SOCIAL AND STRUCTURA L FACTORS AFFECTING N IN PREVENTION OF MOTHER TO CHILD TRANSMISSIO N (PMTCT) PROGRAMS I N MALAWI By Kristan Elwell A DISSERTATION Submitted to Michigan State University in partial fulfillment of t he requirements for the degree of Anthropology Doctor of Philosophy 2015 ABSTRACT SOCIAL AND STRUCTURA L FACTORS AFFECTING N IN PREVENTION OF MOTHER TO CHILD TRANSMISSIO N (PMTCT) PROGRAMS I N MALAWI By Kristan Elwell In 2012, 17.7 million women were living with HIV. Ninety percent of these women lived in Sub -Saharan Africa. Malawi is especially impacted, with some of the highest rates of HIV infection in pregnant women in the world. In 2010, 63,500 pregnant women in Malawi required antiretroviral (ARV) prophylaxis. Of these women, only 53% received ARVs through PMTCT programs. In 2011, in an effort to expand access to treatment services within PMTCT programs, the Malawian Ministry of Health implemented Plan B+, a policy designed to e xpand access to treatment services within PMTCT programs. Plan B+ now offers lifelong treatment to all pregnant women and new mothers who test positive for the HIV virus, regardless of CD4 count. Despite a substantial national scale -up of PMTCT services, r etaining women within PMTCT programs remains a challenge to successful implementation of these interventions. Recent literature suggests that approximately 27 -55% of women enrolled in PMTCT programs fail to remain in them. This dissertation examines the so participation in Prevention of Mother to Child Transmission (PMTCT) programs in Southern Malawi. I employ qualitative methods including semi -structured interviews, focus group discussions and clinical o bservations with PMTCT patients, health care providers and community members to better understand why some women do not remain in these programs. The research questions examine: 1) The experiences and health -seeking strategies employed by pregnant and post participation in PMTCT programs, and conversely, 3) How participation in PMTCT programs wome identities and constraining their health -seeking practices; thus demonstrating how gender inequities become embodied in the health outcomes of HIV -positive women. Copyright by KRISTAN ELWELL 2015 v I would like to give special thanks and dedicate this dissertation to my parents, John and Sarah Elwell, who made all of this possible. I appreciate their endless encouragement, patience, and ongoing support. And to th managed her illness taught me so much about the endurance of the human spirit. vi ACKNOWLEDGEMENTS I would like to sincerely thank my advisor, Anne Ferguson and my dissertation com mittee, Linda Hunt, James Pritchett, and Rebecca Malouin for their ongoing guidance and support throughout my graduate work. I would like to thank my family for all their support throughout my graduate studies . I would like to thank the University of Malaw College of Medicine for allowing this research to be conducted and the staff at the antenatal, postnatal and ART clinics at Queen Elizabeth Central Hospital , and the staff at Mpemba and Ndirande Health Centres for their support. I am grateful for the w ork of Esau and Christina, my research assistants ; Wilson, my Chichewa instructor; and for the participation of men and women from the communities of Blantyre, Ndirande, and Mpemba . I also received support from the staff of the Department of Anthropology, the Center for the Advanced Study of International Development (CASID), and the Center for Gender in Global Context (GenCen). This research was supported by Michigan State University Internal Replacement for the 2012 Fulbright DDRA Fellowship in Malawi; additionally, the entire dissertation process was made possible with funding from the Michigan State University African Studies Center and Foreign Language and Areas Studies (FLAS) Fellowships for Chichewa . vii TABLE OF CONTENTS LIST OF TABLES xi LIST OF FIGURES xii CHAPTER 1: Introduction 1 Living with HIV/AIDS in Malawi: Chitsulo 1 Research Questions 7 Purpose of the Study 7 Conceptual Framework 8 Identity Multiplicity as an Analytic Concept 9 Organization of the Dissertation 10 Chapter 2: Literature Review 12 Chapter 3: Research Methods 13 Chapter 4: Local Understandings of Reproduction and the Construction of Motherhood 13 Chapter 5: Gender Roles and Relati ons in Malawi 14 Chapter 6: Local Understandings of HIV/AIDs and Community -Based Stigma 14 Chapter 7: Dissertation Conclusions 14 CHAPTER 2: Literature Review 15 PMTCT Programs and Patient Adherence 15 Theoretical Framework 21 Impact of the HIV/AIDS Epidemic 23 Impact on the National Health Sector 23 Impact of HIV/AIDS on Households 26 Poverty and Global Inequalities Intersections with AIDS 28 30 Development of Biomedical Health Care in Colonial Africa 31 Development of Western Biomedical System 32 Production of Sexuality in the Colonial Metropole 35 Conclusion 37 CHAPTER 3: Research Methods 39 Part I: Study Objectives a nd Research Questions 39 Part II: Data Collection 40 Study Setting 40 Blantyre 42 Mpemba 42 Ndirande 43 Summary of Methods Used 43 viii Clinical Observations 43 Focus Group Discussions Patients 45 Focus Group Discussions Community Members 45 Patient Interviews 47 Provider Interviews 48 Community Interviews 49 Patient Demographics 49 Contextualizing Kinship Demographics 51 Study Sites and Methods 52 Blanty re 52 Patients 52 Health Care Providers 54 Community Leaders 54 Ndirande 55 Mpemba 55 Patients 55 Patients Who Dropped 56 Health Care Providers 58 Community Leaders 59 Addition al Methods Used 60 Member Checks 60 Analysis 61 Quantitative Survival Analysis 62 CHAPTER 4: Local Understandings of Reproduction and the Construction of Motherhood 64 Part I: Meaning of Reproduction to Local Femin inities and Masculinities 65 Part II: Community Responses to Infertile Women: Pregnancy, Secrecy and Social Risk 69 Infertility as Gendered Embodiment 72 Meaning of Reproduction to Family and Kinship Group 75 Commodifying Infertility: Multiple Wives and Surrogate Sisters 77 Monitoring Fertility: Family Planning Movements and Donor Support 83 84 Part III: Enacting Motherhood: Maternal Sentiment and Prevention of Mother to Child Transmissi on 85 Motherhood Discourse 88 Conclusion 96 CHAPTER 5: Gender Roles and Relations in Malawi 99 Part I: Defining Gender Inequalities 101 Role of the Matrilineage 104 Gender and Household Production 106 Female Gender Roles 106 Division of Labor 107 ix Access to Land 108 Gender Roles and Socialization 109 Gendered Dimensions of Education and Employment 111 Part II: Gender and Power 113 Control over Reproduction 113 Pol itical Power 114 Relationship Power 115 Economic Dependence 115 Decision -Making Dominance 119 Social Norms 124 Fidelity Intersections with Local Masculinities 127 Part III: : Resistance and Accommodation 130 Conclusion 131 CHAPTER 6: Local Understanding of HIV/AIDS and Community -Based Stigma 133 Part I: Defining Stigma 134 -Based Stigma 135 Stigma and the Structure of PMT CT Programs 136 138 Gendered Embodiment and the Clinical Phenomenon of Discordance 141 Forms of Stigma 142 Positionality, AIDS Gossip and Performing the Good Wife 144 Part II: Health Care Provider Attitudes within PMTCT Programs 147 151 Conclusion 156 CHAPTER 7: Dissertation Conclusions 158 Community -Based Stigma 160 Local Understandings of Reproduction and the Construction of Motherhood 161 Gender In equities in Marriage 163 Double Bind: Negotiating the Competing Identities of Wife, Mother, and Patient 164 Employing Secrecy: Mediating Competing Identities 165 Study Limitations 167 Gender and HIV/AIDs Research 168 Suggestions for Future Research 170 Addressing Gender Inequities: Increasing the Relationship Power of HIV -Positive Women 170 Providing Social Support from Other HIV -Positive Women 171 Minimizing HIV/AIDS Stigma through Service Delivery 171 APPENDICE S 173 Appendix A: Survival Analysis of Plan B+ PMTCT Patient Adherence 174 Appendix B: PMTCT Core Services and Patient Profiles 183 Appendix C: Interviews and Instruments Used 186 x REFERENCES 193 xi LIST OF TABLES Table 1: Kapl an Meier Life Table Daily Summary 177 Table 2: Cox Regression Analysis Results 181 Table 3: PMTCT Core Services 184 Table 4: Patient Profile 185 Table 5: Instruments Used 187 xii LIST OF FIGURES Figure 1: Survival Func tion 179 Figure 2: Hazard Function 180 Figure 3: Cox Regression Analysis 181 1 Chapter 1 : Introduction Living with HIV/AIDS in Malawi: Chitsulo When I first met Chitsulo, she was sitting in one of four thatched -roof huts clustered in a semi -circle on a small stretch of land. Chitsulo had a deep and raspy voice, and spoke with a steady cadence, tenacity, and maturity. Her voice belied her age, making her appear older than twenty -two years. Her thin figure, however, revealed her yout h and her recent medical respect. She spoke frankly about her life, her family, her marriage, and her experience with HIV/AIDs. Despite a year filled with several social and medical challenges, she appeared to be doing well. I thought she might be a good candidate for a life history. For the past couple of days, my research assistant and I had been poring over boxes of worn yellow medical ca rds to find those marked had just begun the more difficult and most sensitive stage of my study, locating and then interviewing those women unable to remain on antiretroviral (ART) treatment as participants within PMTCT programs. I asked Chitsulo if I could return again to interview her about her life, and she agreed. her arrival, her neighbors saw her removing ARVs from a hole she had dug in the ground next to her hut, and later told her husband that she was taking ARVs. When her husband learned of her HIV status, he beat her and forbade her to take the ARVs, angered that she had not told him of her positive status and that 2 she had been found to be HIV -positive, which brought shame to their new marriage. She fled to her natal home and her husband soon followed. He begged her to return to him, at which point she consult ed with her family. When I asked if she considered divorce, she explained that her ankhoswe (uncle) came to the village to discuss the situation with her uncle and end thing is that forgive her husband because she did not want to b During the first of three visits, I asked Chitsulo about her extended family living at the that her mother did not i gnore her as her other family members had. Her mother fed her and nsima father brother, in contrast, travelled to the clinic to collect her ARVs and carefully watched over her, eldest male, he would greet us and remain until Chitsulo excused him. As the eldest brother, he followed kinship norms that assigned him r -being. During the course of our interviews, Chitsulo shared with me her challenges with her marriage. At twenty -two years old, she proudly asserted her marital status, even though she was and wanted a husband because her brothers and sisters had all reached this cultural and 3 developmental juncture well before her. She simply wanted to get on with her life. She was sharp, outspok en, and defiant in her zest for life. I was struck by one comment that Chitsulo experiences with HIV/AIDS management. During one of our conversations, Chitsulo tol d me, will worsen. Hiding it (HIV) from your man is not good. It is better you tell them and if they Because of the six months that she had not taken the ARVs, she was switched to the second line medication. 1 Emmanuel had visited Chitsulo on one of his outreach visits as a health services assistant. Her village was part of his catchment area, and I had asked him to stop in to see how she was doing. He told me she had lost a significant amount of weight and was very weak. I returned to the clinic the next day. Emmanuel asked the community health nurse when she was next doing her the Health Services Assistants and nurses who conducted health outreach in the surrounding villages had been an ongoing challenge because of limited resources within the public health system. Yet I knew that a bicycle had been recently supplied by the District Health Office to conduct outreach in the surrounding villages. It took twenty minutes to drive to her home, and probably would have taken forty minutes to bicycle there. When I o ffered to provide the when she was not on duty. Limited health care staff at clinics within the public health system !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!"!All patients who remain off the medication for an extended time are tran sferred to the second line medication to prevent drug resistance. !4 was an additional ongoing challenge. Two conducted an assessment and determined that Chitsulo had edema and moderate malnutrition. The nurse immediately placed her in the nutrition program at the clinic, which meant she would receive Plumpy Nut a ready -to-eat, peanut butter -based therapeutic food that was given to One week later, I asked Emmanuel how Chitsulo was doing and if he had noticed any improvement. She had not been to the clinic since the visit with the community health nurse. That Sunday when I was out of town, I noticed a text from Emmanuel stating that Chit Emmanuel why he thought they were not going to the clinic, he thought perhaps they did not have the money for transport. Chitsulo had mentioned there father decided how money was to be spent. As her mother had little access to this money because local gender norms conferred control over farm in search of ganyu (day labor) to supplement the meager household income. clinic if her family had not done so. When we arrived, the compound was empty except for and her infant Anthony, who hung from her shoulder. The mother and father had taken Chitsulo to the clinic. Her father borrowed a car from his employer at the local water company. It was the first time he had become engaged in her care. When her parents 5 arrived at the clinic, the clinic staff advised her to receive care at the district hospital because of her frail condition. Health clinics are staffed by clinicians, nurses, and HSAs (Health Surveillance Assistants), but not medical doctors. Further, all a cute medical conditions are referred to the district hospital, the only referral center in the region. Despite the often limited resources available at the rural clinic, a brand -new white SUV had recently been supplied to the clinic to serve as the clinic ambulance. On that Monday, however, it was being used by another health clinic. For two days, Emmanuel and I waited to hear about her status. On Wednesday, I received a phone call from her brother that she had died. When I arrived at the clinic the followi ng week, I was greeted by the nurse I had driven and her expressio n morphed from enthusiasm to fear. I walked away and returned to a long line ple no social relations that sometimes worked for her, in the social support from her mother and brother but more often worked against her, when shap ed by gender roles and inequalities, and the social stigma associated with HIV/AIDS. When situated in a context of limited resources, both within the public health system that ultimately failed her, and a household economy structured by gender norms which had little to offer her except on the day before her death her determination was not enough. 6 My study focuses on the experiences of HIV -positive women who are either pregnant or new mothers. I wanted to find out why women would be unable to remain on ARV t reatment both for themselves and their newborn child now that ARVs are free of cost in Malawi. However, I included in my study sample other members of the community with AIDS, like Chitsulo, who could illustrate the lived experience of women who participat e in PMTCT. I community. HIV/AIDS pointed to a similar pattern in the lives o f the PMTCT patients I had met and interviewed. She had intense and unwavering support from her mother and older brother who fed her, washed her, collected her medications, and advocated for her many social and medical needs. But the trajectory of her heal th-seeking was ultimately shaped by gender inequalities in her marriage and a lack of HIV disclosure to her husband that reinforced them, stigma from her extended family, and political economic forces that created fractures in her household economy and wit hin the local public health system. These social and economic forces intersected in a cascade of events many of which had little to do with the bottle of ARVs that sat on the table next to her. My initial interest in studying patient access to HIV/AIDS ser vices began while conducting exploratory research to determine my dissertation topic. I spoke with several health care providers working directly with AIDS patients within the Malawian public health system. I soon learned that one of the main challenges in managing AIDS was the biomedical concept of - attending the Prevention of Mother to Child Transmission (PMTCT) clinics where rates of non -7 adherence were twice that of the general adult population of HIV/AIDs patients. I had a limited the broader social context in which HIV -positive pregnant women and new mothers managed HIV/AIDS fo r themselves and their exposed child at a critical juncture in the HIV/AIDS epidemic when antiretroviral treatment was newly accessible to the poor. Research Questions This public health question of why some patients are able to adhere to treatment and ot HIV/AIDs management tell us about the daily lives of Malawian women? 2) Specifically, how ram? and 3) C et al. 2008; Castro and Farmer 2005; Kasenga et al. 2010; Kuonza et al. 2010; Laher et al. 2012; Mepham et al. 2011; Nyblade et al. 2003; Stinson et al. 2012; Theilgaard et al. 2011; Torpey et al. 2010; Ujiji et al. 2011; Van Hollen 2013; Van der Straten et al. 2014) may interfere with ithin PMTCT programs. The notion that social support may interfere family and community levels. Purpose of the Study 8 to AIDS treatment in PMTCT programs. I analyze the barriers and supports women face within these programs and how these re relations in marriage and the community. I then analyze the implications of these social and structural factors for the health -seeking practice of HIV -positive pregnant women and new mothers wi thin PMTCT programs. Conceptual Framework This research is informed by critical medical anthropology theory, and the theory of revealed by my study lack of HIV disclosure to husbands, fear of community -based stigma, and negative interactions with health care workers using the concepts of structural vulnerability (Quesada, Hart and Bourgois 2011) and the theory of gender and power. Specifically, I examine how gend er and power interact within the social relations of poor peri -urban PMTCT patients and through the lens of HIV/AIDS, producing HIV -related stigma and reproducing gender inequalities. I then explore how the intersection of gender and power produce the heal th-seeking decisions and practices of HIV -positive pregnant women and new mothers who participate in PMTCT programs, and how these social inequalities become embodied in their treatment onnell 1987). I include in this analysis the concept of identity multiplicity to explain how the social identities of HIV -positive pregnant women are located at the intersection of gender and power relations, challenging their ability to seek HIV/AIDS trea tment within PMTCT programs (Sokefeld 1999). 