item cross-loadings, and no factors with fewer than three items were retained.  A correlation matrix was constructed where the variances of all items were equal to 1.0. In this way, the total variance in the matrix was equal to the number of  175  items (i.e., 23). Eigenvalues were calculated, representing the variance explained for each of the 23 items considered in the factor analysis. The eigenvalues were then expressed as a percent of the total variance. To select factors, the Kaiser criterion was applied, which states that only factors with eigenvalues greater than one are retained [38]. As the Kaiser criterion is a less conservative method for retaining factors, the scree test was also used to determine how many factors to maintain [39]. The scree test is a graphical method used to plot the eigenvalues. The point on the plot where the smooth decrease of the eigenvalues appears to level off was located. The “factorial scree” is found to the right of this plot, meaning that the primary factors are found to the left of that point. The information from both the Kaiser criterion and the scree test was combined and factor interpretability was also considered in indentifying which factors to maintain.  After identifying the primary factors, the squared multiple correlation of an item with all other items was used to estimate the communalities for each variable in the factor [37].  Cronbach’s alpha (standardized) is reported to demonstrate the internal reliability of each of the identified factors. While a scale is generally considered reliable if the Cronbach’s alpha coefficient is equal to or greater than 0.70 [40], for exploratory studies such as this one, a coefficient of ≥ 0.60 is considered acceptable  176  [41]. In addition, Pearson’s correlation coefficient was used to calculate an itemspecific correlation for each scale. A correlation exceeding 0.40 was considered satisfactory correlation of items within the scale [42]. We expected reliability coefficients (i.e., Cronbach’s alpha) to exceed inter-scale correlations for the same scale.  For the factor of primary interest identified from this analysis (i.e., that identifying “HAART optimism”), the mean, standard deviation, median, and interquartile range of the scores were calculated for descriptive purposes.  Testing the association between the derived HAART optimism scale and fertility intentions and sexual behaviour outcomes  To assess scale validity, the association between the derived HAART optimism scale and fertility intentions, sexual activity, and unprotected sexual intercourse was tested using the Wilcoxon rank sum test. The associations are reported using a crude odds ratio (OR) with a 95% confidence interval (95% CI).  177  6.2.7 Ethical considerations  All participants provided informed consent and all procedures were approved by the Faculty of Medicine Research and Ethics Committee and the Institutional Ethics Review Board of Mbarara University, the Uganda National Council on Science and Technology, and the University of California, San Francisco Committee on Human Subjects. Study investigators only had access to de-identified data.  6.3 RESULTS  6.3.1 Recruitment A total of 540 women completed the questionnaire. Of these, 49% were HAART users with a median duration of HAART use of 15 months [IQR: 11-25 months].  6.3.2 Baseline characteristics Distributions of baseline characteristics are presented in Table 6.1. As shown, the median age was 34 years [29-39 years]. Overall, the majority of women had a primary school education or less (66%), belonged to the Kiga/Nkole tribe (81%), and  178  were Christian (88%). Only 39% of the women were married, 30% were widowed, 23% were separated or divorced, and 8% were single. Two-thirds (66%) had a monthly household income less than 80,000 UGX (approximately $50 USD). The median number of lifetime livebirths was 3.0 [IQR: 2-5]. Most of the women had advanced HIV disease with 62% presenting in WHO Disease Stage III or IV and 35% reporting a most recent CD4 cell count less than 200 cells/mm3 (of those with nonmissing responses).  6.3.3 Factor analysis The eigenvalues of the correlation matrix for all 23 statements are shown in Table 6.2. The Kaiser criterion results revealed that seven factors (with eigenvalues greater than one) accounted for 55% of the cumulative variance. Figure 6.1 plots the results of the scree test. As shown, the factorial scree occurs at the third factor indicating that the analysis identified three primary and distinct factors. We used varimax rotation to assess the change in grouping of the loadings in each factor. The factorial scree identified factors that were each highly interpretable based on the contributing items. The variance accounted for by these three factors was 34%.  Based on the eight statements that comprised factor 1, the factor was labeled “HAART optimism”. Included in this factor were statements about HAART-related  179  optimism/skepticism regarding HIV transmission to sexual partners and to unborn children through vertical transmission. The second factor comprised four items and was labeled “State of the epidemic”. This factor included items related to changing opinions about the course of the HIV epidemic. The third factor was labeled “Trust of Information” about the epidemic from health care professionals and was made up of four items. Together, the communality between these three factors was 77%, with the first factor accounting for 32% of the communality, the second factor accounting for 25% and the third factor accounting for 20%. The total communality explained by these three factors did not change after varimax or oblique rotation.  The first factor, that identifying ‘HAART optimism’, was of primary interest to this study. This factor included eight statements (shown in Table 6.3), none of which were considered duplicates of one another. This factor, referred to as the derived HAART optimism scale, had moderately high internal consistency, with Cronbach’s alpha = 0.70 in the overall sample. The item-total correlations ranged from 0.33 to 0.50 for the 8-item analysis (Table 6.4).  In the overall sample, the derived 8-item HAART optimism scale had a mean of 11.7 [SD=4.2] and a median of 11 [IQR: 8-13], with a possible range from 8 (indicating high skepticism) to 32 (indicating high optimism).  180  6.3.4 Association between the derived HAART optimism scale and fertility intentions Overall, 14% of women reported that they intended to have (more) children. Women who intended to have (more) children had significantly higher HAART optimism scores (median=13.5 [IQR:12-16]) than women who did not intend to have (more) children (median=10.5 [IQR:8-12]) (p<0.0001).  6.3.5 Association between the derived HAART optimism scale and sexual activity in the previous three months We found that overall, 45% of women reported being sexually active in the previous three months. Women who reported being sexually active in the previous three months had significantly higher HAART optimism scores (median=11 [IQR: 814]) than women who were sexually abstinent (median=11 [IQR: 8-13]) (p=0.0206).  6.3.6 Association between the derived HAART optimism scale and unprotected sexual activity in the previous three months Of women who reported being sexually active in the previous 3 months (n=241), 49% reported practicing unprotected sex. Women who reported practicing  181  unprotected sex had significantly higher HAART optimism scores (median=12 [IQR: 9-15]) than women who practiced protected sex (median=11 [IQR:8-13]) (p=0.0157).  6.4 DISCUSSION  This study developed a culturally appropriate eight-item HAART optimism scale that demonstrated moderately high reliability when applied to HIV-positive women in Uganda. The new scale incorporates concerns related to pregnancy and vertical transmission, in addition to those related to HIV transmission risk to sexual partners and severity of HIV disease. We have termed this new scale the Women’s HAART Optimism Monitoring and EvaluatioN scale, or the WHOMEN’s scale, version 1.  As evidenced from the new scale, overall women expressed low HAART optimism with 75% reporting a HAART optimism score of 13 or lower, out of a possible range from eight (highly skeptical) to 32 (highly optimistic). This suggests that women are still very cautious about the use and effectiveness of HAART on lowering risk of HIV transmission and believing that HIV infection is less of a threat to their health and survival. This finding is consistent with other research which suggests that individuals with high HAART optimism tend to be a minority within the population under study [36, 12, 10, 24, 43]. 182  There is evidence that the brief and reliable WHOMEN’s scale has validity since women who report an intention to have (more) children have significantly higher HAART optimism than their counterparts who do not intend to have (more) children. Similarly, women who are currently sexually active and who practice unprotected sex have significantly higher HAART optimism scores than their counterparts who are sexually abstinent and who practice protect sex. It should be noted, however, that despite statistically significant differences between groups, the absolute difference in HAART optimism scores between groups was small.  Availability of HAART in Uganda has been relatively recent and it is unclear whether HAART optimism will change over time as HAART becomes more widely available. While a trend of increasing HAART optimism over time has not been noted in MSM populations [44], it remains to be seen whether the same trend will occur for women in terms of altering their reproductive intentions and sexual behaviour.  Fertility intentions and sexual behavior among HIV-positive women are, of course, mediated by a number of factors beyond simply HAART optimism [45, 24, 46]. From qualitative interviews with heterosexual women living with HIV and on HAART in Brazil, Kerrigan et al (2006) report that none of the women connected their recent sexual behaviors to HAART optimism [24]. Interestingly, heterosexual  183  men in the same study tended to report fertility intentions as a reason for not wanting to use condoms. No peer-reviewed studies were identified which have investigated gender differentials in HAART optimism and how this might influence reproductive and sexual behaviours. This type of research, explored in a longitudinal fashion, appears warranted.  It remains to be determined whether these findings are generalizable to women in other high HIV prevalence settings. Future studies should test the reliability and validity of the scale in other settings and construct multivariate analyses to determine the relative importance of HAART optimism vs. other factors (such as HAART use) in influencing fertility desires [29].  The limitations of this study must be acknowledged. First, the cross-sectional nature of this analysis precludes us from determining causality. The direction of causality should be interpreted with particular caution since it has been suggested that high HAART optimism may follow risky sexual behaviors, rather than precede them [21]. This finding may be of particular importance to women and fertility intentions since those who intend to have (more) children may be more likely to be optimistic about HAART. Despite the cross-sectional association, our findings offer important information on the association between HAART optimism and fertility intentions and sexual behaviours that may have important implications for HIV  184  prevention, family planning, and PMTCT programming, particularly as HAART access expands in high HIV prevalence settings. Second, there is a risk of social desirability bias whereby women may falsely report their optimism about HAART because they have been educated about HIV transmission and counseled to avoid pregnancy. If false reporting was differential, with those reporting fertility intentions, for instance, more likely to report falsely report HAART skepticism, then our effect estimates are likely somewhat inflated. During the data collection process we took precautions against reporting bias by including standardized questions and using non-clinic staff to conduct the interviews. Third, the associations provided here are crude. Future studies should investigate the reliability and validity of the WHOMEN’s scale on different data sources and conduct multivariate analyses to assess the role of potential confounders in mediating the association between HAART optimism and women’s fertility intentions and sexual behaviours.  Despite these limitations, our brief, reliable, and valid scale, termed the Women’s HAART Optimism Monitoring and EvaluatioN scale (WHOMEN’s scale), may be valuable to broader studies investigating the role of HAART optimism on reproductive intentions and sexual behaviours of HIV-positive women in high HIV prevalence settings.  185  Table 6.1: Characteristics of study population of HIVpositive women in Mbarara, Uganda (n=540) Variable Median Age [IQR] Education Primary school or less Secondary school and higher Tribe Kiga/Nkole Other Religion Christian Other Marital Status Currently Married Divorced/Separated Widowed Single Household Income (per month)§ 0 – 20,000 UGX 20,001 – 80,000 UGX 80,001+ UGX Median Lifetime livebirths [IQR] Most recent CD4 Count < 50 50-199 200-499 500+ Missing WHO Stage¥ Stage 1 or 2 Stage 3 or 4 Current HAART use Yes No  N 34 [29-39]  %  356 184  66% 34%  438 102  81% 19%  475 65  88% 12%  209 125 164 42  39% 23% 30% 8%  166 178 181 3.0 [2.0-5.0]  32% 34% 34%  45 91 147 63 155  8% 17% 27% 12% 29%  188 313  38% 62%  263 277  49% 51%  Notes: SD = Standard Deviation. IQR = Interquartile Range. § 3% of participants did not report household income. ¥ 7% of participants did not have information on WHO Stage of Disease.  186  Table 6.2: Principal component Factor analysis of 23 statements about HAART Factor 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23  Eigenvalue  Difference  3.37 2.62 1.74 1.38 1.30 1.18 1.06 0.99 0.91 0.84 0.79 0.78 0.74 0.72 0.66 0.60 0.56 0.55 0.53 0.51 0.48 0.39 0.29  0.75 0.87 0.36 0.08 0.11 0.12 0.07 0.08 0.07 0.05 0.02 0.03 0.02 0.07 0.06 0.04 0.01 0.02 0.02 0.03 0.08 0.10  % of Total Variance 14.6 11.4 7.6 6.0 5.6 5.1 4.6 4.3 4.0 3.7 3.5 3.4 3.2 3.2 2.9 2.6 2.4 2.4 2.3 2.2 2.1 1.7 1.3  % Cumulative Variance 14.6 26.0 33.6 40.0 45.2 50.4 55.0 59.3 63.3 67.0 70.4 73. 8 77. 0 80.1 83.0 85.6 88.0 90.4 92.8 95.0 97.0 98.7 1.00  187  Table 6.3 Items with Principal Components Varimax Factor Loadings Item Number 21 13 19 1  7 6  15 22 10 9 14 3 18  4 17 2 5 8 11  Item  It would be OK to stop HIV medicines if they made me sick, especially if I were pregnant. I believe that new drug therapies make people with HIV less able to pass the virus to other people. If my partner and I were both on HIV treatments, it would be OK to stop using condoms. A person with virus that has been turned off in the body (a very small amount of virus in the body) cannot pass the virus to someone else People with a very small amount of virus in the body don’t need to worry about infecting others with HIV. If a cure for AIDS were announced, I would stop practicing safe sex (abstinence, being faithful, or using a condom). I am not worried about becoming pregnant because of available HIV treatments. Now that HIV medicines are available, I want to have a baby because I will live long enough to raise a child. HIV/AIDS is a less serious threat than it used to be because of new treatments. HIV is less of a threat because the epidemic is on the decline. I am less worried about HIV infection now that treatments have improved. I’m less worried about HIV infection than I used to be. I think the risks of HIV transmission with “safer sex” (using abstinence, being faithful, and using condoms) are less than what doctors say. New HIV treatments will take the worry out of sex. I think the risks of HIV transmission in pregnancy are less than what doctors say. A person’s CD4 count will remain about 200 (at a safe level) if they stay on therapy. If every HIV positive person took the new treatments, the AIDS epidemic would be over. Until there is a complete cure for HIV/AIDS, prevention is still the best practice. (reverse scored) It’s never safe to have sex without a condom regardless of the amount of virus in the body. (reverse scored)  Factor 1 (HAART optimism) 0.700  Factor 2  Factor 3  0.057  -0.033  0.697  0.100  0.054  0.603  0.121  0.074  0.543  -0.030  -0.077  0.499  -0.034  -0.195  0.491  -0.006  0.118  0.460  0.036  -0.120  0.448  -0.058  0.062  0.102  0.783  0.218  0.196  0.722  0.095  0.002  0.720  0.119  -0.198 0.013  0.570 0.129  -0.181 0.751  0.318  0.087  -0.691  0.206  0.136  0.645  -0.028  0.056  0.463  0.231  0.252  0.076  0.301  -0.382  0.053  0.247  -0.243  0.143  188  12 16  20  23  Because of new treatments, fewer people are becoming infected with HIV. If I become pregnant, I would be worried about passing HIV on to my baby, even if I were taking HIV treatments. (reverse scored). HIV treatments/medicines could be harmful during pregnancy because of possible birth defects. (reverse scored). I am worried that HIV medicines could cause AIDS. (reverse scored)  0.316  0.369  0.253  0.104  0.112  0.017  -0.247  -0.028  0.257  -0.273  -0.038  0.081  189  Table 6.4: Statements constituting the derived HAART optimism scale for use among HIV-positive women of reproductive age living in developing countries Question: Considering the new medicines available to treat HIV, please tell me how much you agree or disagree with the statements I read to you. Do you Strongly Agree, Agree, Disagree, or Strongly Disagree with the statement? Statement It would be OK to stop HIV medicines if they made me sick, especially if I were pregnant. (inverted) I believe that new drug therapies make people with HIV less able to pass the virus to other people. If my partner and I were both on HIV treatments, it would be OK to stop using condoms. A person with virus that has been turned off in the body (a very small amount of virus in the body) cannot pass the virus to someone else. People with a very small amount of virus in the body don’t need to worry about infecting others with HIV. If a cure for AIDS were announced, I would stop practicing safe sex (abstinence, being faithful, or using a condom). I am not worried about becoming pregnant because of available HIV treatments. Now that HIV medicines are available, I want to have a baby because I will live long enough to raise a child.  