9 mothers, wives, and patients . Identity Multiplicity as an Analytic Concept There is a growing body of literature on the social process by which individuals understand and act upon multiple identities (Sokefeld 1999 ; Stirratt et al. 2008 ). According to Sokefeld (1999), managing conflicting identities is an important aspect of the construction of the self. Sokefeld claims tha t this ability to manage multiple identities is a form of agency in which individuals employ strategies that allow them to choose how to act when faced with multiple alternatives. Drawing upon social psychological perspectives, Sokefeld contends that parti cular The notion of identity multiplicity has also been employed by feminist scholars applying the concept to illustrate the multiple forms of oppression linked to gende r, ethnicity, class, race, people must be located in terms of social structures that capture the power relations implied by those structures, and (c) ther e are unique, non -additive effects of identifying with more than one -532). Intersectionality theory posits that each identity (gender, race, sex, etc.) are given equal value in shaping experiences of oppressi on and shaping social identity. The concept of intersectionality suggests that social actors identifying with particular identities produce unique perspectives and consequences. Stirratt et al. (2008) argue that to better accommodate identity multiplicity, researchers must adopt methods that conceptualize individuals at intersections of multiple identities. -seeking at the 10 intersection of their identities as wife, mother, and pat ient. I argue further that the identity multiplicity faced by these women creates a double bind in which enacting each identity may applied the concept to the psychological stressors that contribute to the onset of schizophrenia in children. Bateson argued that competing demands from parents created behavioral dilemmas for children who were Others have drawn upon the concept to explain the double bind of biomedical training when medical students are trained to develop rapport with patients yet also trained to be objective clinicians. To my knowledge, researchers have not applied the idea of conflicting identities to the problem of barriers to medical care. In the following chapters, I discuss how access to HIV/AIDS treatment within PMTCT programs requires women to negotiate multiple conflicting identities that places them in a double bind, c reating barriers to medical care in these programs. Organization of the Dissertation The PMTCT clinic is a site in which limited resources constrain the services received by patients. As in other resource -depleted settings, patients often wait in queue fo r extended periods to be seen by a health care provider. For many women, attending the clinic is a form of forced disclosure in which the community first learns of their HIV/AIDS status. For patients who arrive late or have missed appointments, they may be subjected to the negative attitudes of some health care providers facing worker burnout as they manage larger patient loads with a shrinking health care staff. These experiences at the clinic clearly create barriers for women who wish to return to 11 the cli nic to receive the ARVs and reproductive health care provided in PMTCT programs. However, addressing these structural and programmatic constraints, largely shaped by political IDS treatment programs. To better understand why women have inconsistent participation in PMTCT programs, it is necessary to look outside the clinical context of the PMTCT clinic to the private sphere of ty, and the gender and power relations women must negotiate as HIV -positive mothers and wives. To understand the challenges women may face in negotiating an HIV identity in marriage, community, and clinic, I examine local understandings of HIV/AIDS. In thi s study community, representations of HIV/AIDS are linked to sexual practices that may challenge local notions of morality. Yet gender roles for women as mothers, wives, daughters in law, and community members suggest women fulfill the social role to repro duce, a social identity -seeking is located -positive identities as mothers and wives. Consequently, if a woman seeks care for herself and her child within PMTCT programs, she brings shame to and challenges the stability of her marriage. A gendering of stigma leads women to be blamed for being HIV - Health -seeking at PMTCT prog wives in conflict with each other, situating HIV -positive mothers outside the social categories of good mothers, faithful wives, and compliant patients. The positionality and socioeconomic status of the PMTCT patients in this study place them at risk of losing critical economic support to care for their HIV -exposed child, who is their 12 ultimate priority. To respond to these gender and power inequities, PMTCT patients employ strategies that lead to secrecy and inconsistent adherence as they negotiate multiple power struggles. Each of these encounters, with husbands, health workers, and community members, is nuanced by kinship and political economic structures, demonstrating the shifting nature of pow er -positive mothers and wives. Chapter 2: Literature Review In this chapter, I review the following literature relevant to my study: the development of biomedicine in Southern Africa; the impact of the HIV/AIDS epidemic on the national economy, health sector and households; the public health response to the epidemic the NGOs response and PMTCT scale - social stigma, motherhood, and gender relations. Th e first three sections provide a broad political economic, sociocultural and historical context in which to explore the gender and power relations that emerge in the social relations and health -seeking of HIV -positive pregnant women and new mothers. The fo identities and social relations in marriage, the community and clinic. I share a brief history of the impact of the HIV/AIDs epidemic on gender and community relations in the Malawian health sector to draw attention to the political economic factors gendered biomedical treatment of STDs and contemporary experiences with stigmatization and HIV/AIDS. 13 Chapter 3: Research Methods In this chapter, I describe my research site and the methods employed in the study. I conducted focus group discussions, in -depth interviews, clinical observations, and a quantitative survival analysis of patient retention within the first cohort of Plan B+ patients. Study participants included patients seek ing medical care within Prevention of Mother to Child Transmission (PMTCT) programs and the health care providers who care for them. Study participants also include community members living in Blantyre and surrounding communities. Focus groups and in -depth interviews were conducted with these community members teachers, church leaders, domestic workers, village chiefs, and traditional birth attendants to explore the impact of gender, kinship, and class relations on HIV - -seeking decisi ons and practices choices. Chapter 4: Local Understandings of Reproduction and the Construction of Motherhood In this chapter, I examine the local understandings of reproduction and motherhood in Malawian society by exploring the multiple actors with soci al, cultural, and political interest in society to illuminate the social, cultural, and economic significance of reproduction to the social identities of women who par ticipate in PMTCT programs. To better understand how reproduction is understood in the study communities, I examine how infertility affects women social relations in the family and community HIV/AIDS -related st igma. 14 Chapter 5: Gender Roles and Relations in Malawi and supports in Malawian I explore normative constructions of gender in Malawi and how these constructions are played out Finally, I discuss how 4) are challenged when they participate in PMTCT programs, which may expose infidelities in marriage when an HIV status is disclosed, thereby constrai wifehood. Chapter 6: Local Understandings of HIV/AIDs and Community -Based Stigma In this chapter, I analyze community constructions of HIV/AIDS, exploring the relationship between gender inequalities and local understandings of HIV/AIDS as they intersect experiences with HIV/AIDs stigma and their responses to stigmatizing attitudes and practices encountered in the household, kinship group, an d larger community. In my discussion of -related stigma, I explore how HIV/AIDs stigma Chapter 7: Dissertation Conclusions In this chapter, I bring together my argum ents in the previous chapters and make suggestions for future research. 15 Chapter 2: Literature Review health -seeking practice within PMTCT programs, followed by the theoretical framework that informs this literature. I then discuss the literature related to the impact of HIV/AIDS on broader political, economic, and sociocult ural context in which to explore the gender and power identities and HIV -related health -seeking within PMTCT programs. PMTCT Programs and Patient Adherence In 2012, 17.7 million women were living with HIV (World Health Organization 2014). Ninety percent of these women lived in sub -Saharan Africa. Malawi, a small country within the region, is especially impacted, with some of the highest rates of HIV infection in pregn ant women in the world. In 2010, 63,500 pregnant women in Malawi required antiretroviral (ARV) prophylaxis (UNICEF 2011). Of these women, only 53% received ARVs through PMTCT programs (UNICEF 2011). In 2011, in an effort to expand access to treatment servi ces within PMTCT programs, the Ministry of Health in Malawi implemented Plan B+, a policy designed to expand access to treatment services within PMTCT programs. Plan B+ now offers lifelong treatment to all pregnant women and new mothers who test positive f or the HIV virus, regardless of CD4 count. 16 Despite this substantial national scale -up of PMTCT services, retaining women within PMTCT programs remains a challenge to successful implementation of these interventions. Recent literature suggests that approxim ately 27% of women enrolled in PMTCT programs will - development of opportu nistic infections and resistance to medications. It is estimated that 75% adherence is necessary for ARV treatment to be effective (Nachega et al. 2012). While PMTCT adherence rates are now well -documented, reasons women may be unable to continue in the program are less well understood (Van der Straten et al. 2014). Researchers PMTCT programs. Below I review the social science literature on women and AIDs to sit uate HIV - -seeking and social relations within their broader sociocultural and political economic context. A meta - programs reveals that health system co nstraints, medication side effects, transportation, health care provider attitudes, stigma, and lack of social support may be important (Gourlay et al. 2013). Several studies point to the role of health system constraints (Balcha et al. 2011; Nkonki et al. 2007; Laher et al. 2012; Sprague et al. 2011; Whyte et al. 2004) or clinical hindrances such as side effects (Stinson et al. 2010) in creating barriers to participation. Other studies cited transportation as the most important factor ( ), or lack of education (Kuonza et al. 2010). et al. 2008; Castro and Farmer 2005; Chinkonde et al. 2009; Kasenga et al. 2010; Laher 2012; 17 Mepham et al. 2011; Nyblade et al. 2003; Stinson 2012; Theilgaard et al. 2011; Torpey et al. 2010; Ujiji et al. 2011; Van Hollen 2013; Van der Straten et al. 2014), negative interactions with health care providers (Nyirenda 2010; Painter et al. 2004) and fear of HIV/AIDS disclosure (Chinkonde et al. 2009; Van Hollen 2013) gender inequalities, which may prohibit women from disclosing an HIV/AIDS status to their family and community. In Malawi, an AIDS diagnosis for a woman may not only connote rooted in marriageability and fertility. The potential role of stigma has been noted by several researchers who suggest stigma -seeking (Bwiriire 2008; Castro and Farmer 2005; Chinkonde 2009; Kasenga 2010; Laher 2012; Mepham et al. 2011; Nyblade et al. 2003; Stinson 2012; Theilgaard et al. 2011; Torpey 2010; Ujiji 2011; Van Hollen 2013; Van der Straten 2014). Farmer (Castro and Farmer 2005) claims that stigmatization is a tool by which individuals leverage power differentials within the social hierarchy, reinforcing t he marginalized positions of the poor, women, and others most vulnerable. He calls upon scholars to reveal the lived experience of patients with AIDS and to explore community perceptions of AIDS to counter the detrimental effects of stigma. Government poli cymakers and donors have begun to knowledge of AIDS to communities with the goal of altering stigmatizing attitudes toward AIDS. Recent literature suggests a gendere d experience of HIV/AIDS stigma, claiming that women experience greater HIV -related stigma than men (Nyblade et al. 2003; Van Hollen 2013). 18 The studies point to the fact that an HIV/AIDS diagnosis disrupts family and kinship relations, which may manifest i n stigmatizing attitudes and practices toward women. Bond (Nyblade et al. 2003) found that women were more likely to be stigmatized for being HIV -positive because they failed to fulfill their roles as caregivers to members of the extended family because of their -positive mothers in India. Van Hollen (2013) found that women were stigmatized by mothers -in-law who blamed them for bringing HIV/AIDS into the marriage, thus threatening the moral integrity and continuity of the lineage. Some studies have identified signifiers of an AIDS status that women must hide from members of the community. For example, in many parts of Central and Southern Africa, a until the child is two years old signals to the community that she may be HIV -positive, and may be subjected to criticism from community members if her breastfeeding practice conflicts with local norms (Thorsen et al. 2008). In this study, women experience stigma in their social relations with husbands, community members and health care providers. the potential role of gender relations and inequalities. Further, few s tudies examine the impact of Studies exploring the role of gender relations and inequalities in HIV - domain of prevention (Baylies 2000; Boo th 2004; Tallis 2000). These studies largely focus on the role that gender relations and inequities play in heterosexual transmission. Studies focusing on Central and Southern Africa, for example, suggest that gender inequalities place women at risk for co ntracting HIV/AIDS (Bwirire et al. 2008; Hunter 2007; Iliffe 2006; Jewkes and 19 Morrell 2012; Schatz 2005; Susser 2009). These studies have shown that negotiating safe sex may lead to domestic violence (Jewkes and Morrell 2012), threaten local ideas of mascu linity The studies support the notion that gender inequities place women at higher risk of contracting HIV because women fear negotiating with husbands who may be unwilling to practice safe sex. For example, Chinkonde (2009) finds that women fear disclosing their HIV status because disclosure requires they negotiate condom use, creating tension in the marriage. Schatz (2005) found that women who failed to negoti ate safe sex simply opted for divorce rather than remain within relationships that placed th em at risk for contracting HIV. Booth (2004) highlights how development agendas assume gender equality, leaving health care workers to reinterpret and reformulate h ealth policy to fit with local realities. Booth found that nurses in AIDS prevention programs in Kenya decided to not distribute condoms to d ideologies rooted in biomedical training and/or positions of privilege may compel them to restructure medical protocols based on their own cultural assumptions (Booth 2004; Tsing 2005). While several studies examine the role of gender and risk of HIV/AID S transmission, few Moreover, fear of disclosure (Bwirire et al. 2008; Chin konde et al. 2009; Van Hollen 2013). Van Hollen -seeking within P 20 status as pregnant women and future mothers target them for reproductive health services prenatal or antenatal services which offer HIV testing to all pregnant women attending thes e services. As a result, many women receive an HIV/AIDS diagnosis prior to their husband simply because they attended the antenatal clinic and agree to be tested. Husbands, on the other hand, are less likely to be tested, either because they refuse, or hav e fewer opportunities to be tested discuss how this aspect of PMTCT services contributes to a gendering of stigma for women which reinforces existing gender and powe community. -seeking in Mozambique suggests that scarce resources, a pattern t -seeking within prenatal health services because of economic marginalization and competition for husbands whose economic power may buffer the effects of limited resources in central Mozambique (Chapman 2012, 238). Other studies conducted in rural Southern Malawi (Bwirire et al. 2008); and urban Central Malawi (Chinkonde et al. 2009) suggest a lack of support from husbands stemming fr om f gender relations and programs. 21 Theoretical Framework To better understand how gender roles and relations contribute to gender inequities, I deconstruct the broad concept of gender inequalities, drawing upon the work of scholars whose research examines gender and kinship relations within Southern Africa (Conroy 2013; Crehan 1997; Jewkes and Morrell 2012; Mohanty 2003; Poewe 1980; 1981). I analyze gender inequalitie s in terms of economic and political relations such as control over production, reproduction, political power, (Crehan 1997; Poewe 1980; 1981) and relationship power (Conroy 2013; Jewkes and Morrell 2012). In Chapter 5, I argue that gender inequalities for women participating in PMTCT are - HIV disclosure and PMTCT treatment -seeking challenging. I discuss power place women living with HIV at risk of divorce when they disclose a positive status to husbands and the community. focuses on how gender relations shape and are shaped by economic dependence. This dimension of relationship power demonstrates how gender inequities are produced by un equal access to education and employment, resulting in limited economic power within the household. I discuss shape gender inequalities within marriage. The social stru - -making power within the household (Crehan 1997; Poewe 1981). Decision -making dominance is 22 relations such as the negotiation of condom use and HIV disclosure to a spouse. This social structure helps explain why many poor women do not disclose their HIV status, perceiving few options for social and material security outside the marital relationship. The final soci ities each of these social contexts demonstrate how PMTCT health -seeking is located at the intersection of these power relations: economic status, little decision -making authority, and -seeking and within PMTCT progr ams. Rooted in political economic theory, and drawing upon notions of symbolic violence (Bourdieu and Passeron 2000) and normativity (Foucault 1978), structural vulnerability refers to indirect forms of violence (Quesada et. al. 2011, 1). Similar to the concept of structural violence, the notion employs a depoliticized tone while demanding social and political responses to unequal health outcomes. Structural vul nerability extends the concept of structural violence by capturing the multiple inequalities that lead to poor health outcomes. A key distinction between the two concepts is the emphasis upon social process in which positionality and power relations contri bute to social suffering. While structural violence denotes a static social structure, structural vulnerability 23 treatment programs in its broader political, economic, and sociocultural context, I discuss the impact of broader economic reforms and repercussions of the HIV/AIDS epidemic on resource constraints within the public health system. Impact of the HIV/AIDS Epidemic Impact on the National Health Sector The impact of the AIDS epidemic on health sectors within the Global South demonstrates the concept of biopower 2 (Foucault 1978), or the biosocial process by which the "power of biology" invites intervention int o the African continent and results in the creation of new social forms (Ferguson 1994 ; Lock and Nguyen 2010 ). The global response to the HIV/AIDS epidemic has fueled an emergence and application of biotechnologies to advance efforts in the prevention and treatment of AIDS (Biehl 2005) . In many southern African countries, continued high rates of AIDS combined with the social and economic fallout from structural adjustment policies (SAPs) and other political economic changes attracted an influx of non govern mental organizations (NGOs) and faith -based organizations to the region, providing health services while creating new opportunities for intervention in the lives of patients with AIDS. Trade liberalization and privatization have increased poverty, migratio n, low wages, and decrease in access to social services (Mohanty 2003). SAPs had a detrimental effect on state -sponsored health services, as downsized government programming conflicted with health policies positioned to "scale -up" health infrastructure. As a result, decentralization of health services and privatization largely restricted health care access for the poor (Pfeiffer and Chapman 2010). !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!#!Biopower, defined as the regulations of life at the population and individual levels, illustrates the productive nature of disciplinary power. Examples of biopower include the regulation o f fertility, sexuality, modern technologies. !24 I discuss how shifts in the national economy and a decline in spending for state -sponsored health services crea ted opportunities for a transnational response to the AIDS epidemic in Malawi, including the expansion of AIDS -related services. The process illustrates the role of political -economy in the national scale -up of prevention and treatment -based services. The process further illustrates the power differential between economic powers in the Global North WTO, IMF, the Global Fund, faith -based organizations, and transnational corporations at the center of antiretroviral distribution as the power of self -governance among poorer nations declines and that of transnational institutions is on the rise (Pfeiffer and Chapman 2010). These global political and economic processes associated with neoliberalism and capitalistic expansion exacerbate economic, racial, and gender inequalities in many southern African nations (Mohanty 2003). The mass distribution of AIDS prevention and treatment services such as PMTCT programs illustrates the decline in state power and authority in many African nations. As many African states attem pted to repay debts from SAPs, support for AIDS -related services such as the production and distribution of ARV medications ushered in an influx of donor -funded NGOs that assumed new forms of governance as they fulfilled policy obligations. In recent years , public health researchers have called for a greater emphasis on health systems strengthening by international donors. These researchers claim that donor aid tends to favor vertical, non -sustainable programs supported by international NGOs. Meanwhile, lit tle aid is funneled to ministries of health so that existing programs can be reinforced with additional unsustainable (Pfeiffer et al. 2008, 2137). Such projects cater to the goals of international donors 25 rather than the goals of local ministries of health. Consequently, these international projects create parallel progr ams that are ultimately not linked to existing local health services (Pfeiffer et al. 2008). NGOs and faith -based organizations operate as forms of biopower , responding to the health service gap with an extensive array of services dedicated to promote th e health and well -being of thousands of men, women and children living with AIDS (Farmer 1992; Millen et al. 2000). In recent years, AIDS health policy has shifted from prevention -based approaches to therapeutic approaches with the mass distribution of AID S medication. This shift in policy has created new global power relations opportunities for pharmaceutical companies now located at the intersection of downsized state -funded health services and transnational and national treatment -oriented health policies . As a result, the pharmaceutical industry and NGOs have played a larger role in governance, shaping national health policy as state governments attempt to minimize the impact of AIDS on individual patients and the larger body politic. In Chapter 4, I disc uss how resource participate in PMTCT programs. Just as NGOs have responded to service gaps with the provision of health and other services, pharmaceutical companies have gained entry to African markets through partnerships with transnational NGOs (Bornstein 2003). Pharmaceuticals such as ARVs provide a lens through which to observe the social processes by which structural inequalities are reproduced; who gets access t o pharmaceuticals, and who doesn't, often reflects social hierarchies within the larger society. The construction, distribution, and implementation of AIDS intervention strategies has been framed as a "biopolitical assemblage" of "discourses, bodies, drugs , and technologies (Nguyen 2004). Authors claim the circulation of pharmaceuticals provides insight into the intersection of science, the state, and private industry, linking the 26 value of commodities with the authoritative knowledge of medical experts (Ngu yen 2004). As state governments, donors and NGOs realign health policy strategy toward treatment -based methodologies, pharmaceutical companies have gained entry into a new market by increasing demands for antiretroviral medication. While authors remark on the potential contribution of lower -cost treatment, they also suggest the detrimental effects of transnational corporations (TNC) in shifting health policy away from the promotion of community health to more privatized individualized health strategies whic h rely heavily upon biotechnologies (Millen 2000, 220). Scholars suggest the effects of the "biotechnical response" to the AIDS epidemic distract from the underlying structural inequalities that have fueled AIDs. Drawing upon notion of the "critical e vent," biotechnical responses erase the suffering of individual bodies, and place the responsibility for health on the individual body or in "the power of biomedicine" such as pharmaceutical fixes while larger political economic forces remain unchanged. Impact of HIV/AIDS on Households As discussed earlier in the chapter, many of the economic reforms designed to promote economic growth have disproportionately affected the poor, particularly poor women and children. In Southern Africa, men and women have in creasingly engaged in informal economies, devising creative strategies to compensate for failed economies from economic reforms such as structural adjustment policies (SAPs) (Tranberg Hansen 2000; Kalipeni 2009; Schatz 2005, Hunter 2007). Scholars have inc reasingly drawn attention to the failure of the state to create viable jobs, which has contributed to migration and the formation of new sexual networks. In countries such as Malawi, d 27 (Mohanty 2003). HIV/AIDs primarily has affected the health of adults in their productive years, leading to dramatic economic and social changes in the struc ture of households. Family members who are too ill to work contribute less to the household economy through lost wages and are often no longer able to contribute to household food production. Furthermore, these individuals may need other family members to withdraw from the workforce to provide ongoing care and support. The loss of adults in their productive years changes the overall structure of social relations within families. For example, adult children who once provided for elderly parents and are no lo nger able to provide for them due to HIV/AIDS -related illness may need their elderly parents to assume caregiving responsibilities for their children, once they have passed away or are unable to care for their children independently (Drimie 2002). In addit ion to its impact on traditional support systems, the HIV/AIDS epidemic has had a significant impact on food security at the household and national level. Households who lose the productive labor of family members too ill to contribute to agricultural prod uction may shift the type of crop cultivated from cash crops such as maize and tobacco to food crops such as cassava because these food crops require less intensive manual labor (Dorward and Mwale 2011). The economic impact of HIV/AIDS not only affects the economic well -being of HIV/AIDS households. The decline in household income negatively affects the broader macroeconomy as households consume fewer goods with reduced incomes (Bryceson and Fonseca 2006.) Consequently, reductions in household income are mo st severe in poor households. Research suggests that the extent to which HIV/AIDS impacts wages depends on 28 whether the work of the deceased family member is skilled or unskilled. When wages for unskilled labor decline, inequalities within rural communities are exacerbated (Dorward and Mwale 2011). Early in the HIV/AIDS epidemic, a series of political economic factors intersected, leading to acute food insecurity. Low agricultural productivity, high rural population density, land shortage, and the removal of fertilizer subsidies, collectively threatened subsistence food production in many communities and intensified ganyu labor so that families could meet their food needs. The HIV/AIDS epidemic played a significant role in the increase in labor migration as large numbers of individuals sought ganyu labor. This migration was primarily assumed by male members of the family, leaving many households as female -headed for extended periods of time. rise in the informal sector. In recent years, the donor climate has shifted from traditional donors from the Global North to those from the Global South. For example, India, China, Brazil, and South Africa have recently become enga ged in the development of agricultural, health and other sectors of many African countries (Kragelund 2010). Poverty and Global Inequalities Intersections with AIDS The ideologies and practices constituting the AIDS epidemic provide a lens through which to view developing AIDS. Political economic inequalities and relations of power characterize the circumstances that have fueled the AIDS epidemic in southern Africa. The se same inequalities 29 HIV/AIDS long after they have initiated treatment. A number of scholars argue that the structural inequalities that led to the growth of the H economic processes that reproduce poverty (Quesada, Hart and Bourgois 2011 ). For example, structural inequalities linked to poverty, gender, and racism often d etermine who is at risk for AIDS and who gets access to treatment for AIDS (Desai and Johnson 2005; Farmer 2005; N guyen 2004; Petros et al. 2006; Schatz 2005; Tallis 2000). Hunter (2007) argues that structural inequalities play an important role in the gro wth of the HIV/AIDS epidemic, and calls upon scholars to deemphasize the often -belabored connection between AIDS and sexuality (Castro and Farmer 2005; Whyte et al. 2004) in public health and anthropological discourse, which obscures the role of poverty an d unemployment in the AIDS epidemic. Many women have responded to the effects of structural violence by creating informal economies to supplement limited incomes. In urban South Africa women generate income through selling food, selling secondhand clothing , brewing beer, and providing sex work ( Hunter 2007) . As women attempt to compensate for limited incomes, some women are placed at increased risk for contracting AIDS and experiencing sexual abuse when earning additional income through sexual networks. Und eremployment triggers a trajectory of poverty and domestic violence, causing women to lose their marriages and economic stability. AIDS places women at increased risk of exploitation by employers who deploy power relations to take advantage of young women seeking to secure livelihoods. Kalipeni (2009) explores how women navigate the structural factors that contribute to AIDS, claiming that globalization has fueled inequalities that put women at greater risk of contracting the disease. For example, in Malawi , AIDS 30 deaths have transformed the structure of families as orphaned children relocate with members of the kinship network or fend for themselves through alternative strategies. Young girls who seek employment through domestic labor are placed at risk of s ex abuse as economic dependence binds them to unsafe work, while others exert agency and construct new identities in sexual networks that draw upon the economic prowess of sugar daddies (Cole and Thomas 2009). Kalipeni (2009) found that women who lost fami ly members to AIDS and sought work as domestic workers were more vulnerable to exploitation by their female and male employers, experiencing verbal and sexual abuse from female and male employers, respectively. Some women who have lost productive members o f the family may seek additional income through transactional sex, or the exchange of sexual relations for cash income. While women may supplement household income with transactional sex, men and women may seek additional income through ganyu , or the excha nge of day labor for cash income. Gender roles and relations often confer caregiving responsibilities to women and girls. Women play an important role in agriculture and food security, however, they are often pulled from these forms of productive labor to provide caregiving for relatives with HIV/AIDS (Drimie 2002). Similarly, girls are more likely to be withdrawn from school to assist with childcare for younger siblings and other domestic responsibilities. Such losses in education potential have long-term consequences on the earning potential of AIDS orphans. These changes in caregiving responsibilities have weakened traditional support systems (Drimie 2002). Motherhood is central to the identity of women living in Southern Africa (Walker 1995). focused on HIV - 31 status leads women to mini mize the number of pregnancies they choose to have. Employing the metaphor of the infecting body, Long reports that HIV -positive South African mothers fear HIV transmission to the unborn fetus. These women also fear being unable to care for their HIV -posit ive infant because of their illness (Long 2009; Van Hollen 2013. In contrast, Yeatman and Trinitapoli (2013) find that fear of mother -to-child transmission does not influence Malawian oose to continue Importantly, this 2013 study was conducted after the PMTCT expansion, specifically after Option B+ implementation, which initiates mothers o n ARVs immediately after diagnosis, and anticipates significantly reduced MTCT and fewer HIV -positive infants (Fasawe et al. 2013). example, some studies examine whether HIV -positive women are blamed for being poor mothers (Liamputtong 2013; Long 2009). These studies examine how placing a child at risk for mother -to-child transmission may be understood by members of the community as a failure to fulfill the caretaking role lin childbirth and HIV and the challenges of being both pregnant and HIV -positive during Option B+ where fewer infants are contracting HIV/AIDS through mother to child transmission. Th e -seeking and their intersection with HIV/AIDS stigma has been largely unexplored. Below I explore the existing literature related to gender and HIV/AIDS to illustrate this gap in the literature. Develop ment of Biomedical Health Care in Colonial Africa 32 processes: the development of biomedical health care in Southern Africa and the social construction of population growth in the imperial and colonial eras. The development of biomedical health care in Southern Africa was based on a set of ideologies and practices that crystallized around the domain of population health as a marker of the health of imperial nations and their col onial states. The early 1900's was characterized by a period in which the body, women's sexuality in particular, became of central concern to the colonial state and mission doctors. A global concern with fertility emerged in Europe as infectious disease epidemics caused a decline in the European population. At the same time, expansions in scientific knowledge and the emergence of the field of statistics facilitated state monitoring and assessment of population health. As a symbol of the health of population s, fertility has historically been a key interest to nation states whose interests center on productivity and growth. During the imperial era, birth rates marked the economic potency of nation states, allowing states to statistically track the success of n ation building. While concerns with underpopulation plagued the European metropole, low levels of fertility within many African colonies prompted concerns that African colonies would be unable to sustain a potent labor force to support a growing economy. A s a result, European states exerted increasing control over African colonies in an effort to reflect idealized representations of a productive nation state (Vaughan 1991). To address the concern of underpopulation, state and biomedical actors turned their attention to the female body as the sites of reproduction and production. Development of Western Biomedical System Globally, the growth of biomedicine is marked by four phases: imperial [ -1920], colonial [1920 -1960], nationalist [1960 -1980], and non -gover nmental [1980 -present] (Lock and Nguyen 33 2010, 148). Biomedicine gained prominence when clinical success with infectious disease control bolstered the credibility of western biomedicine in Europe. Vaccination campaigns, the development of antibiotics, and t he development of hygienic practices mitigated against illness and mortality from infectious disease epidemics within European nations and their colonial conquests. After World War I, the ideology and practices associated with population growth became of increasing global concern as nation states feared the decline of European populations. At this time, birth rates became a proxy for the health of nations and the pronatalist policies designed to increase birth rates evoked a medicalization of childbirth, wi intervention (Lock and Nguyen 2010, 158). In Central, Western, and East African colonies, concerns with infertility fueled fears of an promoted safe childbirth populations. Such civilizing discourse characterized the transnational focus on fertility within African colonies (Tsing 2005). This discourse further served the political and economic needs of nation states seeking a viable African labor pool to further their economic interests. State interest in birth rates fueled colonial and mission health policy and the transnational growth of the maternal a nd child health movement (Vaughan 1991; Stoler 1995). In the former colonies, rules and proscriptions related to maternal and child health and childrearing became central to colonial health policies, leading to the formation of the maternal and child healt h movement. Family life, with an emphasis on mothering practices, was carefully monitored as the state sought to shape its future citizens within the European metropole, and a viable labor force in its African colonies. According to mission and colonial ad ministrative authorities, mothers were constructed as the source of morality within the family. Aligned with the missionizing goals of constructing the 34 African family, mission doctors reinforced constructions of mothers whose moralities evoked "pro per" and "hygienic" practices. Such constructions were commodified in products promoting personal hygiene, such as soaps and lotions aligning consumers with modernist identities (Burke 1996). Conceptions of the ideal family, however, were rooted in Western European notions of the nuclear family, which paid little attention to the role of kinship and extended family within African society. The institutions and practices that constituted mission reform of maternal and child health instilled values and practic es associated with motherhood that failed to incorporate larger networks of social relations critical to enacting motherhood in Africa (Vaughan 1991). These proscriptions for future mothers instead espoused micro -level changes at the household level in the values and practices of women. As a result, women were assigned responsibility for conforming to normative ideals rooted in Western models of individualism. notions of -seeking as mothers and wives is conducted independent of their social relations with spouses, kinship group, and community. In the 1930s, African child rearing and feeding were monitored and managed to cultivate new subjectivities grounded in moral principles central to the ideology of Christianity. Mission doctors monitored and transformed child rearing and marital practices through discourse that incited social anxieties about the "difficult child" and "laxity in marriage." Infant feeding became central to this project as mothers were trained how to properly feed their infants to cultivate "proper character" in their children (Vaughan 1991, 67). Stoler (1995) argues that ideologies related to childhood socializati on reinforce racial boundaries and structure the development of future citizens by linking everyday practices such as bathing, feeding, and breastfeeding to anxieties over personhood, race, and national identity . Coining the