Corrected item-total correlation 0.53 0.50 0.45 0.36 0.34 0.33 0.35 0.35  190  4  3.5  Eigenvalue Value  3  2.5 Factorial scree  2  1.5 1  0.5 0 1  2  3  4  5  6  7  8  9  10 11 12 13 14 15 16 17 18 19 20 21 22 23 Eigenvalue Number  Figure 6.1: Results of the scree test  6.5 REFERENCES  1. Hogg RS, Heath KV, Yip B, Craib KJ, O'Shaughnessy MV, Schechter MT, et al. Improved survival among HIV-infected individuals following initiation of antiretroviral therapy. JAMA 1998; 279: 450-4.  2. Mocroft A, Vella S, Benfield TL, Chiesi A, Miller V, Gargalianos P, et al. Changing patterns of mortality across Europe in patients infected with HIV-1. EuroSIDA Study Group. Lancet 1998; 352: 1725-30.  3. Palella FJ Jr., Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998; 338: 853-60.  4. Seyler C, Anglaret X, Dakoury-Dogbo N, Messou E, Toure S, Danel C, et al. Medium-term survival, morbidity and immunovirological evolution in HIVinfected adults receiving antiretroviral therapy, Abidjan, Cote d'Ivoire. Antivir Ther 2003; 8: 385-93.  5. Braitstein P, Brinkhof MW, Dabis F, Schechter M, Boulle A, Miotti P, et al. Mortality of HIV-1-infected patients in the first year of antiretroviral therapy: comparison between low-income and high-income countries. Lancet 2006; 367: 817-24.  192  6. Spacek LA, Shihab HM, Kamya MR, Mwesigire D, Ronald A, Mayanja H, et al. Response to antiretroviral therapy in HIV-infected patients attending a public, urban clinic in Kampala, Uganda. Clin Infect Dis 2006; 42: 252-9.  7.  Wools-Kaloustian K, Kimaiyo S, Diero L, Siika A, Sidle J, Yiannoutsos CT, et al. Viability and effectiveness of large-scale HIV treatment initiatives in sub-Saharan Africa: experience from western Kenya. AIDS 2006; 20: 41-8.  8. Dilley JW, Woods WJ, McFarland W. Are advances in treatment changing views about high-risk sex? N Engl J Med 1997; 337: 501-2.  9. Kelly JA, Otto-Salaj LL, Sikkema KJ, Pinkerton SD, Bloom, FR. Implications of HIV treatment advances for behavioral research on AIDS: protease inhibitors and new challenges in HIV secondary prevention. Health Psychol 1998; 17: 310-9.  10. International Collaboration on HIV Optimism. HIV treatments optimism among gay men: an international perspective. J Acquir Immune Defic Syndr 2003; 32: 54550.  11. Remien RH, Smith RA. HIV prevention in the era of HAART: implications for providers. AIDS Read 2000; 10: 247-51.  12. Elford J, Bolding G, Sherr L. High-risk sexual behaviour increases among London gay men between 1998 and 2001: what is the role of HIV optimism? AIDS 2002; 16: 1537-44.  193  13. Kalichman SC, Nachimson D, Cherry C, Williams E. AIDS treatment advances and behavioral prevention setbacks: preliminary assessment of reduced perceived threat of HIV-AIDS. Health Psychol 1998; 17: 546-50.  14. Kravcik S, Victor G, Houston S, Sutherland D, Garber GE, Hawley-Foss N, et al. Effect of antiretroviral therapy and viral load on the perceived risk of HIV transmission and the need for safer sexual practices. J Acquir Immune Defic Syndr Hum Retrovirol 1998; 19: 124-29.  15. Van de Ven P, Kippax S, Knox S, Prestage G, Crawford J. HIV treatments optimism and sexual behaviour among gay men in Sydney and Melbourne. AIDS 1999; 13: 2289-94.  16. Vanable PA, Ostrow DG, McKirnan DJ, Taywaditep KJ, Hope BA. Impact of combination therapies on HIV risk perceptions and sexual risk among HIVpositive and HIV-negative gay and bisexual men. Health Psychol 2000; 19: 134-45.  17. Ostrow DE, Fox KJ, Chmiel JS, Silvestre A, Visscher BR, Vanable PA, et al. Attitudes towards highly active antiretroviral therapy are associated with sexual risk taking among HIV-infected and uninfected homosexual men. Aids 2002; 16: 775-80.  18. Van de Ven P, Rawstorne P, Nakamura T, Crawford J, Kippax S. HIV treatments optimism is associated with unprotected anal intercourse with regular and with casual partners among Australian gay and homosexually active men. Int J STD AIDS 2002; 13: 181-83. 194  19. Elford J, Bolding G, Maguire M, Sherr L. Combination therapies for HIV and sexual risk behavior among gay men. J Acquir Immune Defic Syndr 2000; 23: 26671.  20. Crepaz N, Hart TA, Marks G. Highly active antiretroviral therapy and sexual risk behavior: a meta-analytic review. JAMA 2004; 292: 224-36.  21. Huebner DM, Rebchook GM, Kegeles SM. A longitudinal study of the association between treatment optimism and sexual risk behavior in young adult gay and bisexual men. J Acquir Immune Defic Syndr 2004; 37: 1514-9.  22. World Health Organization (WHO). Towards universal access: Scaling up priority HIV/AIDS interventions in the health sector: Progress report 2008. Geneva: WHO; 2008.  23. van der Straten A, Gomez CA, Saul J, Quan J, N Padian. Sexual risk behaviors among heterosexual HIV serodiscordant couples in the era of post-exposure prevention and viral suppressive therapy. AIDS 2000; 14: F47-54.  24. Kerrigan D, Bastos FI, Malta M, Carneiro-da-Cunha C, Pilotto JH, Strathdee SA. The search for social validation and the sexual behavior of people living with HIV in Rio de Janeiro, Brazil: understanding the role of treatment optimism in context. Soc Sci Med 2006; 62: 2386-96.  25. UNAIDS. AIDS epidemic update. December 2007. Geneva: UNAIDS; 2007.  195  26. Kirshenbaum SB, Hirky AE, Correale J, Goldstein RB, Johnson MO, RotheramBorus MJ, et al. "Throwing the dice": pregnancy decision-making among HIVpositive women in four U.S. cities. Perspect Sex Reprod Health 2004; 36: 106-13.  27. UNGASS. UNGASS country progress report Uganda, January 2006 to December 2007 (accessible through UNAIDS Uganda country report). Government of Uganda; 2008.  28. World Health Organization (WHO). Core health indicators. Geneva: WHO; 2006.  29. Maier M, Andia I, Emenyonu N, Guzman D, Kaida A, Pepper L, et al. Antiretroviral Therapy is Associated with Increased Fertility Desire, but not Pregnancy or Live Birth, among HIV+ Women in an Early HIV Treatment Program in Rural Uganda. AIDS Behav 2008; 13 (Suppl 1): 28-37.  30. Bongaarts J. The measurement of wanted fertility. Popul Dev Rev 1990; 16: 487506.  31. Pritchett LH. Desired fertility and the impact of population policies. Popul Dev Rev 1994; 20: 1-55.  32. Tan PC, Tey NP. Do fertility intentions predict subsequent behavior? Evidence from Peninsular Malaysia. Stud Fam Plann 1994; 25: 222-31.  196  33. Miller WB. A framework for modelling fertility motivation in couples. Population Studies 2004; 58: 193-205.  34. Andia I, Kaida A, Maier M, Guzman D, Emenyonu N, Pepper L, et al. Highly active antiretroviral therapy and increased use of contraceptives among HIVpositive women during expanding access to antiretroviral therapy in Mbarara, Uganda. Am J Public Health 2009; 99: 340-7.  35. Scaling up antiretroviral therapy in resource-limited settings: treatment guidelines for a public health approach. 2003 Revisions. Geneva, Switzerland: WHO; 2004.  36. Van de Ven P, Crawford J, Kippax S, Knox S, Prestage G. A scale of optimismskepticism in the context of HIV treatments. AIDS Care 2000; 12: 171-6.  37. StatSoft. Principal components and factor analysis. Available from: www.statsoft.com/textbook/stfacan.htm. Accessed: October 1st, 2007.  38. Kaiser HF. The application of electronic computers to factor analysis. Educ Psychol Meas 1960; 20: 141-151.  39. Kline P. An easy guide to factor analysis. New York: Routledge; 1994.  40. Nunnally JC, Bernstein IR. Psychometric Theory, 3rd ed. New York, McGraw-Hill; 1994.  197  41. Bland, JM, Altman DG. Cronbach's alpha. BMJ 1997; 314: 572.  42. Ware JE Jr., and Gandek B. Methods for testing data quality, scaling assumptions, and reliability: the IQOLA Project approach. International Quality of Life Assessment. J Clin Epidemiol 1998; 51: 945-52.  43. Sullivan PS, Drake AJ and Sanchez TH. Prevalence of treatment optimism-related risk behavior and associated factors among men who have sex with men in 11 states, 2000-2001. AIDS Behav 2007; 11: 123-9.  44. Elford J, Bolding G, Sherr L. HIV optimism: fact or fiction? Focus 2001; 16: 1-4.  45. Bedimo AL, Bessinger R, Kissinger P. Reproductive choices among HIV-positive women. Soc Sci Med 1998; 46: 171-9.  46. Craft SM, Delaney RO, Bautista DT, Serovich JM. Pregnancy decisions among women with HIV. AIDS Behav 2007; 11: 927-35.  