concept of In Chapter 6, I discuss how a moral discourse 35 related to sexuality and fidelity is central to concepts of wifehood, contributing to st igma toward women living with HIV/AIDS. I show how these historical processes the production of an African contrast with historical and contemporary understandings of motherhood and wifehood. I find that the tension between these identities produce unequal power relations and community -based stigma toward mothers living with HIV/AIDS. replaced with concerns o Nguyen 2010, 160). Population control and family planning became central to government economic policy, resulting in the family planning movement which linked a nation's prosperity and sta bility with the reproductive capabilities of its families (Lock and Nguyen 2010, 121). Production of Sexuality in the Colonial Metropole In the British metropole, the threat of underpopulation fueled a "pronatalist discourse" dedicated to population expan sion. Pronatalist ideologies reflected racist sentiment as concerns over "dying" races intensified. Campaigns to promote white prestige led to the mobilization of the Eugenics movement, a scientific project grounded in racist ideology which sought to prese rve the purity of the European "stock" in the context of the profound economic and social change accompanying industrialization (Stoler 2002, 120). As symbols of reproduction and production, women and children were depicted as bodies particularly vulnerabl e to "degradation." Discourse on the "degradation" of bodies and populations led to new constructions of the sexuality of working class women, linking sexuality with racial 36 degeneration as the British colonial elite sought to cultivate a population that me t the needs of the political and economic interests of the imperial state. In the former African colonies, such concerns with sexuality were equally embedded within racist discourse on relations between black men and white women. European officials and cit izens threatened black men against engaging in sexual relations with European women. Assumptions about the sexuality of black men were rooted in racist fears culminating in the "Black Peril," a systematic attempt to deploy racial categories to establish so cial boundaries between white men and black women. The Black Peril parallels the anxieties manifested in the discourse on African sexuality. In Colonial Africa, the Zimbabwean stat e institutionalized similar racial divisions Institutionalized forms of racism continued to emerge as African men were charged with rape when their proximity to European wo men threatened racial boundaries. Invoking the concept of the social body, Setel (Setel et al. 1999) claims such attacks on the individual bodies of white women signified an attack on the larger white community. Representations of African sexuality further operated as a strategy for maintaining political and social control of the colonial empire. Following Stoler's work on racist sentiment in the ideology and practices of the former Dutch colonies, allegations of sexual indiscretion facilitated control over the behaviors and practices of certain populations, demonstrating the processes by which sexuality operates both as a metaphor for domination as well as a marker of race and class (Stoler 2002, 45). Health -seeking for the treatment of sexually transmitted diseases was a humiliating and shameful process in which both men and women were ridiculed by colonial authorities for roduced in the attitudes and practices of 37 colonial health officials who attributed this health - 148). In Chapter 6, I demonstrate that many of these attitudes persist today in the gendering of stigma experienced by PMTCT patients as mothers and wives living with HIV/AIDS. I discuss ho rooted in the history of STDs in colonial Malawi (Vaughan 1991). I argue that manyazi (shame) has been reproduced discursively and is well alive today in the treatme nt seeking of these women living with HIV/AIDs. Conclusion In this chapter, I discuss several key domains in the literature on women and HIV/AIDS that explain the barriers and supports, grounded in gender and power relations, that shape -seeking within PMTCT programs. I provide a review of the current literature related to studies of patient participation in PMTCT programs. I share a brief history of the impact of the HIV/AID s epidemic on the exploring the relationship between gender and HIV/AIDS tra nsmission, and the socioeconomic 38 Africa to illustrate the historical roots of a gendered access to AIDS treatment, drawing a link bet experiences with stigmatization and HIV/AIDS. I then include literature on motherhood and gender relations to foreground my discussion on the power inequities PMTCT patien ts encounter in marriage, the community, and clinic. In the following chapter, I discuss the methodology used to explore the main research questions linked to this literature. 39 Chapter 3: Research Methods This research study was conducted at Queen Elizabe th Central Hospital, surrounding health centers and several community locations within the Blantyre district of Southern Malawi. The study included both qualitative and quantitative methodology. The qualitative methods employed during the study include sem i-structured interviews, focus group discussions, and clinical observations. A quantitative program survival analysis was conducted at the end of the study period at one of the study sites, Queen Elizabeth Central Hospital. This analysis is included in the Appendices section of the dissertation. The study participants included patients attending PMTCT programs, health care providers providing care in these programs, and community members from the surrounding Blantyre district. This chapter begins with a dis cussion of the research questions that guided this study and summarizes the methods used to attempt to answer those questions. This discussion is followed by a description of the study sites divided by region, and provides a more detailed discussion of the research methods in each of the main three geographic areas studied here: Blantyre, Ndirande and Mpemba. Blantyre is a densely populated city in the Southern region. Ndirande is an urban township outside of Blantyre, and Mpemba is a peri -urban township in the Blantyre district. Data was collected at health facilities and community settings. I discuss additional research methods used in the study in the final section of this chapter. Part I: Study Objectives and Research Questions While conducting explorat ory research to determine my dissertation topic, I spoke with 40 health care providers working directly with AIDS patients in the Malawian government -run health system. I learned that one of the challenges with the clinical management of AIDS was clinics, where rates of non -adherence were twice that of the general adult HIV/AIDS population. o consistently access medication for themselves and their HIV -exposed children, when ARV treatment was now available at no cost after years of being accessible only to those who could afford it. The overall goal of the study was to identify factors that af participate in PMTCT programs in Malawi. economy discussed in Chapter 2 informed the overarching research question: Why do only some women in PMTCT programs complete treatment? More specific questions included the following: What strategies for AIDS treatment and management do pregnant and postpartum program? gender relations? Part II: Data Collection Study Setting Between March 7 and December 5, 2012, ethnographic interviews, focus group discussions, and clinical observations were conducted at the health facilities and community settings described above: Blantyre, Ndirande, and Mpemba . I spent the first five months at QECH (Queen Elizabeth Central Hospital), followed by six months at two health centers. Patient 41 interviews and focus group discussions were conducted in Chichewa by two research assistants: a nurse midwife from QECH, and a Health Services Assistant (HSA) from one of the participating rural clinics. The nurse midwife conducted all interviews with current PMTCT patients a t QECH and Ndirande, and the HSA independently conducted interviews with all and attended all patient interviews except those interviews conducted with patien The study was approved by the institutional review boards at Michigan State University and the College of Medicine at the University of Malawi. Interviews lasted between forty minutes and one hour, and all interviews were audio -taped. Two h ealth centers in addition to QECH, Ndirande Health Centre and Mpemba Health Centre were included in the sample to capture a sample of PMTCT patients who are referred back to health centers after the standard six month follow -up appointment and to ensure representation from both peri -urban and urban communities. I selected these particular health centers for several reasons. First, each had a consistent patient load to guarantee I would be able to recruit patients for interviews. I collected statistics on th e number of patients attending PMTCT clinics from the District Health Office in Blantyre to identify clinics with the highest patient load, and health centers that received few PMTCT patients were not considered. Furthermore, one of the clinics, Mpemba Hea lth Centre, was willing to allow me to volunteer to assist with administrative tasks during the weekly adult antiretroviral ART/PMTCT clinics. Finally, I wanted to include in the sample satellite health centers, as many patients receiving HIV/AIDS services travel back and forth between Blantyre and the surrounding communities. 42 Blantyre Blantyre is the second largest city in Malawi, and is located in Southern Malawi. With a population over 700,000, Blantyre is the commercial capital in Southern Malawi. Bla ntyre is located in the Shire highlands at an elevation of 3,000 meters, and is surrounded by several mountain ranges: Michiru, Soche, Chiradzulu, and Ndirande. The largest hospital in Malawi Queen Elizabeth Central Hospital (QECH) is located in Blantyre, serving a patient population of 1,300 patients at a time, and over 100,000 children a year. Several factors aligned to make QECH and the surrounding health centers an ideal location for the study. The city of Blantyre has some of the highest rates of HIV/A IDS in Malawi, with an average rate of 20.5% (Ministry of Health 2008). Since Blantyre functions as the commercial capital, QECH attracts a diverse patient population representing a cross -section of urban, semi -urban, and rural patients. QECH is a major pu blicly -funded tertiary -level teaching hospital and is a primary referral facility specializing in neurosurgery, orthopedics, radiology and other subspecialties (Muula 2003). The PMTCT program at QECH is based out of the prenatal clinic within the Departmen t of Obstetrics and Gynecology (OBSGYN). The department includes on average eleven physicians including six pediatricians, three -to-five physician interns, three midwives, and three -to-five nurses. The Chatinkha Maternity Unit housed within OBSGYN offers p renatal care to approximately 1,300 women per year (Changole et al. 2010). Mpemba The majority of this five -month research period was spent at Mpemba Health Health Centre for reasons described later in this chapter. Mpemba Health Centre is a peri -urban he alth 43 center approximately ten kilometers from Blantyre. This clinic served all patients with HIV/AIDs, both men and women, in a combined ART/PMTCT clinic. Ndirande Ndirande Health Centre is an urban health center located in a high density township approxi mately two kilometers from central Blantyre. Summary of Methods Used Clinical Observations In order to fully understand the nature of the PMTCT program and patient experiences in the program, I conducted clinical observations at each participating clinic. I observed the interaction between PMTCT staff and the women who attend the programs, permitting comparison between patient self -reports and their observed behaviors. While conducting participant observation, I noted the general atmosphere of the clinic, exploring attitudes women health education sessions for PMTCT patients to o bserve patient -provider interactions, and to listen to the questions asked by women to better understand which health information elicited clarification. Finally, I observed whether women attended the clinic with a self -appointed guardian as suggested by P MTCT protocol, and whom the patient chose to be their guardian. medication management and encourage their adherence. Most clinics dismiss this guideline as 44 not relevant, a nd I did not observe health care providers denying treatment to women who failed to attend with a guardian. Clinical observations were conducted at the district hospital during the weekly prenatal and postnatal clinics from which patients were recruited. T he prenatal clinic was typically bustling with mothers and infants waiting for the health education sessions. Once registered, women waited in line for themselves and their infants to be weighed. Prenatal and postnatal clinics began with health education s ongs by Mothers to Mothers, a peer -led support group for women living with HIV/AIDS (WLWA). PMTCT nurses would conduct health education sessions as women gathered around a small courtyard. After the group health education (IEC) session, health care provide rs and clinicians conducted clinical exams with mothers and their infants in one of the surrounding private rooms. Women registered in PMTCT would then be called into a private room dedicated to the distribution of ARVs. Three small clinical exam rooms surrounded a large room lined with wooden benches which served as the patient waiting area, prenatal clinic and postnatal clinic. Clinical observations were also conducted at participating health centers. At Mpemba Health Centre, I provided administrative sup port during weekly PMTCT clinics to observe the interactions between health care providers and patients. After volunteering for a period of months, I came to know the clinic staff well, particularly nurse midwives and health services assistants, and health center staff treated me as part of the staff. I weighed adults and babies to calculate BMI, filed medical records, counted pills, and distributed Plumpy Nut, the nutritional supplement for AIDS patients with acute malnutrition. 45 Focus Group Discussions Patients In addition to semi -structured interviews, four focus group discussions were conducted with patients. The goal of the patient focus groups was to examine the relevance of specific research questions for PMTCT patients, to focus clinical observation s and patient interviews, and to allow women to discuss potential facilitators and barriers to completing PMTCT treatment. Patient focus groups reached saturation at the fourth group discussion. Focus groups consisted of seven -to-ten current PMTCT mothers initiated on the 5A regimen. Early in the study period, it became clear that women were more forthcoming in expressing their opinions in the presence of other PMTCT participants. The social context of focus group discussions allowed for discussion of inter view questions in greater depth without the pressure of a one -on-one, semi -structured interview. To capture this greater depth in participant responses, a larger number of focus group discussions were conducted than had been originally planned. Patient foc us group discussions were conducted by the nurse research assistant in Chichewa, and I was present during all the focus group discussions. It is likely that given the sensitivity of the research topic, the anonymity granted within a group context allowed w more candidly expressed both positive and negative aspects of their PMTCT program experiences. Further, during the focus group discussion women appeared to develop a se nse of community. Several women cultivated new alliances with other focus group discussion (FGD) participants sharing their identity as HIV -positive pregnant women or new mothers. Focus Group Discussions Community Members To gain a more in -depth understan ding of the broader context of the lives of Malawian 46 women outside of the PMTCT clinic, semi -structured interviews and focus groups were conducted with community members within Blantyre and the surrounding community setting. Community members included teac hers, church leaders, domestic workers, village chiefs, and traditional birth attendants (TBAs). Community interviews and focus groups provided the opportunity to gain more information on the perspectives of women in the community in order to understand th eir everyday lives and responsibilities related to pregnancy, motherhood, and spousal and family relations. Attention was paid to common responsibilities of motherhood, within the community. The goal of these interviews was to capture the day -to-day experiences of women who may typically attend a PMTCT program. Interview questions explored the structure lial relationships and the role of motherhood. A total of six focus group discussions were conducted with teachers, church leaders, and male and female domestic workers. One of the teacher focus groups consisted of all women while the other focus group inc luded both men and women. In order to find enough teachers to comprise a focus group at the second school, both men and women were included. The aim of values, ideals, and practices imparted to women in PMTCT program guidelines may or may not Focus group discussions were held with primary and secondary school teachers, both male and female, in the study communities. Teachers were selected because as leaders in their communities they would be able to provide in -depth information about gender and kinship relations, local understandings of HIV/AIDs, and the responsibilities and experienc es of men and 47 women within the household, kinship group and community. Questions explored the roles of husbands, wives, mothers and fathers within the community, and also examined the meanings of marriage, divorce, pregnancy, and fertility in order to unde rstand community constructions of PMTCT programs. I first sought permission from the headmaster of the school to interview six to eight teachers. Teachers were selected based upon their willingness to be interviewed and their availability, and the permission to conduct focus groups was given by the headmaster on days when the school was fully staffed . Establishing rapport with the teachers was relatively easy. Many of the teachers seemed to appreciate my interest in their perspectives on gender and family relations in their communities. Moreover, in contrast to interviews and focus groups with women with AIDS, there appeared to be a significantly smaller power differential between myself and the teachers. I chose to interview teachers because as key socializing agents in their communities, teachers articulate key values and ideals of their community, convey them to students, and thus reproduce those ideals for the larger co mmunity. Patient Interviews A total of 71 patients were included in the patient sample. All patients were purposively oncepts and goals regarding HIV/AIDs, their experiences in the program, the strategies they employ, the challenges they may face, and how AIDS status and management of AIDS for themselves and their children; which aspects of 48 PMTCT programs were particularly challenging; what material, social, or religious strategies women employ as program participants; and who, in their social network, knows about their AIDS status a nd how they respond to this knowledge. Information was also gathered on patient demographics including age, marital status, number of children, number of household members, employment, and religious affiliation. Plan B+ patients are prescribed medication r egimen 5A immediately after testing HIV positive, regardless of their CD 4 count. (CD 4 cells are a type of white blood cell that protects immune system is responding to the HIV infection.) The 5A regimen includes the medications Tenofovir, Lamivudine, and Efavirenz. Provider Interviews A total of eighteen in -depth interviews were conducted with health care providers, consisting of nurse midwives, clinicians, PMTCT co ordinators, and health services assistants (HSAs). Provider questions focused on the following topics: 1) perceptions of why some women are unable to complete the PMTCT program and why others successfully complete; 2) what is required of women as participa nts in this program; 3) what providers perceive as the responsibilities of women in PMTCT programs; 4) what strategies providers employ to promote completion; 5) how providers perceive successful program completion; 6) how these perceptions are conveyed to women as prescriptions for a treatment regimen; and 7) what recommendations providers have to improve program completion rates. These questions were designed to reveal overlap. 49 Community Interviews A total of six semi -structured interviews were conducted with village chiefs, traditional birth attendants (TBAs), and church leaders. As noted above, I interviewed two church counselors from the Catholic community and the Ch urch of Central Africa Presbyterian (CCAP) community. Church leaders were included in the sample because of the prominent role that religion plays in the daily lives of most Malawians, through which churches are powerful institutions within the community a nd play a significant role in imparting values and beliefs kinship group, and community. I interviewed two traditional birth attendants (TBAs) in the peri -urban Mpemba region who were recruited by snowball sampling. The HSA asked the village chief in Mpemba for the names of TBAs in the village. These interviews were conducted with simultaneous translation by my research assistant, and both TBAs were interviewed at their homes in one of the surrounding villages. TBAs were included because of their pivotal role in the reproductive health of women living in rural Malawi. Eighty - roductive health practices during pregnancy and delivery two Patient demographics I interviewed and conducted focus groups with a total of 71 patients between the ages of 18 an d 49. The majority of women (75%) were between the ages of 26 and 40, 84% of the women were married, one woman was divorced and two women were widows. 50 Thirty percent had some secondary education, and 30% had completed only primary education. Most of the wo men attending the peri -urban PMTCT clinic had attended but not understand their ability to gain access to employment and other sources of economic support, as these factors may affect their ability to participate in PMTCT programs. or secretaries (13%), as small business vendors selling market produce (28%), or housewives who focu sed on domestic responsibilities of childcare, cooking, cleaning, water hauling, and collecting firewood (31%). The majority of women who farmed subsistently lived in Mpemba. One woman sold charcoal prior to her illness, another worked as a sex worker. Gua rdians were interviewed on behalf of each of these women. The mother of the woman working as a sex worker participated in the interview. For the woman who sold charcoal, her mother completed two interviews after the daughter died suddenly before the end of the study period. Most PMTCT patients were married (83%). One woman was divorced and two women were widows. The majority of patient informants from both urban and rural sites were of matrilineal descent (75%), whereas 25% were of patrilineal heritage. Alt hough, as noted earlier, these descent and residence patterns were significantly altered in this urban context. The most common ethnic affiliations were Lomwe (n=16) and Ngoni (n=13). The second yanga, four were Chewa, and and community -based stigma were found not to vary by ethnic affiliation. Women who reported 51 challenges to decision -making authority in mar riage, and those who reported shared decision -making power in marriage represented the same ethnic groups Lomwe, Chewa, and Ngoni. However, a few health care providers and community members claimed that in some social contexts, patrilineal kinship groups l example, community members and health care providers claimed women in patrilineal kinship groups experience greater stigma, arguing that these women were more likely to suffer from discrimination from mot hers -in-law who may ill -treat them when they are unable to bear children or are sick. As I discuss in Chapter 4, health care providers and community members report that women from patrilineal communities experience greater stigma when unable to reproduce. Pressure from in -laws is most pronounced in patrilineal kinship systems where the social and material value of children weighs more heavily. Contextualizing Kinship Demographics While the majority of the men and women in this study come from matrilineal k inship groups, many of the gender relations described by informants reflect a shift in matrilineal kinship structures, where vestiges of the matrilineal system persist in the context of urbanization which has altered the practice of matriliny. The majority of the women interviewed in this study came from urban and peri -urban areas in the Blantyre district of Southern Malawi, where the matrilineal system and residence patterns have been changing over time (Phiri 1983; Desai and Johnson 2005) and matrilineal practices have broken down, thus reshaping kinship relations. Phiri notes that the Chewa family system has been greatly affected by modern socioeconomic changes from colonialism and the capitalist economy which have shifted the structure of families in Mal awi. He states, 52 Malawi today does the mother's brother exercise the kind of domestic authority he may once (Phiri 1983, 257). A stronger relationship between wife an d spouse delegitimized (Phiri 1983), who traditionally held primary responsibility for the wellbeing of his sister and his altering a primary source of material and social sup port available to women. As Phiri states: In its ideal form, the Chewa family system is supposed to place considerable emphasis on the mother right, the avunculate, uxorilocal marriage, husband's subordination to wife's kin, and importance of female child ren as future reproducers of the lineage. In practice, however, there are a number of factors which have tended to mitigate the impact of these tendencies in the system, as far as men or husbands are concerned. These changes include changes in the marriage contract, family residential patterns, exercise of domestic authority, and control or custody of children. (Phiri 1983, 274) My data demonstrate the malleability of matrilineal kinship structures, and illustrate how matriliny is functioning in peri -urban areas of Blantyre, where men and women have fewer opportunities to earn a living, particularly for those with little education. Consequently, these economic and social support. The research methodology used at the three study sites, Blantyre, Ndirande, and Mpemba, are discussed in more detail in the next section. Study Sites and Methods Blantyre Patients Twelve interviews were conducted with current PMTCT participants at Queen Elizabeth 53 Central Hospital (QECH). All participants were purposively sampled. All patients interviewed were selected from the first cohort of PMTCT patients registered on Plan B+. I attended weekly prenatal and postnatal clinics with my research a ssistant to solicit experiences with the PMTCT program. Those women who agreed to participate met the research team in a private room after receiving their in itial check - confirmed for Plan B+ participation after they agreed to participate. Four out of 45 patients declined to participate. Each participant received a 2kg bag of flour or 1kg bag of sugar for their participation. Preg nant women met inclusion criteria if they were 18 -49 years old, were HIV positive, and were initiated on the Plan B+ 5A medication regimen which began in July, 2011. 5A treatment was initiated at study sites after October 3, 2011. Women were asked why they had initiated treatment for themselves and for their child; household and familial responsibilities; challenges and supports to AIDS management; thoughts on why some women are unable to adhere; whether they had ever missed two or more consecutive PMTCT ap pointments and the reasons for doing so; experiences with HIV -related stigma; decision -making roles in the family/household; and reproductive decision -making and practice, including intended length of breastfeeding period. A total of three focus group disc ussions were conducted with current PMTCT participants within the prenatal and postnatal clinics at QECH. I attended weekly prenatal and postnatal clinics to solicit volunteers. Each woman registered at the clinics was invited to participate in a study on 54 participate met the research team in a private room after they received their check -up. Focus group discussions began once a minimum of seven women volunteered. Health Care Providers Eleven in -dept h interviews were conducted with health care providers at QECH. Interviews were carried out with nurse midwives, clinicians, PMTCT coordinators, and health services assistants (HSAs). Nine informants were female and two informants were male. To recruit hea lth care providers, I applied a purposive sampling strategy (Marshall and Rossman 2006) and relied upon networking through my contacts from my initial pre -dissertation research conducted in 2010. Shortly after I received College of Medicine Research and Et hics Committee (COMREC) approval, I was introduced to the head of the PMTCT clinic at QECH. The PMTCT clinic manager then introduced me to PMTCT staff who were interested in participating in interviews. Community Leaders In Blantyre, I conducted one focus group discussion with teachers at a local secondary school. This discussion included a convenience sample of men and women who were working as full-time teachers and available to participate in the focus group during their lunch hour. I also conducted sem i-structured interviews with two church leaders, one from the Catholic Diocese and the Church of Central Africa Presbyterian (CCAP) church. Church leaders were selected through the process of snowball sampling. At one of the Catholic churches I attended du ring the fieldwork period, I asked for the name of the Church Counselor as I thought 55 lives. CI Montford is the largest Catholic church in the Blantyre district, able to seat 2,500 parishioners. Sunday morning services include one service conducted in English followed by a service conducted in Chichewa. I conducted one focus group discussion with female domestic workers and one focus group discussion with male dom estic workers. These included a convenience sample of men and women who were employed at a nearby lodge as housekeepers, gardeners, and guards. Ndirande Four interviews were conducted with current PMTCT participants at Ndirande Health Centre. Patients wer e selected following the methodology employed at QECH. Three in -depth interviews were conducted with health care providers at Ndirande Health Centre. These included a convenience sample of midwives, clinicians and health services assistants (HSAs). Two hea lth care providers were female, one was male. In the Ndirande community, I conducted one focus group discussion with teachers at a public secondary school. These included a convenience sample of men and women who were working as full -time teachers. Mpemba Patients I conducted one focus group discussion with current PMTCT participants within the prenatal and postnatal clinics at Mpemba Health Centre. My research assistant and I attended weekly prenatal and postnatal clinics to solicit volunteers. Each woman registered at the clinics 56 agreed to participate met the research team in a private room after they received their check -up. Focus group discussions began once a minim um of seven women volunteered. I attended the HIV/AIDS clinic in Mpemba for a period of five months. I spent an extended period of time in Mpemba because it became an ideal place in which to conduct participant observation. I was unable to conduct particip ation at the district hospital because there was little need for my volunteer assistance at the clinic. However, the clinic staff at Mpemba needed administrative support at the clinic, and asked me to participate. Further, Mpemba was the one clinic that wa s conducting outreach to follow - and this recruitment phase took a significantly greater amount of time than the recruitment of active patients. Patients Who Dropped Perhaps the most sensitive aspect of the study was the proce ss of identifying, locating, guidelines as those patients who have not attended the PMTCT clinic for two consecutive months since their ARV medication has ru n out. In Mpemba, a total of fourteen interviews were conducted with women categorized as and constraints from women most clearly challenged by the demands of PMTCT participation. To identify patient defaulters, medical records were reviewed to create a list of patients determine eligibility: registered in PMTCT after Octobe r 3, 2011, between the ages of 18 and 49, and pregnant or new mother. 57 Patients who default are typically followed up by health center personnel. I learned that due to limited staff, only one health center in the Blantyre district, Mpemba, conducted outreac h and health education in nearby villages to follow -up on defaulters. One HSA at this clinic was trained to conduct open -ended interviews with these women during their scheduled follow -up visits. For this reason, all interviews with patients who dropped ou t of the program (defaulters) were conducted at Mpemba Health Centre. I had intended to hire a female for this sensitive position. However, I found one of the male HSAs, Esau, to have a quiet, nonjudgmental demeanor which seemed to place patients at ease. While volunteering at the health center at which he worked, I saw how he interacted with both male and female patients, and noticed that patients seemed very comfortable with him. I asked Esau to conduct a couple of pilot interviews, and he typically retur ned from the interviews armed with gifts each time sugarcane sticks or bags of mangoes. r home. Fortunately, Esau was responsible for conducting TB outreach to the surrounding villages and so combined these two goals in one home visit. To deflect any attention to him as a representative of the AIDS clinic, I asked him to say that he was condu cting community health education if asked what he was doing in the village. Fortunately, his presence in the villages was so common because of his many non -AIDs -related activities, that he attracted little attention. I did not participate in these intervie ws, because of the tendency for my presence to attract attention from the villagers when I first accompanied Esau to follow -up with patients identified as defaulters. community was expected, but as a foreign visitor, my presence was unexpected and was likely to 58 bring unwanted gossip from villagers toward the homes we visited. For this reason, I did not attend the interviews with these former patients. Still, four of th e 18 defaulters were not found, resulting in a total of fourteen women who qualified as defaulters and completed the interviews. Of the four women not found, we learned that one woman was a traditional healer who moved to the capital city, Lilongwe. While searching for the homes of the three remaining patients whose addresses were difficult to find, some villagers told Esau that sometimes at the time of registration, AIDS patients will change their name for the medical record so that they cannot be identifi ed as HIV positive in their home villages. One woman is believed to have changed her name when reporting to the clinic, and was not found because she did not want to be found. I quickly dropped this patient from the study. The other two women lived several hours from the rural clinic. Because of the distance between the two villages and the clinic, it is likely that geography may be the reason for not returning to the clinic. When I spoke to clinic staff about the distance to these villages, they described them - traditional birth attendants (TBAs). Health Care Providers Four in -depth interviews were conducted with health care providers at Mpemba Health Centre. Interviews were conducted with nurse midwives, clinical assistants, and health services assistants (HSAs). These included a convenience sample of three female informants and one male informant. 59 Community Leaders In Mpemba, I conducted interviews and focus group dis cussions with teachers, church leaders, village chiefs, and traditional birth attendants (TBAs). I conducted one focus group discussion with teachers at a local secondary school. Participants included a convenience sample of men and women who were working as full -time teachers. In addition, I conducted one focus group discussion with church leaders. Mpemba had a CCAP church within walking distance of the health center at which I was based. I held a focus group with church leaders because they preferred to p articipate in a focus group rather than individual interviews. I interviewed two chiefs from the Mpemba region, both selected through convenience sampling, one of whom was male and the other female. Both interviews were conducted through simultaneous trans lation. Village chiefs were included in the sample because they hold political authority within Malawian communities and are powerful social agents who handle household disputes between men and women when ankhoswes, marital counselors from within the kinsh ip group, are unable to negotiate a resolution. I interviewed two traditional birth attendants (TBAs) in the peri -urban Mpemba region who were recruited by snowball sampling. The HSA asked the village chief in Mpemba for the names of TBAs in the village. T hese interviews were conducted with simultaneous translation by my research assistant, and both TBAs were interviewed at their homes in the surrounding villages. TBAs were included in the study because of their pivotal role in the reproductive health of wo two of the 60 Additional Methods Used Member Checks At the end of the study period, member checks were employed to validate study findings. Member checks involve bringing together a group of participants from the populations sampled PMTCT patients and PMTCT health care pro viders separately presenting preliminary themes, and soliciting feedback on the themes presented. Member check discussions were conducted in both urban and peri -urban study sites. Two focus group discussions were conducted with health care providers and tw o focus group discussions were held with patients from two of the three study sites to elicit feedback on initial findings from representatives of the study population. Health care providers confirmed the preliminary research themes. However, one theme tha t was understandably uncomfortable for health care providers was the apparent fear of negative interactions with some providers. Health care providers did not comment significantly on this theme. After developing a list of preliminary themes to present to patients in the member check discussions, I realized that geography and lack of transportation never emerged as a barrier to care. This surprised me, as I had traveled throughout the region on foot and in a four -wheel drive vehicle to see where patients li ved. Several times while exploring the rural region, I would find patients from the clinic who were physically impaired and struggling to walk home. I would deliver these patients home to assist them, and found the distances from the rural clinic to be unrealistic for patients who must walk long distances to the bicycle taxi stop. Furthermore, the bicycle taxi cost was prohibitive for many of the patients who frequented the clinic. These patients chose to walk the additional miles because of the cost. A thi rd, but more expensive option in the rural region was the minibus which traveled on the paved highway, and not the dirt 61 patient focus groups, people reminded me that walking long distances and not having the cash to access transportation on a regular basis was a normal process and likely not mentioned because management . Analysis Transcripts were analyzed and analysis was conducted to answer the overall research questions examining the reasons women continue or do not continue treatment within PMTCT programs. Grounded theory was used to identify additional themes that e merged from the data to explore potential relationships between