198  CHAPTER 7: SUMMARY, CONTRIBUTIONS, RECOMMENDATIONS, FUTURE RESEARCH, AND CONCLUSION  7.1 Summary of Objectives  The objectives of this thesis were: (1) to review the literature describing the potential impact of highly active antiretroviral therapy (HAART) on fertility in subSaharan Africa; (2) to assess whether the prevalence of recent sexual intercourse (and sexual abstinence) varied according to use of HAART among HIV-positive women of reproductive age from Brazil, South Africa, and Uganda; (3) to assess whether the prevalence of fertility intentions varied according to HIV status and use and duration of HAART among women in Soweto, South Africa; (4) to assess whether the prevalence of contraceptive use and method mix varied by HIV status and use and duration of HAART among women in Soweto, South Africa; and (5) to develop a reliable HAART optimism scale for use among HIV-positive women and to test the scale’s validity against measures of fertility intention, sexual activity, and unprotected sexual intercourse among HIV-positive women in Mbarara, Uganda. All analyses were conducted within an over-arching framework that recognizes and supports the sexual and reproductive rights of people living with HIV [1].  199  7.2 Summary of Findings  My PhD dissertation research makes an important and lasting contribution to the emerging literature regarding the impact of expanding access to HAART on sexual and reproductive health decision-making and behaviours of HIV-infected and -affected women in high HIV prevalence settings in sub-Saharan Africa.  At the beginning of this thesis, I conducted a review of the available literature pertaining to the known and potential impacts of expanding access to HAART on the fertility of HIV-positive women in sub-Saharan Africa. As presented in Chapter 2 (and published in Current HIV/AIDS Reports in 2006) [2], the review revealed that while a plethora of studies have examined the impact of HIV infection on fertility, there were limited empirical findings regarding the impact of increased HAART availability. Thus, the review was extended to use Bongaarts’ proximate determinants of fertility framework (adapted for conditions of generalized HIV epidemics) [3, 4] to examine how use of HAART may mediate the impact of HIV infection on fertility through its biological and behavioural proximate determinants. Based on this examination, a conceptual framework (Figure 2.1) was proposed to guide hypothesis generation and subsequent investigations aimed at understanding how expanding access to HAART may impact the proximate determinants and ultimately fertility of HIV-positive women in sub-Saharan Africa. This published  200  framework formed the basis of the theoretical paradigm that I used to frame the objectives and hypotheses in the following chapters of my dissertation.  Chapters 3 to 6 examine relationships between receipt of HAART and several key determinants of fertility as outlined and prioritized in the conceptual framework including, sexual activity (Chapter 3), fertility intentions (Chapter 4), contraceptive use and method mix patterns (Chapter 5), and HAART optimism (Chapter 6).  Chapter 3, published in AIDS Care in 2008 [5], presents the results of a pilot study which examined the prevalence of recent sexual activity among cohorts of HIV-positive women in three global settings (Rio de Janeiro, Brazil; Soweto, South Africa; and Mbarara, Uganda) and whether sexual activity varied by current HAART use. The findings indicated that overall less than half (46%) of HIV-positive women reported recent sexual intercourse (defined as vaginal intercourse in the previous month). In adjusted analyses, sexual activity was not associated with receipt of HAART in the overall sample or in country-specific models. It was, however, positively associated with a measure of “HAART optimism”, suggesting that being optimistic about the effect of HAART on reducing HIV transmission risks and reducing threats to health and survival may be a better predictor of sexual behaviour than actual receipt of HAART. This early finding prompted further  201  examination of HAART optimism among HIV-positive women and led to the development and testing of the HAART optimism scale presented in Chapter 6.  Chapter 3 also revealed that while HAART use wasn’t associated with recent sexual activity, it was positively associated with other sexual behaviours including contraceptive use and practicing protected sex (i.e., consistent condom use). Despite a limited sample size, the findings from the pilot study highlighted the importance and feasibility of further investigating the role of receipt of HAART on sexual and reproductive behaviors among women in high HIV prevalence settings. Subsequently, larger fertility-related studies were developed and implemented in all three sites included in the pilot study. The analyses in Chapters 4 and 5 emerged from the study that I developed and led in collaboration with the Perinatal HIV Research Unit (PHRU) in Soweto, South Africa while the analyses in Chapter 6 emerge from a study led by investigators at Harvard Medical School and the Mbarara University of Science and Technology in Mbarara, Uganda. Chapter 4, in press at the American Journal of Public Health [6], reveals that the adjusted odds of fertility intentions of HIV-positive women in Soweto are 60% lower than those of HIV-negative women, and this difference is largely irrespective of use and duration of HAART. This finding contrasts with those from other sub-Saharan African sites, which showed an increased likelihood of fertility intentions among  202  HIV-positive women receiving HAART compared with HAART-naïve women [7, 8]. The observed lack of association between HAART use and fertility intentions in Soweto may relate to the nature of HAART and HIV care services at the PHRU. Namely, HAART has been available at the PHRU since July 2004, nearly three years longer than at government clinics in South Africa [9] and at least two years longer than many sub-Saharan African settings [10]. The lack of association between HAART use and fertility intentions noted in this study may reflect an emerging reality that HIV-positive women who are not yet receiving HAART can be confident that treatment is available once they are medically eligible, thereby minimizing differences in fertility goals between groups.  This analysis also highlighted that while stated fertility intentions of HIVpositive women are significantly lower than HIV-negative women, they are not negligible. Nearly one-third of HIV-positive women reported intention to have (more) children, a proportion that increased significantly by younger age and fewer living children. These findings speak to the urgent need for policies and services that support HIV-positive women to achieve their reproductive goals in the healthiest and safest possible manner. It is similarly critical that effective, non-coercive, and non-judgmental family planning services be available to the 70% of HIV-positive women who reported that they do not intend to have (more) children.  203  The results in Chapter 4 also reveal that over two-thirds of HIV-negative women intend to have (more) children. In many regions of sub-Saharan Africa, HIV incidence and prevalence are highest among young women in their peak reproductive years. In South Africa in particular, HIV prevalence peaks among women in the 25-29 year age group among whom fully one-third have acquired HIV infection [11]. The substantial fertility intentions of HIV-negative women, viewed within the existing conflict between the unprotected sexual activity required for conception and the risk of HIV acquisition from sexual partners with unknown or known HIV-positive status, demands consideration of targeted reproductive health services to minimize their risks of HIV acquisition through realization of their reproductive goals. Chapter 5, currently under review, describes the prevalence of contraceptive use and associations with HIV status and HAART use among sexually active women in Soweto, South Africa. Overall, a high proportion of women reported using contraception (78%) with significant variation by HIV status and use of HAART. In adjusted models, compared with HIV-negative women, women receiving HAART were significantly more likely to use contraception while HAART-naïve women were marginally more likely (AOR: 2.40; 95% CI: 1.25-4.62 and AOR: 1.59; 95% CI: 0.88-2.85, respectively). These findings are broadly consistent with the results from the pilot study presented in Chapter 3 and results  204  from a paper that I co-authored from the Mbarara study [12]. Although receipt of HAART was positively associated with contraceptive use, there was no association between increasing duration of HAART use and contraceptive prevalence.  Chapter 5 also highlights the important differences in the types of contraceptive methods used by women based on HIV status and receipt of HAART. Such data are severely lacking in the literature. The results show that not only were HAART users significantly more likely to use contraceptive methods overall, they were more likely than HAART-naïve and HIV-negative women to use dual protection (defined as use of both a barrier method (primarily the male condom) and a hormonal or permanent contraceptive method [13]). Women on HAART also reported the highest prevalence of consistent condom use (with or without other methods). Indeed much of the difference in contraceptive prevalence between the three groups was accounted for by the significantly lower prevalence of condom use among HIV-negative women. Compared with HAART-naïve women, HAART users were more likely to use hormonal contraception, and uptake of injectables exceeded oral contraceptive use. Collectively, these results suggest that the regular care, counseling and support associated with HIV care and treatment is contributing to contraceptive uptake among HIV-positive women in this setting. These findings highlight the great potential and on-going need for improved integration of HIV  205  testing, treatment, and care services with sexual and reproductive health services, including the provision of contraceptive options. In Chapter 6, published in AIDS and Behavior in 2009 [14], I developed a reliable and culturally-appropriate HAART optimism scale among HIV-positive women in Mbarara, Uganda and tested the scale's validity against measures of fertility intentions, sexual activity, and unprotected sex. Uniquely, the scale incorporates measures of optimism about becoming pregnant and risks of vertical transmission of HIV, both of which are absent from previously validated and available HAART optimism scales. Overall, HIV-positive women in Mbarara expressed a low level of HAART optimism, with a median score of 11 [IQR: 8-13], out of a possible range from eight (highly skeptical) to 32 (highly optimistic). This suggests that women in this setting remain cautious about the use and effectiveness of HAART on lowering risk of HIV transmission and believing that HIV infection is less of a threat to their health and survival. Nonetheless, higher HAART optimism scores were positively associated with fertility intentions, recent sexual activity, and risky sexual behaviour. These findings constitute an important progression from the relationships explored in previous chapters and suggest that among HIV-positive women with access to HIV treatment services, optimism about the effect of HAART on improving health and reducing transmission risks may be as or more important than actual receipt of HAART. This scale was termed the Women's HAART  206  Optimism Monitoring and EvaluatioN scale (WHOMEN's scale v.1) and is currently being incorporated into other studies aimed at investigating the role of HAART optimism on influencing sexual and reproductive behaviours of HIV-positive women in settings with generalized HIV epidemics.  Together, the evidence provided here suggests that HIV-positive women receiving HAART are more likely to use contraception overall and dual protection in particular, with minimal differences in fertility intentions or rates of sexual activity, relative to their HAART-naïve counterparts. Indeed, optimism about the effects of HAART, rather than actual use or non-use, appears to be more strongly associated with fertility intentions and sexual activity of HIV-positive women. In addition, there remain important differences in sexual and reproductive decisionmaking and behaviours by HIV serostatus. Namely, HIV-positive women are significantly less likely to report fertility intentions and significantly more likely to use contraception (and condoms in particular) relative to HIV-negative women from the same community.  These empirical findings are at times inconsistent with the hypothesized effects of HAART described in the conceptual framework (Figure 2.1). It must be noted that the conceptual framework was developed by reviewing literature from the very early days of HAART scale-up initiatives in sub-Saharan Africa. Since then the scale-  207  up process has been dramatic; in 2008, an estimated 2.9 million people in the region were receiving antiretroviral therapy, a 30-fold increase since the end of 2003 [15]. Thus, early findings and initial hypotheses about the effects of HAART use on the proximate and underlying determinants of fertility may be incomplete, as they were indirectly predicated on the novelty and anticipated scarcity of HAART availability. Moreover, the role of increased regular contact with health care, support, and counseling services as a function of the clinical follow-up required to monitor the health of individuals receiving therapy was not adequately considered in the development of the conceptual framework. While the framework addressed the actual receipt or non-receipt of HAART at the individual-level, the findings presented here suggest the level of availability and accessibility of HIV treatment services in a given community and the role of HAART optimism may be additionally relevant variables. Thus, while the framework remains a useful tool to help guide future research regarding the impact of HAART on fertility and its determinants, current work is underway to update it to include recently available empirical findings.  Nonetheless, the findings of this thesis collectively reveal the critical need and great value of improved integration of HIV testing, treatment, and care services with reproductive and sexual health programming to address the diverse reproductive  208  and sexual needs of HIV-infected and affected women living in HIV endemic settings with expanding access to HAART.  7.3 Unique Contributions  The strengths and limitations of this research project have been detailed in each of the study chapters (2-6). However, taken collectively, the research papers yield several unique contributions to the literature worth highlighting.  An important and lasting contribution of this research lies in its assessment of the impact of HAART on the sexual and reproductive health of women in settings experiencing highly feminized, generalized epidemics. It has been argued that approaches to addressing the HIV epidemic in sub-Saharan Africa have been too heavily weighted on experiences and policies from countries where the disease affects particular “risk groups” [16]. The review in Chapter 2 also cautioned against generalizing findings about the impact of HAART on fertility in North American and European settings to sub-Saharan Africa given great differences in the epidemiology of the epidemic and the cultural, economic, social, and gender dynamics that influence the context of reproductive decision-making between the two settings. Fittingly, this thesis revealed novel patterns of the impact of HAART access on sexual and reproductive health that differed from those previously 209  described in settings not experiencing generalized HIV epidemics.  This thesis also yielded the first reliable and validated HAART optimism scale developed to measure HAART optimism among HIV-positive women in high HIV prevalence settings. Uniquely, the scale incorporated concerns related to pregnancy and vertical transmission of HIV, neither of which was included in previously validated scales. Among HIV-positive women higher HAART optimism was shown to be associated with fertility intentions and sexual behaviours, regardless of actual receipt of HAART. As HAART access continues to rapidly expand in high HIV prevalence settings, this finding suggests a critical need to address perceptions and beliefs about HAART within HIV prevention messaging in the general community, not just among those accessing treatment services.  Although the primary interest of this thesis was to investigate the impact of expanding access to HAART among HIV-positive women, this thesis included comparisons with HIV-negative women from the same community. No other published studies were identified that considered information across all three groups. The study findings revealed important differences and trends in reproductive and sexual health by HIV status and receipt of HAART. It was also possible to document the concerning context of reproduction for HIV-negative women in high HIV prevalence settings. As shown in Chapters 4 and 5, HIV-  210  negative women were unlikely to know the HIV status of their primary sexual partner, the most likely to report fertility intentions, and the least likely to use contraceptive methods overall, and condoms in particular (compared with their HIV-positive counterparts). This study highlighted the critical need to prioritize efforts to keep HIV-negative women negative and the important role of sexual and reproductive health programming in supporting HIV-negative women to safely fulfill their reproductive goals while minimizing their risks of HIV acquisition.  This thesis yielded an opportunity to incorporate findings across two diverse high-needs settings. South Africa and Uganda represent important and appropriate countries within which to investigate the impact of HAART on sexual and reproductive health. As one of the first countries to experience and document a high prevalence of HIV, Uganda is now one of few countries to have shown a sustained decreased HIV prevalence over the last decade from 18% in the mid 1990s to 6.7% today [17]. In contrast, HIV prevalence in South Africa remains stable at 19% [11]. The relative difference in HIV prevalence is also critical in absolute terms. There are an estimated 400,000 women living with HIV infection in Uganda compared with approximately 3.2 million in South Africa. Similarly, while antiretroviral coverage is currently comparable in the two countries (approximately one-third of HIV-positive individuals who need treatment are receiving it) [10], the scale of existing and required treatment services is markedly different. Recent estimates report that over  211  700,000 HIV-positive South Africans are receiving treatment compared with 140,000 Ugandans [10]. While fewer women are living with HIV in Uganda compared with South Africa, the total fertility rate of Uganda is over three times greater (TFR = 6.7 in Uganda; TFR = 2.0 in South Africa) [19].  