stated barriers, and to uncover barriers and supports not identified in the literature. A common technique within grounded theory analysis is the application of the constant comparative method (Bernard 2000) to refine these theoretical constructs. Using this method, I continually compared units in the data to refine the core and secondary theoretical concepts. By continually comparing theoretical concepts to newly collected data, I was able to identify the dimensions of core and secondary concepts, explore their interrelationships, and integrate them into a coherent set of generalizations (Bernard 2000). This is an iterative process in which the analysis of data guides subsequent data collection (Bernard 2000). Comparisons were made against negative cases to assure all possible variations in the data were accounted for (Bernard 2000). Each interview was reread to determine whether interview responses fit into the existing theoretical concepts and continued until all data were accounted for. 62 To conduct the analysis, relevant text was selected that either related to my research a name or phrase that I d eveloped or stemmed from participant language. More focused coding was conducted with codes used most frequently. Codes were written as gerunds to eliminate relevant text from transcripts, based upon how they related to each other whether by causation, as consequence, etc. Codes were clustered based upon similarities, and clusters of initial codes were renamed as themes. In the final stages of analysis, I used p articipant quotes to write a (Auerbach and Silverstein 2003). Finally, I compared these theoretical narratives with the existing literature to answer my researc h questions and to explain other themes that emerged from the data. Coding began once a number of interviews were conducted and patterns had begun to emerge from the data. Throughout the interview process, I used memoing to document potential codes and exp lore relationships among themes (Bernard 2000). Data codes reflect both the questions in the interview guide and broader themes that emerged from the interviews with patients and providers, focus groups with patients, and clinical observations . Quantitati ve Survival Analysis A quantitative survival analysis was conducted in the field to measure the frequency with which patients drop out or default from the PMTCT program at Queen Elizabeth Central Hospital (See Appendix A). The analysis was conducted to pro vide QECH with a record of patient adherence rates. Given that the focus of my study is on the social context of patient 63 defaulting within the primary PMTCT clinic in Southern Malawi, calculating specifically how many women are unable to remain in this PMT CT program allowed me to determine whether or not defaulting is a problem at my main study site, and if so, to calculate the extent of the problem. The quantitative analysis addresses the following research questions: 1) What proportion of women within the study population default within the study period, and 2) Does the likelihood of defaulting depend upon age of the patient? The survival analysis is included in the Appendix. In Appendix A, Table 6 (Patient Profile) includes all of the methods used in the study as well as a summary of patient demographics. In the following chapter, I discuss the discourse and PMTCT programs. I discuss the meaning of reproduction to the social identities of men and women within Malawian society, and to the family and kinship group. 64 Chapter 4: Local Understandings of Reproduction and the Construction of Motherhood rmants in this women and men to enact social identities associate d with motherhood and fatherhood. This chapter is divided into three sections. In Part I, I discuss the meaning of reproduction to the social identities of men and women within Malawian society. Part II nfertility. I demonstrate the centrality children to marriage, kinship group and community. Part III explores constructions of motherhood, primarily from the perspec tive of PMTCT patients. I discuss how women in this study enact this important social role by prioritizing the caretaking of their HIV -exposed child, conceptualizing compliance as key to their mothering role. Finally, I examine the social contexts that cal identity as wife. infertility. A discussion of infertility is relevant to my larger discu with HIV/AIDS treatment because both health conditions share common themes of a gendered stigmatization rooted in the social obligation to reproduce. Further, each of these health e role, construct infertile women and pregnant women with HIV/AIDS as failing to conform to ideals of motherhood and wifehood. 65 In addition, I demonstrate that the reification of motherhood has its opposite in the marginalization of women who are infertile and cannot have children. Just as women with HIV/ AIDS are perceived as promiscuous and immoral, infertile women are stigmatized and mocked for not conforming to the normative constructions of womanhood. In both instances the gendering of stigma and the ge ndering of infertility show how gender inequities become community. For the kinship group, community, and larger Malawian society. Gender roles for wome n as mothers, wives, daughters, and community members maintain that women fulfill their reproductive role to the family, kinship group, and community. For the husbands of these women, enacting the role the household, kinship group, and community. As in other sub -Saharan societies, childbirth in Malawi procures marriage, marks the transition to adulthood for men and women, and symbolizes the continuity of the lineage (Yeatman and Trinitapoli 1999). The da ta presented in this chapter was drawn from interviews with health care providers and current PMTCT patients, and focus group discussions with community men and women. Part I: Meaning of Reproduction to Local Femininities and Masculinities Being a mother is central to the construction of womanhood in Malawian society. Fertility is both a biological and social process, assigning women greater status within the 66 community for their reproductive capabilities. The meaning of fertility becomes clearer, however, when looking at how infertility is conceptualized, and how infertile women are regarded. Women experience a range of negative responses from husbands, family and community when they are unable to meet these social pressures. Informants describe infertility as a significant social burden to women. Health care providers, community members, and PMTCT patients claim that women without children are targets for a range of social responses from the family and community. Stigmatizing discourse towards infertility m ay include mockery and name -calling, with such expressions as the Ulikutayiska waka economic val ue of fertility as a commodity to be exchanged between lineages. More derogatory Ukuzuzga waka chimbuzi pano Human Rights Commission 2005). Among some matrilineal communities in Zambia, i f a couple is unable to have children the woman may divorce the husband (Poewe 1980; 1981). This kinship structure thus confers notes that some women enact procreative p ower by constructing gendered boundaries around the while wome Poewe notes that the Bemban and Luapulan communities explicitly differentiate between Such opportunities to challenge the source of infertility were absent among the PMTCT patients I 67 mitigated by a lack of reproductive potential. The notion that infert ility may be located in the infertility is the same as that d iscussed with respect to the women receiving treatment for source of this disproportionate stigma is another example of how gender inequities become embodied in wom These women lack the power to demand divorce and find another husband with whom to have children. Moreover, some women may be replaced by another wife or fiisi (male surrogate) if they are unable to have a child. If it is believed tha t a woman cannot provide a child for her husband, he will seek another wife, or negotiate to obtain the reproductive capabilities of her sister. as the subject of socia l commentary by the local community, thus demonstrating the diffusive nature of biopower as it operates through the monitoring eyes of the community. Just as the community monitors for HIV, the community monitors for infertility, thus exercising control over their social relations and marking deviance through various disciplinary mechanisms mocking or community gossip, marking social spaces that include and exclude, and sanctioning the abnormal with medicalizing discourses of infertility and witchcraft (Fou cault 1977). As noted 68 -positive, they also mock those women unable to bea r children. Informants report that those most likely to deride women include various forms of gossip with often disparaging comments, claiming that community members located at the margins of society, although there is variation in this as the following statement from a health care provider illustrates. Below, the health care provide r asserts class status by constructing those who mock infertile women as lacking in education: They are mocked by their fellow women that they can't have children. It depends on the educational level of the people you are living with because when you are living with people who have gone to school, at least they understand. But in Malawian communities we have a lot of people that are not well educated so they MOCK at the woman saying she can't produce. She can't bear babies. And the woman has a really tough time with the relatives at the man's side. (Female health care provider) care provider conflates the social risk of infertility with the biological risk of HIV transmission, suggesting the cultural pressure to reproduce places men and women at greater risk of contracting HIV: They will be shouted everywhere they go. Even small ch ildren will be pointing at them, promoting HIV infection being highly transmitted because the husband will look for a ving a child and then they will keep on -positive. (Female health care provider) 69 Part II: Community Responses to Infertile Women: Pregnancy, Secrecy and Social Risk Some women mar ked as infertile are marginalized by their communities by community witchcraft discourse employed by community members whose watchful eyes reproduce the monitoring practices of the local clinic, thus ensuring the social and biological reproduction of society. The following statements were made during a focus group discussion I conducted with men from a rural study site: Sometimes giving birth will take place in the villages because of delays. They say even though the woman is in pain, you cannot go to the hospital during the day. You see, they do come during the night. Because of that sometimes, they will give birth right in the village when they're not planning on it. Maybe children are told that babies are born in the hospital. So they will try to have children not know that the mother had to leave for the hospital. So they leave during the night. Why is it during the night? (Interviewer) There should be some form of secrecy. Sometimes you avoid the witchcraft. People should not see them going because they took some necessities. That you took things for the delivery. If they see that certain things are taken, they know that this one is going to deliver. So witchcraf t can happen. Giving birth is secretive. Why is it secretive? (Interviewer) Maybe it's part of culture. So if you go at night, the community doesn't see you deliver, and you are less at risk for witchcraft. Informants living in one rural region claim the secrecy of pregnancy is maintained by not going to the hospital during the day for fear of exposing oneself to witchcraft in daylight. In this community, informants maintain that women continue to use TBAs because of the distance to the health clinic and f ear of negative attitudes of health workers at the nearest health clinic. l 70 worlds, stigma, gossip, and witchcraft discourse reinforce and reproduce normative ideals of social and biological reproduction. Witchcraft discourse may demonstrate state power as it constitutes its subjects through the reinforcement of reproductive ide als. Such discourse may also serve as a leveraging mechanism for members of the community with little social, economic, and political power to manage social inequalities within the community (Luedke and West 2006). In her study of traditional healing pract ices in Mozambique, Luedke explains the witchcraft discourse surrounding the sale of traditional healing services by curandeiros . When these healing services begin to replace free spiritual healing provided by local churches, anxieties about social competi tion and the commodification of the local economy emerge (Luedke and West 2006, 82). in which witchcraft accusations emerge during times of commoditization and economic inequality (Ashforth 2005; Lwanda 2005). I have a sister who...that one we do think she is a witch because we always wonder wi th her behavior. She always says anything against those people who have children. So sometimes I can tell maybe it's witchcraft. Sometimes it's because of abortions. Others say she is being bewitched. Some may wish that she does not have children. Does you r sister think that? (Interviewer) Even myself I do think. Sometimes I won't get in the car with her. Apart from Malawians. If you don't have a child, people look at you with some eyes. At this age, she should have a child. They don't respect you. All sort s of insults would come to you. (Male teacher) This pattern emerged primarily within one rural community where informants framed pregnancy as a social practice vulnerable to witchcraft. These informants described pregnancy as 71 pregnant women will travel to the traditional birth attendant (TBA) at night, in the dark to avoid the wat chful eyes of the community. If pregnant women are seen carrying cloth, basin, and soap, it is believed they are going to the hospital for delivery. One community informant describes the practice below: There should be some form of secrecy. Sometimes you avoid the witchcraft. People should not see them going because they took some necessities that you took things for the delivery. If they see that certain things are taken, they know that this one is going to deliver. So witchcraft can happen. Giving birth is secretive. (Male teacher) In her study of reproductive risk in Mozambique, Chapman (2012, 222) describes a similar pattern of monitoring practices against women unable to have children. Chapman found that pregnancy evoked jealousy and distrust among en vious infertile women and inquisitive neighbors who monitor the allocation of good and ill -fortune within the community. Perhaps those targeted for witchcraft are not only targeted for the social wealth they will receive in childbirth, but also for materia l wealth they carry as the ability to afford the supplies demanded of them at the public facilities signifies access to scarce resources. function of reproducing norma tive ideals of social and biological reproduction which further the infertility. Below, an informant reflects upon the local manifestation of a global pattern of gendering infertility (Van Hollen 2013), placing the blame of being unable to bear children on the wife and not the husband: 72 Culturally, people think that it's usual ly the wife who has got problems. The one who receives a lot of stress as part of this whole thing... in the community... is the wife. 'Cause even the relatives to the man or the relatives to the wife, think she is not performing in the house, it's just th e way our culture has brought us up. (Male health care provider) Infertility as Gendered Embodiment In many parts of the world, women are most likely to be blamed for infertility (Lock and Nguyen 2010). In this study, informants suggest that women, rathe r than men are most likely to be blamed for infertility. Such patterns point to the gendering of infertility, where normative alities in gender relations between men and women. In men in Malawi, when it comes to these reproductive issues, the woman is mostly the one to blame. The man is infertile, the woman can be fertile, but if there are no babies in the family, the first one to be suspected is the woman. (Female health care provider) Unfortunately, the Malawian community tends to blame the woman and not the man they say that the woman cannot have children. So it's not a nice experience. Some people tend to look down at you li ke you are a failure because you aren't able to have children. You have patients who are forty who are still hoping that one day they are going to have a baby. You cannot really offer them much... they still have to prove to the community that they can hav e children. They would never believe that infertility is a male's problem. (Male health care provider) The belief that wives and not husbands bear responsibility for being infertile illustrates the concept of gendered embodiment which captures the social process by which gender inequalities become internalized (Connell 2012). For women unable to reproduce, the inability to fulfill the maternal responsibility places them at the margins of society as they have failed to reach a critical sociocultural and dev elopmental marker of womanhood. 73 infertility to normative ideals of womanhood. In their own words, these individuals describe infertile women as experienced those sociocultural rituals marking critical transitions in the life cycle of women. Such local constructions are reproduced in community sanctions that forbid women from holding a child until they have fulfilled this social role, thus marking social spaces that include and exclude, and sanctioning the abnormal with medicalizing discourse of infertility. Other in responsibility the biological and social reproduction of Malawian society. When asked to framed these women as lacking in purpose and social value. She reiterates the comments of other married at the time to her best friend from childhood. She had a vivacious personality, and was always quick to laugh. One day I found her in tears as she was performing her domestic work. As pressure to bear a child so that they could build a family. She then discussed how much she feared losing her marriage. W 74 When I returned to Malawi a year and a half later, Evelyn had still not given birth to a child. She appeared much more subdued, and although her personality recalled her earlier vivacity and playful nature, she was clearly burdened by her worries. When I a sked her in private husband had gone to impregnate another woman, but did not yet want to end the marriage. Evelyn was overwhelmed by fear that the marriage would eve ntually end, and was devastated that her husband had fathered a child with another woman. Enacting motherhood is central to the identity of the majority of the women in this study. A woman who has failed to reproduce does not have the social capital in whi ch to enact her responsibilities within marriage, solidify social relations within the lineage or leverage respect to reproduce signifies a failure to fulfill t heir roles as wives, mothers, sisters and daughters. Women unable to provide children for their husband fail to fulfill one of the central tenets of wifehood for their uncle or older brother fail to provide a bvumwe over which he may expand upon his social wealth. To other women in the community, women with reproductive capabilities may serve as social threats as they are known to have greater marital potential. Similarly, men experience the social pressure to reproduce and bear children. Within the PMTCT patient population in this study, by ensuring a family over which he may provide social and economic security. Informant s claim may lead some men to delay seeking 75 ability to demonstrate fertility by producing children for the family and lineage. In the following quote, a male health c are provider confirms the pressure placed upon Malawian men to reproduce: Culturally if you are a man, you should show that you have children. You should have children, it's just the way we are trained, brought up. That you should see some children... run Meaning of Reproduction to Family and Kinship Group The shifting nature of power and gender relations is evident in the structure of the matrilineal kinship system and the significance of reproduction to the socia l and economic power to the matrilineage. Therefore the status and labor power of children is granted to the larger matrilineal kinship system (Crehan 1997, 92 ). When describing the meaning of reproduction, community members and health care providers equated having additional children with amassing greater social and economic wealth. al informants situated reproduction within a religious context, claiming that having children signifies a family informants employed religious discourse, claiming that whether they would have another child 76 According to normative ideals within communities adhering to matrilineal practices, children expand the social wealth of the lineage and enhance the status of the uncle or avunculate . Within both the matrilineal and patrilineal kinship systems, children signify social and material wealth, and confer greater status to the lineage. Children do so by expanding the size of the lineage and by contributing the products of their labor to the kinship group with which they live. Within patrilineal kinship groups, the birth of a child sanctions the institution of children are In communities adhering to matrilineal practices, the birth of children confers greater stronger bonds. Uncles within matrilineal kinship systems become more powerful as size of the lineage is equated with social, political family grows, for the uncle has expa nded the lineage by expanding his clan (Van den Borne 2005). These normative ideals suggest men achieve their status through the role of malume (uncle) rather than that of the husband (Mtika and Doctor 2002). The social status of the malume is augmented by providing him with a bvumwe over which he exercises political, social and children, such as paying school fees and paying for wedding costs. One health care provider expands upon this social process, emphasizing the desirability of a larger clan to a matrilineage, stating: 77 the uncle is a bit more powerful. The tribe is growing bigger. That man has Aphwanga. So he has more status. How about if she has only one child? (Interviewer) Having children contributes to the social status of the household banja and lineage. Situating the social value of the child within both mat rilineal and patrilineal contexts, the following health care provider describes the social wealth attributed to having children. The provider links marriage and reproduction as an opportunity for female children to amass bridewealth within the matrilineage : Other parts of Malawi they would take that as a wealth. And more especially female they will make little children and get married quickly. Because they expect that girl to go get married and then give bridewealth to the family. This is for the Chewas. (Female health care provider) In the following comment, the informant explains how in patrilineal kinship structures such as that of the Ngoni, greater social value ma y be placed on males as boys can inherit the wealth of their fathers: woman is giving birth to only girls then the husband has to go find another woman for him to have a Ngoni. The very same thing happened in my family. We are three. We are all girls and now my Dad is looking for a boy. (Female health care provider) Commodifying Infertility: Multi ple Wives and Surrogate Sisters She is saying sometimes, God created us but sometimes it happens that there are other problems that can make us not to have the gift of a child. We are married okay, but the gift of a child is not there. For such people, wh at do they do? (Female health care provider) 78 potential is made clear in the multiple strategies employed by spouses and kinship groups to ensure that a couple will repro duce. Informants suggest the couple and the extended family may employ a range of strategies to ensure the reproduction of the family when a couple is believed to be infertile, including use of a male surrogate or fiisi, appropriating the reproductive capa sister, use of multiple wives, divorce, use of drugs/medical treatment, traditional medicine, or prayer. Fiisi is a cultural practice in which a male member of the community is asked to impregnate a woman who has not become p regnant in her marriage. These men serve as surrogate fathers so that a couple may have children. In rural communities, if the husband is suspected to be infertile, the family will secure a surrogate to provide a child for the couple. However, if the wife is suspected of being infertile, the family will call for the husband to divorce her. Thus, in these communities, women are more likely to lose the marriage. Given that women are most likely to be blamed for infertility, the risk of losing marriage is part icularly high for them. In some areas of Malawi, the use of fiisi is practiced. While discussing the social practice of fiisi Fiisi was actively practiced in one study community, but was more commonly They will just look for any woman. And some there are some other cultural practices. the husband the wife is not man whom they have seen that in his family he has children. They will call him, come heal my family. You should sleep with my wife and th en you should impregnate my wife. They call it as a fiisi . If they do the sex unprotected, definitely they will do it unprotected 79 because they want a child. If that man is HIV - infected. Those are some other ba d cultural practices. (Female health care provider) Bearing children is essential to the foundation of family and the construction of local masculinities and femininities. Therefore, seeking pregnancy outside of marriage may be a coping strategy by which women preserve the stability of their marriage. Female teachers interviewed within a focus group discussion claim that infertile women living in patrilineal societies have considerably less decision -making power in marriage and must negotiate greater socia l challenges. These women are more likely to have the husband find a infertility, or to be divorced: In the patrilineal type of marriage, they will tell them that the woman has contributed to that and they will ask for a second marriage. Usually they will ask for a separate marriage, maybe the sister of the woman provides the children for the family. Sometimes it might not be a sister. They may look for somebody el se to carry it. Sometimes the other family, they have a baby, and it becomes difficult for them to let you hold the baby. (Female teacher) As noted in the example above, husbands may exchange the reproductive body of his wife for that of her younger siste r if she fails to provide a child early in the marriage. These younger siblings either serve as a surrogate or join the marriage as a second or third wife. Husbands may also take on a second wife if the first wife is unable to provide children. For many of the PMTCT patients in this study, husbands often decide where a woman will give birth at home or at the hospital. Husbands also strongly influence the number of children in the family. In contrast, for many of the professionals and nurses in the study, husbands and wives shared the decision -making authority when making family planning decisions. 80 Decisions related to family size are not only made by husband and wife, they are also severe, placing a - -in-law in particular. Pressure from in -laws is most pronounced in patriarchal kinship systems where the social and material value of children weighs more heavily. The following informant situates the meaning of infertility beyond the household to the larger kinship system: She can't bear babies. And the woman has a really tough time with the relatives at the man's side. Because the mo ther in law aphongozi. So when you are married they are looking forward that you should have a baby in some time and when you're not producing, they are asking you, or they even ask the man to leave you, that you should go to somebody else who can have bab ies. So it's really like a mockery. Being infertile is not good here in Malawi. (Female health care provider) They can tell the son to look for another woman. Would he get a new wife? Yes, he would get a new marriage. You can even be in the house while an other is having an affair. Why do they want babies? They just want babies. They want babies. They want children in this house. Why are children important? They are a big asset. A house without a child in Malawi is useless. (Female health care provider) The social and economic value of the child to the lineage is evident in the following example of a failed delivery recounted by a female health care provider in which the mother -in- ly she deploys her power within the lineage to enact violence on the young woman who is viewed as withholding a child from the family and patrilineage with a failed pregnancy. cal care from a traditional birth attendant (TBA) rather than a hospital delivery. After two failed deliveries, the mother -in-law co -opts the assistance of the TBA to ensure a successful pregnancy. The quote provides an example of the sometimes violent for ms in which reproductive pressures 81 may be experienced by women whose mother -in-laws exert control over their reproductive health to ensure the social and biological reproduction of the lineage. By ensuring reproduction, the aphongozi reinforces her status within the family when the son has brought children to the lineage. I have one example. This patient, she came from Mulange. This was the third pregnancy. The first pregnancy she started labor. She wanted to go to the hospital. The mother -in-law said no y ou will not go to the hospital. You will go to the traditional birth attendant. They went there, the labor was not progressing. A stillbirth. The second pregnancy, the same ill each and every time she is pregnant her mother -in-law takes her to the birth attendant. Next time around she went there, she labored, she labored, nothing. And then the ?(descent)? was high. The mother -in-law and the T BA took an mpanga ( knife), they inserted it into the vagina, it will bring the baby down like this (demonstrates) but they failed and then the woman had ruptured the uterus. She bled. They went to the hospital because the woman was gasping. She arrived her e with offensive discharges of the uterus. We kept her for almost two months. The mother -in- coming here because you are not giving us babies. We gave MK 5000 to your mother and to your brother, but because you are not giving us babies, we will take thirteen year think igno rance also leads to some of these problems. (Female health care provider) In the above example, fertility becomes a commodity to be exchanged between lineages. When the productivity of one body fails, it is replaced by a younger sister whose fertility rep ays and of society (Chapman 2012). The reproductive value o f the anticipated birth goes beyond the 5000 kwacha, to the social and material power of the family and lineage, transacted through the exchange of one reproductive body for another. Forced delivery was not a common experience 82 among the women I interviewed . However, it illustrates how societal and kinship pressures to reproduce interact with gender inequalities and poverty, situating the young woman in a social context in which she has little agency to assert control over her own social and reproductive power. While recruiting women to participate in the study, one woman appeared for an interview to talk about her experiences within the PMTCT program. She had heard that a researcher wanted to talk to mothers about their experiences in the program. I often be gan the interview with a discussion of the baby and his or her health to take the focus off of the mother until she was comfortable. When I began to ask questions of the mother, she hesitated, then nervously offered the information that she was not the mot her of this child. The infant was in the PMTCT program, but she was not the mother. She then began to tell me the story of her younger sister who had died after initiating and then stopping ARV treatment within PMTCT. At eighteen years old, the young HIV -positive patient had been in an unstable relationship with the father of her baby, and tested positive in the antenatal clinic at the district hospital. However, she had been positive for some time, and was inconsistent in her use of the medicine. The young er sister then died in labor. At 38 years old, the sister explained that she had not yet been able to have a child, These extended notions of kinship are not uncommon to families who have lost family members to HIV/AIDS (Lock and Nguyen 2010). The woman above signifies the growing responsibility placed upon family and kinship to assume parenthood for children whose HIV -positive parents have died. She also signifies the social status allotted to women who have enacted their reproductive responsibilities, even if through the ties of kinship. The woman above is unmarried, and offered that she may have her own child, but clearly believed she had assumed 83 why she stepped forward to join the cohort of HIV - performatively as she align ed herself with the shared social identity of other new mothers despite the fact that she was not HIV -positive, and not the mother of her infant. Monitoring Fertility: Family Planning Movements and Donor Support The commodification of reproduction is also evident in the state and transnational efforts to manage reproduction through reproductive health and family planning initiatives. Ideas about family and population size drive the agendas of family planning and reproductive health policy of the state and transnational donors, demonstrating how relations of power are embedded within Social scientists have brought attention to the plight of many women in sub -Saharan Africa who are burd ened by infertility, suggesting support from social networks has declined because transnational donors increasingly link economic support to state control of population growth. While state support to treat infertility has weakened, clinics are provided wit h contraceptives to ensure population control. In many Southern African countries, social support mechanisms such as fostering have emerged a form of reciprocity in which the extended family procures a child for households unable to have children. This sam e mechanism of reciprocity provides a support system for parents when they are older. -seeking within PMTCT programs brings attention to the tension that 84 contradictory social roles that call upon their desire to protect their child from mother -to-child transmission (MTCT) while preserving their marriage. Navigating th ese contradictory roles is all discussed at the end of the chapter. Again, women are constructed as failing to conform to ideals of motherhood and wifehood, demonstrati ng how gender inequities become embodied in mother challenges her identity of being a wife when she discloses an HIV -positive status to her husband. onses to Reproductive Pressures Health care providers and community women claim that some women will hide the infertility of their husbands by seeking other men with whom to become pregnant. Female health care providers shift the location of reproductive p ower by reframing this social process as Finding a surrogate male with whom to conceive was framed as both a may prove her own fertility. Teacher s within a community focus group discussion stated that women may try to become pregnant with other men, bringing the child into the family as though it were the child of her husband. Women seeking pregnancy outside the boundaries of marriage do so without the awareness of the husband. For those who do not tell their husbands, the intent causal link between these manifestations of gender inequities responses to inf ertility and responses to women with HIV/AIDS shed light on the social obligation imposed upon families to have children. 85 so, the informant creates new social spaces for women to resist the cultural practices that point to her reproductive failure. Instead of a husband or extended family demanding she reproduce with a fiisi, the informant constructs women as going out to prove their own fertility. Such reprodu ctive potential. If she goes out and proves she's pregnant...she will pretend that this pregnancy is from the problem or the husband. Sometimes they agree to have a second wife...or fiisi . (Female health care provider) Part III: Enacting Motherhood: Maternal Sentiment and Prevention of Mother to Child Transmission As of my side, the most important thing in my life is to have a healthy body so I can work and serve my chil dren. (Current PMTCT patient, Blantyre) The PMTCT patients in this study clearly prioritized their social role as mothers and shaped their health -seeking decision -making accordingly. When asked why they participate in PMTCT, women expressed an overwhelmin g acceptance of a positive status. Women pointed to the marked improvement in morbidity and mortality for PLWA (people living with AIDS) in -posit ive women participate in PMTCT to gain access to medication 86 prophylactic ARVs, postnatal Nevirapine, and antibiotics to extend their lives, to prevent the HIV -exposed infant from acquiring the virus through mother -to-child transmission (MTCT), and to exten of capitalist and colonialist social forces (Walker 1995). Christian and Victorian discourse on family, in service of the spouse and children. While the ideals of motherhood have remained stable, researchers note that the institution has shifted with an increase in the number of single mothers having children outside the institution of marriage (Walker 1995). For the HIV -positive mothers in this study, consuming ARVs was described as a strategy to protect children from the HIV virus and nurture them well into their childhood years. The majority of the patients interviewed conceptualize their attendance at the PMTCT clinic as key to exposed child sheds light on their most significant social role as mothers. In thi s study, the -making processes related to PMTCT By extending their own lives, women believed that they could protect the health of their exposed infant. Thus, the y conceptualized their own health as a means to care for the child, with the health of the child as the ultimate goal; these women prioritized the health of the baby, viewing their own health as secondary. These conceptualizations of mothers the idea that is supported by similar research on HIV -positive motherhood (Long 2009; Van Hollen 2013). 87 y in HIV -positive motherhood. This study predates the universal roll -out of mortality. The study also took place at a time when PMTCT was newly available, and patient care for her children involved securing alternative caregivers to take o n the mothering role after her expected death. These mothers went so far as placing children up for adoption or in orphanages to ensure they would be cared for if they themselves did not survive. The women in this study, while sharing the same maternal sen timent and enacting the same maternal practices of caretaking, enjoy the privilege of accessing ARV medication at a time when access to ARVs is no longer limited to the wealthy. ten contested discourse of motherhood that prevailed within PMTCT programs in recent years (AbouZahr 2003). Highlighting the importance of the health of the baby, these programs were criticized for contributing to an erasure of the maternal body (Long 2009). When asked why women would not want to continue treatment within PMTCT programs, several health care providers claimed that women are more likely to discontin ue treatment for themselves, but do not wish to withhold treatment from their child. One health care provider explains an encounter she had with a mother who came to collect her medication for the child, but not for herself. 