These differences in HIV, HAART access, and fertility profiles enabled consideration of the relative ways in which HAART may influence sexual and reproductive health decision-making, behaviours, and outcomes and highlight the great differences in the nature and scale of required sexual and reproductive health programming for HIV-positive women in the two countries. In addition to the fertility intentions and contraceptive use papers presented in Chapters 4 and 5 based on data from Soweto, I co-authored two related papers on the impact of HAART among HIV-positive women in Mbarara, Uganda [8, 12]. While there was no HIVnegative comparison group in the Uganda study, we found that there were important differences in the overall prevalence of fertility intentions among HIVpositive women between the two sites (30% of HIV-positive women in Soweto expressed fertility intentions compared with 15% of women in Mbarara) yet similar rates of overall contraceptive use (84% and 85% of HIV-positive women in Soweto and Mbarara, respectively). In addition, receipt of HAART was associated with significantly increased fertility intentions in Mbarara but not in Soweto, and with contraceptive use in both settings. Engaging in investigations across these two sites  212  has offered a unique opportunity to highlight that sexual and reproductive decisionmaking and behaviours of HIV-positive women are likely to vary across settings and, to be effective, the required corresponding sexual and reproductive health programming must similarly consider these differences.  The opportunity to work with collaborators across sites yielded an additional unique contribution of this thesis with respect to knowledge translation and dissemination activities. Upon publication, the early findings of this research endeavour contributed to a growing interest in investigating the impact of HAART on sexual and reproductive health and fertility. Our group of research collaborators felt it was important to share these early research findings with a broader community of stakeholders such that more nuanced methodologies, analyses, and interpretations could be developed. As such, in collaboration with investigators from the Perinatal HIV Research Unit in Soweto, the BC Centre for Excellence in HIV/AIDS, Harvard Medical School, Simon Fraser University, and the University of British Columbia, I co-chaired a public forum and research conference held in Soweto between April 21st-23rd, 2008 entitled: Switching gears in HIV research: Building an international agenda on the impact of highly active antiretroviral therapy (HAART) on fertility in sub-Saharan Africa [20]. The purpose of the conference was to foster international knowledge exchange on emerging research and policy priorities related to reproductive and sexual health, fertility decision-making, and fertility of  213  HIV-positive individuals, during a time of expanding access to HAART. The event brought together internationally renowned HIV/AIDS and reproductive and sexual health stakeholders, including researchers, clinicians, policy-makers, community activists, and representatives of international development and health research funding agencies. The conference and the presented papers functioned to raise the profile of sexual and reproductive health issues in the context of expanding access to HAART and provided an opportunity to contribute to knowledge translation activities. Chapters 2 and 4 of this dissertation were shared at this conference to solicit feedback and to discuss implications of the findings.  As the conference drew to a close, it became clear that although exciting and important research was being conducted in the region, there was a dearth of published literature available. As such, a commitment was advanced to support local researchers and conference delegates to develop their research findings into manuscripts and to circulate a broader international call for related papers on the topic of HIV, HAART, and fertility in sub-Saharan Africa. The end result of this commitment was a published supplement edition of the journal AIDS and Behavior (published in print in July 2009) [21]. I was the lead Guest Editor for the supplement. The supplement provides an array of research findings from across the region to be used towards the design and implementation of evidence-based reproductive and sexual health policies and programs that support women, their families, and their  214  communities in reproductive decision-making in the context of HIV. The engagement of a larger community of stakeholders, particularly women living with HIV/AIDS in sub-Saharan Africa, greatly strengthened the interpretation and local relevance of the research findings presented here.  7.4 Recommendations  Each of the study chapters provides recommendations for policy and programming specific to each research finding; however, there are several recommendations worth highlighting upon considering the cumulative evidence presented here.  First, all women, including women living with HIV, should be supported to achieve their reproductive goals in the healthiest and safest possible manner [1]. Given the high prevalence of fertility intentions among HIV-positive women, it is critical that factual and non-stigmatizing information and support be incorporated into HIV treatment services to optimize sexual and reproductive outcomes for mother, father, and baby. This includes counseling services regarding use of HAART during pregnancy, safer labour options, accessible and comprehensive PMTCT services, antenatal and postnatal care, and support for safer infant feeding practices. No clear guidelines exist to support HIV-affected couples in limited215  resource settings who want to become pregnant to do so safely and this information is urgently needed. In particular, HIV-affected women require and have requested information regarding the ideal time for pregnancy with respect to CD4 level, stage of treatment, treatment regimen, viral load, and the HIV- and health-status of her sexual partner [22]. Without this information and support, HIV-affected couples may knowingly risk transmission in order to conceive, unaware that there are a range of strategies to mitigate the risk of infection. More must be done towards adopting a harm reduction approach to fertility counseling in resource-limited settings to minimize risk of HIV transmission between sero-discordant couples [23].  Since a substantial proportion of women do not want to have more children, it is also critical that effective, non-judgmental, and non-coercive family planning services (including access to termination of pregnancy services) are available to all women who need them. As shown in the contraceptive use paper, the higher prevalence of contraceptive use overall and dual protection methods in particular among women receiving HAART highlights a critical opportunity to assist women living with HIV/AIDS prevent unintended pregnancy and prevent HIV transmission. Overall, more work must be done on the provision of contraceptive options to close the gap between reported fertility intentions and fertility.  216  This work also highlights the great need for integration of HIV services with sexual and reproductive health services [24]. Integration is critical to support the clearly intertwined sexual and reproductive needs of HIV-affected women living in the context of generalized epidemics. Moreover, integration is also of operational value since it holds potential to reduce service duplication and could help reduce the stigma of accessing separate HIV services. Globally, an inspiring degree of effort and resources are currently being directed towards the critical need to expand access to HIV treatment services [10]. The additional expenditure associated with incorporating family planning services represents a highly cost-effective means of reducing unintended pregnancies and further minimizing vertical and horizontal HIV transmission, thus contributing to healthier individuals and families in HIV endemic regions [25].  Finally, the findings presented here have particular relevance to pioneering “Treatment as Prevention” strategies, which aim to curb HIV transmission through early, sustained, and successful antiretroviral treatment of people living with HIV [26]. As shown, expanding access to HIV treatment involves more than the provision of antiretroviral drugs; it can be an opportunity to address other aspects of health and well-being including sexual and reproductive health. In this study, women on HAART were more likely to use contraceptives overall, and barrier, permanent, and dual protection methods in particular. Of particular relevance to HIV prevention  217  efforts, 68% of sexually active HAART users reported consistent condom use, compared with 59% of HAART-naïve women, and 39% of HIV-negative women. Current models estimating the impact of expanding access of HAART on reducing HIV transmission assume rates of condom use reported in the general population [27]. However, the higher prevalence of condom use among HAART users suggests a further reduced risk of HIV transmission to discordant sexual partners than previously estimated given that 80% of women receiving HAART in our study are also virally suppressed. Explicitly incorporating sexual and reproductive health concerns of women into ‘Treatment as Prevention’ initiatives offers an opportunity to extend the value of HAART scale-up services to address the desperate state of maternal and child health in high HIV prevalence settings.  7.5 Future Research  Based on the findings presented here, there are a number of priority areas for future research on this topic. First, there is a great need for longitudinal data to monitor the ongoing impact of expanded access to HAART on the sexual and reproductive health of HIV-infected and –affected women. In particular, there is great value in conducting a large observational cohort study with an open recruitment strategy that enrolls women at various stages of their HIV disease and  218  reproductive lives. Such a study could be responsive to changes in HIV testing and incidence rates, changes in rates and strategies of HAART provision, and changes in the policy and programming environments that affect sexual and reproductive health of women living in conditions of generalized epidemics. It would be valuable to enroll both HIV-positive and HIV-negative women in such a cohort study as an important means of understanding systemic differences between groups that could not be fully accounted for in the analyses presented here. Moreover, a longitudinal assessment, which includes data pre- and post- HIV seroconversion and pre- and post HAART initiation, would be more appropriate to accurately assess a timeoriented potential impact of HAART access on sexual and reproductive decisionmaking and behaviors and ultimately, outcomes. For example, the systemic and genital tract changes associated with progesterone exposure have suggested a biological plausibility to increase risk of HIV acquisition [28-31], transmission [3234], and disease progression [35] among oral contraceptive users. Although the epidemiologic evidence has yielded inconsistent findings in the general female population [36-40], it is possible that women using oral contraceptives may be at increased risk of HIV acquisition and disease progression, necessitating earlier receipt of HAART. The longitudinal cohort study proposed here would provide an opportunity to address this question of great public health priority.  It is acknowledged that women do not make reproductive decisions in  219  isolation. Thus, future research, including the cohort study described above, should include men both as individuals and as sero-concordant or discordant partners, as part of investigations investigating the impact of expanded access to HAART.  Another important future research direction is for a community-based study to complement the findings of clinic-based studies. Existing research has primarily enrolled HIV-positive and HIV-negative women from research-oriented clinical settings, which may limit the generalizability of the findings. Women who are actively seeking out HIV testing and treatment services may constitute a selective sample since most have disclosed their status and are seeking preventative health services. A community-based study of the potential impact of HIV status and HAART on sexual and reproductive health may yield more representative findings of the general population of women living with HIV in endemic settings.  Future research should utilize the WHOMEN’s scale developed here to measure HAART optimism among HIV-positive women and to assess the role of optimism, versus actual receipt of HAART, on women’s sexual and reproductive decision-making, behaviours, and outcomes. Use of the WHOMEN’s scale would provide opportunities to investigate additional aspects of sexual and reproductive behaviour that may have implications for HIV transmission. In particular, it would  220  be of great value to use the scale to measure trends in HAART optimism corresponding with population-level changes in access to HIV treatment services.  Finally, despite the great value of epidemiologic analyses presented here, such quantitative data fails to capture the salient influence of cultural and gender dynamics on sexual and reproductive decision-making. As such, there is a great need for qualitative studies that explore the reasons and processes behind the outcomes measure here. Cultural beliefs and practices have often and consistently been discussed as critical determinants of sexual and reproductive behaviour [41, 42] and HIV status and HAART use alone are unlikely to be the sole or even primary drivers of reproductive decision-making [2]. Indeed, qualitative studies have highlighted the critical role of male partners, opinions and support of health care workers, and community and peer perceptions of positive parenting as critical factors influencing fertility decision-making of HIV-positive women [43, 44]. Additional such studies are desperately needed to provide a more complete and contextual picture of the impact of expanding access to HAART on sexual and reproductive decision-making and behaviours.  221  7.6 Conclusion My PhD dissertation offers an important and lasting body of evidence pertaining to the role of expanding access to HAART in influencing sexual and reproductive decision-making and outcomes among women in two high HIV prevalence settings in sub-Saharan Africa. It has highlighted that HIV-positive women receiving HAART are more likely to use contraception overall and dual protection in particular, with minimal differences in fertility intentions or rates of sexual activity, relative to their HAART-naïve counterparts. Moreover, the findings reveal that optimism about the effects of HAART, rather than actual use or non-use, may be a more important predictor of fertility intentions and sexual activity of HIVpositive women. In addition, this thesis documents important differences in sexual and reproductive decision-making and behaviours by HIV serostatus. In particular, HIV-positive women are significantly less likely to report fertility intentions and significantly more likely to use contraception (and condoms in particular) relative to HIV-negative women from the same community. The findings have underscored the great value that integrated sexual and reproductive health services and HIV care and services can have towards improving HIV prevention efforts and supporting the rights of all women to be sexually active and achieve their fertility goals, while minimizing risks of HIV transmission.  222  7.7 References  1.  Gruskin S, Ferguson L, O'Malley J. Ensuring sexual and reproductive health for people living with HIV: an overview of key human rights, policy and health systems issues. Reprod Health Matters. 2007; 15:4-26.  2.  Kaida A, Andia I, Maier M, Strathdee SA, Bangsberg DR, Spiegel J, et al. The potential impact of antiretroviral therapy on fertility in sub-Saharan Africa. Curr HIV/AIDS Rep. 2006; 3:187-94.  3.  Boerma JT, Weir SS. Integrating demographic and epidemiological approaches to research on HIV/AIDS: the proximate-determinants framework. J Infect Dis. 2005; 191 Suppl 1:S61-7.  4.  Bongaarts J, Potter RG. Fertility, biology, and behavior: An analysis of the proximate determinants. New York: Academic Press; 1983.  5.  Kaida A, Gray G, Bastos FI, Andia I, Maier M, McIntyre J, et al. The relationship between HAART use and sexual activity among HIV-positive women of reproductive age in Brazil, South Africa, and Uganda. AIDS Care. 2008; 20:21-5.  6.  Kaida A, Laher F, Strathdee SA, Janssen PA, Money D, Hogg RS, et al. Fertility intentions of HIV-positive women of reproductive age in Soweto,  223  South Africa: The influence of expanding access to HAART in an HIV hyperendemic setting. Am J Public Health. In press.  7.  Cooper D, Moodley J, Zweigenthal V, Bekker LG, Shah I, Myer L. Fertility intentions and reproductive health care needs of people living with HIV in Cape Town, South Africa: implications for integrating reproductive health and HIV care services. AIDS Behav. 2009; 13 Suppl 1:38-46.  8.  Maier M, Andia I, Emenyonu N, Guzman D, Kaida A, Pepper L, et al. Antiretroviral therapy is associated with increased fertility desire, but not pregnancy or live birth, among HIV+ women in an early HIV treatment program in rural Uganda. AIDS Behav. 2009; 13 Suppl 1:28-37.  9.  Kapp C. South Africa unveils new 5-year HIV/AIDS plan. Lancet. 2007; 369:1589-90.  10.  WHO. Towards universal access: scaling up priority HIV/AIDS interventions in the health sector: Progress report 2008. Geneva: World Health Organization (WHO); 2008.  11.  Shisana O, Rehle T, Simbayi L, Zuma K, Jooste S, Pillay-van-Wyk V, et al. South African national HIV prevalence, incidence, behaviour and communication survey 2008: A turning tide among teenagers? Cape Town: Human Sciences Research Council (HSRC); 2009.  224  12.  Andia I, Kaida A, Maier M, Guzman D, Emenyonu N, Pepper L, et al. Highly active antiretroviral therapy and increased use of contraceptives among HIVpositive women during expanding access to antiretroviral therapy in Mbarara, Uganda. Am J Public Health. 2009; 99:340-7.  13.  United Nations (UN). World Contraceptive Use. New York: Department of Economic and Social Affairs, Population Division; 2007.  14.  Kaida A, Lima VD, Andia I, Kabakyenga J, Mbabazi P, Emenyonu N, et al. The WHOMEN's scale (Women's HAART Optimism Monitoring and EvaluatioN Scale v.1) and the association with fertility intentions and sexual behaviours among HIV-positive women in Uganda. AIDS Behav. 2009; 13 Suppl 1:72-81.  15.  World Health Organization (WHO). Towards Universal Access: Scaling up priority HIV//AIDS interventions in the health sector - 2009 Progress Report. Geneva: World Health Organization (WHO); 2009.  16.  De Cock KM, Mbori-Ngacha D, Marum E. Shadow on the continent: public health and HIV/AIDS in Africa in the 21st century. Lancet. 2002; 360:67-72.  17.  UNAIDS. Report on the global HIV/AIDS epidemic 2008. Geneva: UNAIDS; 2008.  225  18.  Uganda Bureau of Statistics and Macro International Inc. Uganda Demographic and Health Survey 2006. Calverton: Uganda Bureau of Statistics and Macro International Inc.; 2007.  19.  SADHS. South Africa Demographic and Health Survey 2003. Pretoria: Department of Health; 2004.  20.  Forrest J, Kaida A, Conference summary report: Switching gears in HIV research: Building an international research agenda on the impact of HAART on fertility in sub-Saharan Africa. Soweto: Perinatal HIV Research Unit (PHRU); 2008.  21.  Kaida A, Bangsberg DR, Gray G, Hogg RS, King R, Miller CL. Editorial: Introduction to the Supplement on HIV, HAART, and Fertility in sub-Saharan Africa. AIDS Behav. 2009; 13 Suppl 1:1-4.  22.  Nduna M, Farlane L. Women living with HIV in South Africa and their concerns about fertility. AIDS Behav. 2009; 13 Suppl 1:62-5.  23.  Matthews LT, Mukherjee JS. Strategies for harm reduction among HIVaffected couples who want to conceive. AIDS Behav. 2009; 13 Suppl 1:5-11.  24.  World Health Organization (WHO). Strategic considerations for strengthening linkages between family planning and HIV/AIDS policies, programs, and services. Geneva: World Health Organization; 2009.  226  25.  Halperin DT, Stover J, Reynolds HW. Benefits and costs of expanding access to family planning programs to women living with HIV. AIDS. 2009; 23(Suppl 1):S123-S130.  26.  Montaner JS, Hogg R, Wood E, Kerr T, Tyndall M, Levy AR, et al. The case for expanding access to highly active antiretroviral therapy to curb the growth of the HIV epidemic. Lancet. 2006; 368:531-6.  27.  Granich RM, Gilks CF, Dye C, De Cock KM, Williams BG. Universal voluntary HIV testing with immediate antiretroviral therapy as a strategy for elimination of HIV transmission: a mathematical model. Lancet. 2009; 373:4857.  28.  Louv WC, Austin H, Perlman J, Alexander WJ. Oral contraceptive use and the risk of chlamydial and gonococcal infections. Am J Obstet Gynecol. 1989; 160:396-402.  29.  Miller L, Patton DL, Meier A, Thwin SS, Hooton TM, Eschenbach DA. Depomedroxyprogesterone-induced hypoestrogenism and changes in vaginal flora and epithelium. Obstet Gynecol. 2000; 96:431-9.  30.  Morrison CS, Bright P, Wong EL, Kwok C, Yacobson I, Gaydos CA, et al. Hormonal contraceptive use, cervical ectopy, and the acquisition of cervical infections. Sex Transm Dis. 2004; 31:561-7.  227  31.  Wira CR, Fahey JV. A new strategy to understand how HIV infects women: identification of a window of vulnerability during the menstrual cycle. AIDS 2008; 22:1909-17.  32.  Mostad SB, Overbaugh J, DeVange DM, Welch MJ, Chohan B, Mandaliya K, et al. Hormonal contraception, vitamin A deficiency, and other risk factors for shedding of HIV-1 infected cells from the cervix and vagina. Lancet. 1997; 350:922-7.  33.  Prakash M, Kapembwa MS, Gotch F, Patterson S. Oral contraceptive use induces upregulation of the CCR5 chemokine receptor on CD4(+) T cells in the cervical epithelium of healthy women. J Reprod Immunol. 2002; 54:117-31.  34.  Wang CC, McClelland RS, Overbaugh J, Reilly M, Panteleeff DD, Mandaliya K, et al. The effect of hormonal contraception on genital tract shedding of HIV-1. AIDS. 2004; 18:205-9.  35.  Asin SN, Heimberg AM, Eszterhas SK, Rollenhagen C, Howell AL. Estradiol and progesterone regulate HIV type 1 replication in peripheral blood cells. AIDS Res Hum Retroviruses. 2008; 24:701-16.  36.  Morrison CS, Richardson BA, Mmiro F, Chipato T, Celentano DD, Luoto J, et al. Hormonal contraception and the risk of HIV acquisition. AIDS. 2007; 21:85-95.  228  37.  Myer L, Denny L, Wright TC, Kuhn L. Prospective study of hormonal contraception and women's risk of HIV infection in South Africa. Int J Epidemiol. 2007; 36:166-74.  38.  Womack J, Williams A. Hormonal contraception in HIV-positive women. AIDS Read. 2008; 18:372-7, 381.  39.  Stringer E, Antonsen E. Hormonal contraception and HIV disease progression. Clin Infect Dis. 2008; 47:945-51.  40.  Stringer EM, Levy J, Sinkala M, Chi BH, Matongo I, Chintu N, et al. HIV disease progression by hormonal contraceptive method: secondary analysis of a randomized trial. AIDS. 2009.  41.  Sonko S. Fertility and culture in sub-Saharan Africa: a review. Int Soc Sci J. 1994; 46:397-411.  42.  Caldwell JC, Caldwell P. The cultural context of high fertility in sub-Saharan Africa. Population and Development Review. 1987; 13:409-437.  43.  Cooper D, Harries J, Myer L, Orner P, Bracken H, Zweigenthal V. "Life is still going on": reproductive intentions among HIV-positive women and men in South Africa. Soc Sci Med. 2007; 65:274-83.  229  44.  Kirshenbaum SB, Hirky AE, Correale J, Goldstein RB, Johnson MO, Rotheram-Borus MJ, et al. "Throwing the dice": pregnancy decision-making among HIV-positive women in four U.S. cities. Perspect Sex Reprod Health. 2004; 36:106-13.  230  APPENDIX 1: HUMAN ETHICS APPROVAL CERTIFICATES The University of British Columbia Office of Research Services Behavioural Research Ethics Board Suite 102, 6190 Agronomy Road, Vancouver, B.C. V6T 1Z3  CERTIFICATE OF APPROVAL - FULL BOARD PRINCIPAL INVESTIGATOR:  INSTITUTION / DEPARTMENT: UBC BREB NUMBER: UBC/Medicine, Faculty of/Health Patricia Janssen H06-04019 Care & Epidemiology INSTITUTION(S) WHERE RESEARCH WILL BE CARRIED OUT: Institution  N/A  Site  N/A  Other locations where the research will be conducted:  Perinatal HIV Research Unit (PHRU) of the University of Witwatersrand's Chris Hani Baragwani Hospital in Soweto, South Africa  CO-INVESTIGATOR(S): Angela Kaida Deborah Money Robert S. Hogg Steffanie Strathdee  SPONSORING AGENCIES: BC Centre for Excellence in HIV/AIDS - "Investigating the impact of Highly Active Antiretroviral Therapy (HAART) on sexual and reproductive behaviours and fertility among HIV-positive women in South Africa " Canadian Institutes of Health Research - "Measuring the impact of highly active antiretroviral therapy on HIV-positive women's sexual behaviour and fertility in a high prevalence setting" PROJECT TITLE: Investigating the impact of Highly Active Antiretroviral Therapy (HAART) on sexual and reproductive behaviours and fertility among HIV-positive women in South Africa REB MEETING DATE: CERTIFICATE EXPIRY DATE: January 25, 2007 January 25, 2008 DOCUMENTS INCLUDED IN THIS APPROVAL: DATE APPROVED: March 21, 2007 Document Name  Protocol: Research Proposal Consent Forms: Revised consent form Questionnaire, Questionnaire Cover Letter, Tests: Questionnaire_Gender, Fertility, HAART study_South Africa Other Documents: Ethical Approval from the Human Ethics Research Board (Medical): University of the Witwatersrand  Version  Date  N/A  December 15, 2006  4  March 21, 2007  N/A  December 1, 2006  N/A  December 1, 2006  The application for ethical review and the document(s) listed above have been reviewed and the  231  procedures were found to be acceptable on ethical grounds for research involving human subjects. Approval is issued on behalf of the Behavioural Research Ethics Board and signed electronically by one of the following:  Dr. Peter Suedfeld, Chair Dr. Jim Rupert, Associate Chair Dr. Arminee Kazanjian, Associate Chair Dr. M. Judith Lynam, Associate Chair Dr. Laurie Ford, Associate Chair  232  233  234  235  236  237  238  239  240  APPENDIX 2: DATA COLLECTION TOOL FOR THE SOWETO, SOUTH AFRICA STUDY  241  242  243  244  245  246  247  248  249  250  251  252  253  254  255  256  257  258  259  260  261  262  263  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0058273/manifest

Comment

Related Items