88 There was this little one of n baby should not have the same problem. But myself no need, sister. (Female health care provider, Blantyre) Motherhood Discourse The future and care of my children was my main concern, now I just believe and pray to God to grant me long life so that I can continue taking care of my children. (Current PMTCT patient, Blantyre) constitute the social identity of patients constructed the good mother as a woman who is compliant with PMTCT/ARV adherence, aligning the roles and responsibilities within PMTCT with representations of the Wome n who do not participate in PMTCT - other HIV - demonstrates the construction of subjectivities that enable some HIV -positive mothers to al ign with the dominant caregiving practices of good mothers. Thus, when seeking treatment identity as a compliant patient. A similar pattern was discussed in Corn 89 2002). emaining compliant with the clinical obligations within PMTCT programs. Below, health care providers equate the dominant motherhood discourse of child good mot her: PMTCT to protect the baby, and they love the baby. (Female health care provider) who have accepted and they have that feeling that my child should survive, they take the medication. (Female health care provider) While women are blamed for being unable to reproduce, some health care providers blame HIV -positive women who fulfill this soc ial role, again placing PMTCT patients in a double bind where they are blamed for being HIV -positive and for being pregnant. Informants suggest that with the arrival of the 5A regimen, HIV -positive women are reporting pregnancies that result in uninfected infants. This new medical intervention is expected to reduce mother -to-child transmission (MTCT) by up to 90% from the previous 50% (Schouten et al. 2011). PMTCT policy advocates that women should be provided a choice whether to have additional pregnancies when found HIV -positive (Ministry of Health 2011). However, there is a significant difference between these national and transnational ideals and their actual practice by the health care providers who implement them. Some PMTCT patients maintain that heal th care providers convey the notion that they are placing themselves and their future child at risk 90 body as infecting body (Long 2009) emerges when some HIV -positive w omen decide to have additional children despite their HIV -positive status. The pregnancy itself it is a problem whether you are positive or negative. Then with the HIV it is worse. Because your immunity is still low then with the HIV it becomes worse. Bec as we know. So once you give birth, you need to take care of that baby carefully, also baby is prone to infections. Because you say I have to take care of this child. After the child is grown up to 2 years, you say you I should get another pregnant. You are still risking your life and also the life of the baby. (Female health care provider) Employing biomedical risk discourse (Farmer and Kleinman 1989), female health care providers at several clinics asserted that patients may be exploiting greater a ccess to ARVs to facilitate additional pregnancies by which they would otherwise not be capable. These health care providers suggested that women are leveraging the presumed efficacy and scientific legitimacy of ARVs in reducing mother -to-child transmissio n (MTCT) to justify their decision to have more children. Grounding their statements in biomedical risk discourse, these providers assume that HIV -positive women should not have more children because of the nature of their positive status. But others they come because they want to get pregnant. So they know, I want to get another pregnancy. I will still be giving birth because there is a medication. They want to birth because we are told in PMTCT that we can even have babies that are not infected. Oh, we will go and get some medication for the treatment. Because we are told in that I can have 2 or 3 babies, who are not infected, so I can be giving birth. There is no problem, because there is medication here knowing that they are risking their lives. (Female health care provider) 91 Reports of negativ e attitudes toward women who are HIV -positive and continue to have children also emerged in discussions with current PMTCT patients during a focus group discussion when they were asked about whether they had ever experienced stigma. Yes they talk about us . Yes, they say look at that one, they know they have the disease that, we knew of it here. (Current PMTCT patient) I was once coming for the antenatal clinic and the nurse who was supposed to check my BP before going into the examination room. She was also impolite to me. When she saw my book [Health Passport], she asked like that [HIV -positive], why do you do this? Do you not k now others die? You are just giving birth? You will die as a result of that! That can make you suddenly sick. You see t we are like that, so we get Yes, to say the truth, that is the biggest problem we have here. Okay. Stigmatizing us. Stigma? (Interviewer) Yes. In well expressed Chichewa! What happens for you to know they are disc riminating against you ? (Interviewer) Maybe when they take your book [Health Passport] and when they just read it, the way they handle your book, what they will do. The way they react? (Interviewer) Right! Your friends then immediately know that there is s omething wrong with you. That worries you to the point that I wanted to change where to get the medication. Maybe they have a different view of us because of our status. (Current PMTCT patient) The HIV -positive mothers, they continue to become pregnant ev en if they know they are positive. After the child is grown up to 2 years, you say I should get another pregnant. You are still risking your life and also the life of the baby. (Female health care provider) I too think with the coming of Nevirapine, medic ation provided to infants within PMTCT programs, they think that I can bear a child, he or she will be taking the drugs, so he cannot be tested positive. And some NGOs will assist you if you bear a child. They say you can get pregnant if you are positive. At the health facilities there are health workers to assist you. (Female health care provider) biotechnologies demonstrates how government is exercised in the form of technologies tha t 92 generate objects for management as well as new knowledge (Agrawal 2005). As technologies of dividual responsibility for being pregnant, they are reworking who is deemed eligible to reproduce. Yet few of these women make the choice to reproduce independently. Husbands, members of the extended family, and societal pressures to reproduce all figure into this decision. transformed what it means to be HIV -positive and pregnant. Having additional pregnancies when HIV -positive reveal binary constructions of motherhood in which risk (Scheper -Hughes 1992, 36) whereas the non - to do so. However, the above quotes assum e a level of agency that is not within reach for many of the women in this study. Negotiation of condom use is often outside of the control of many of these women, particularly those with little decision -making authority within the marriage. Risk discourse risk discourse maps assumptions of individual choice and freedom onto the attitudes and practices of patients while erasing the social and cultural context of these practices (Farmer and Kleinman 1989). Yet many women in this study have little control over their own reproductive practices when they often must prove fertility to husband, kinship group and community. As a result, many women experience a double bind in which they are blam ed when they are unable to reproduce, yet also blamed for deciding to reproduce when HIV -positive. While the majority of current PMTCT patients women viewed taking ARV medication within PMTCT as a means to ensure child survival in future pregnancies, some current PMTCT 93 patients did not want to have children when HIV -positive. These women point to the shifts in how some HIV - imparting harm to the health of their child with mother -to-chil d transmission during gestation, birth, or breastfeeding (Long 2009). Such discourse is reproduced in the larger community. Several current PMTCT patients I interviewed reported fear of illness or death during delivery because of their HIV status. As one i mufuna wofera discourse that continues to surround childbirth, framing childbirth as a high risk biological event, duction of society. In the following examples, patients themselves reproduce risk discourse, highlighting their HIV status as a factor in determining family size, and choosing to limit the size of their family because of the assumed biological risks associ ated with pregnancy for HIV -positive women. The following women construct their HIV/AIDS diagnosis as a reason to no longer continuing to give birth: Why have you decided that you will never fall pregnant again? (Interviewer) On my part I never thought th at I would one day be HIV -positive; therefore I find it justifiable never to have another child again. (Current PMTCT patient) Why do you want this one to be the last born? (Interviewer) What actually d isappointed you when were found to be HIV -positive? (Interviewer) My worries were that, I was found with the virus when pregnant, how would I deliver? After being supported, I was told there is no problem, I needed to start medications after coming with my guardian and being counseled, and when the child will be born I will 94 have to take medications for the child until the child is six weeks and tested. (Current PMTCT patient) I was also worried because a lot of people say when you have the virus, bearing c hildren is dangerous, one can die, that worried me but also I had difficulties to find a guardian since all my relatives do not stay near, until one doctor volunteered to be my guardian urrent PMTCT patient) The discussion of a biological risk of mother -to-child transmission reveals an unexpected consequence of PMTCT interventions (Ferguson 1994). As development projects, health interventions such as PMTCT may rework social relations bet ween husbands and wives, and health workers and patients, redefining the social categories of the biological and the social. From a biomedical perspective, HIV/AIDS as a biological and social process places a lens on the dynamics between men and women with in marriage and the various kinship and community actors whose political, economic, and social interests lay at the intersection of these relations. PMTCT patients who are scolded for having additional children when HIV -positive reveal the tension in women who desire to provide a child to the family and lineage, yet at the same time are framed as poor mothers because they have exposed their future child to additional medical risk of contracting HIV. In this context, women who comply with program recommendat ions by attending the program to access treatment are constructed as non -compliant patients, yet are simply enacting their caretaking role as mothers, illustrating the double bind of PMTCT patients who comply with the biomedical standards underlying their health care are at the same time asked to take on a set of ideas and practices that are closely tied to these medical recommendations, suggesting a medicalization and re -ordering of their family an d social life. For example, in Information Education and Communication (IEC) sessions designed to 95 following discussion, one prenatal nurse attempts to debunk tradi tional beliefs about reproduction that are deemed harmful to the health of the mother and baby. In doing so, she takes the opportunity to provide marital advice to the young mothers, suggesting they continue to have sexual relations with their husbands, or provide ongoing sexual relations during pregnancy highlights the unspoken fragility of marital relations, and the normative gender roles that reproduce the gender inequalities within many marriages that I discuss in Chapter 6. In the following discussion, a health care provider asks a group of pregnant women, both HIV -positive and HIV - (Participant): A pregnant woman is not supposed to have sex, bec ause it might harm the baby. (Nurse): This is a common practice, when your husbands want to have sex, you deny for fear of this, is she lying? (Participant): No. (Nurse): You should think about the repercussions of this behavior with HIV/AIDS around, if yo u are denying your husband sex where do you want him to get it from? And after you give birth you are told to stay for some four months for you to get rid of disease, months? other woman is caring do you think he will come back to you? (Participant): No. (Nurse): And if he comes back he might come back with STIs including AIDS, then you give birth to a baby with eye problems, you now start blaming hot chilies. Usually at times like these you find that that there is no love between the two of you and this is one of the contributing factors, for example I cannot feed my husband with porridge only and deny a man sex for 6 months just because you are pregnant. (Female health care provider, health education session) s conflated with ideas about HIV prevention, demonstrating how reproductive health discourse is remade as providers interject 96 their own ideals related to marriage and family. Further, eotypes of men as unable to live without sex to debunk local systems of knowledge, constructing the mothers as uneducated by distancing them from an processes medicalized. The social relations managed in this encounter evoke the intimate colonialism of the colonial era when political officials and health care providers imparted domestic advice on the sexual relations between men and women and the socialization of children to construct an imagined ideal citizen (Stoler 1995). This same process appears in the attitudes and practices towards men and women living with HIV/AIDS. HIV/AIDS di scourse reproduces sociomoral ideals within the larger society, managing the sexual relations of men and women, shaming women for being HIV -positive as they were once shamed for carrying STDs. Efforts to guide women through the psychosocial and sociocultur al challenges of marital relations again emerge in health worker efforts to assist women unable to disclose to their HIV status to husbands, demonstrating the blurring of categories of the social and the biological. As discussed in Chapter 6, during semi -structured interviews, several health care providers offered that they often helped women to disclose an HIV status to their husbands. For those women who were afraid to tell the husband, health workers would offer to tell the husband herself: Usually what we are doing here we encourage the woman to tell the husband or if she is not comfortable to tell the husband we advise her when the husband comes she can inform us and we can do the counseling together. (Female health care provider) Conclusion In this chapter, I have discussed local understandings of reproduction within Malawian 97 I examine the meaning of reproduction to the social identities of men and women within Malawian society and to the family and kinship group . I discussed the meaning of reproduction to the social identities of men and women within Malawian society. I demons by discussing the social significance of children to marriage, kinship group and community. The birth of a child in Malawian society signifies social and material wealth within the inst itution of marriage and within the kinship network. According to normative ideals, childbirth anchors both kinship system regardless of class, education, and ru ral vs. urban residence. I then explored constructions of motherhood, primarily from the perspective of PMTCT patients, demonstrating how women in this study enact this important social role by prioritizing the caretaking of their HIV -exposed child and con ceptualizing PMTCT participation as key to their mothering role. I explore how women experience a range of pressures to reproduce by family and kinship managed by multiple interests most pronounced when a couple does not have children. However, gender roles, relations, and inequalities reinforce power inequities when women are infertile. When faced with in fertility, it is often the wife who is blamed for her inability to reproduce. Therefore, the failure to reproduce has severe consequences to women particularly poor women, who may risk losing their marriage. 98 In fact, their reproductive power is explicitly enacted in their efforts to protect their HIV -positive child. The pressures placed upon men and women to have children illustrate the political economic and sociocultural significance of reproduction in Malawian society. Importantly, these pressures point to the shifting nature of power relations in the realm of reproduction as women enact their social roles as wives, mothers, community members, and patients. I argue that women are striving to be good mothers and do so by choosing to seek treatment within P MTCT programs. In the following chapter, I demonstrate how enacting the 99 Chapter 5: Gender Roles and Relations in Malawi In the previous chapter, I discussed the importance of motherhood to wom identities. To understand how gender and power relations relate to pregnant women and new understood within the study community, and how these constructio ns of HIV/AIDS may wife and mother, ultimately shaping their coping strategies and health seeking practices. In this chapter, I discuss how such constructions situate HIV -positive women outside the h and publicly attend the clinic to receive HIV/ AIDS treatment. her HIV status is disclosed. I examine how HIV disclosure to husbands, the most commonly reported as mothers and HIV/AIDs patients. This chapter describes the power relat ionships within marriage which give greater authority to men in the realm of economic decision -making in the household and relationship power. This chapter further explores how these power relationships shape women's responses to a positive HIV/AIDS diagno sis and health -seeking practices in clinics; as will be discussed, women's dilemmas in health -seeking behavior are often linked to gender inequities in the marital 100 r social identity as wife, when her marriage may be threatened by a positive HIV/AIDs status. This chapter addresses the following research questions: 1) What is the impact of gender do the barriers and -seeking be used as a lens to understand how gender roles and relations affect the lives of women in my sample? The information discussed in this chapter is drawn from interviews and focus group discussions conducted with PMTCT patients, PMTCT health care providers, and community members from the city of Blantyre and surrounding communities. This chapter is divided into four parts. In Part I, I will discuss constructions of gender in and within the community. In discussing constructions of gender, I explore how gend er relations are understood in the context of matriliny. I focus on the gender roles, relations and ideologies around AIDs management, discussing the impact of so cial norms related to sexuality, marriage In Part II, I discuss how these constructions contr ibute to or mitigate against gendered inequalities within the institution of marriage. The barriers and supports women face with AIDS management within PMTCT programs shed light on the gender and social inequalities they experience in their daily lives. Wh ile many active participants claim to have disclosed an HIV -positive diagnosis to husbands, most also claim to know other PMTCT patients unable to do so. 101 Health care providers and patients assert that most women who attend PMTCT are afraid to tell their hu sbands of their HIV status, fearing the consequences of disclosure. For many women, a positive diagnosis signals problems within the marriage and the possibility of marital conflict and divorce (Bwirire et al. 2008). Part III explores how these constructio ns are challenged or reproduced by the women who participated in my study: PMTCT patients and professional women living within the study given the importance of ma In the conclusion, I discuss the implications of these gender and power inequalities on -positive pregnant women and new mothers. Part I: Defining Gender In equalities To deconstruct the broad concept of gender inequalities, I draw upon the work of scholars whose research examines gender and kinship relations within Southern Africa (Conroy 2013; Crehan 1997; Jewkes and Morrell 2012; Poewe 1980; 1981). These sc holars point to the multiple domains within gender relations in which gender inequalities emerge and the shifting nature of power relations as women leverage power in certain domains (Poewe 1981). Within this section, I discuss gender relations across the domains of economic and political relations control over production, reproduction, political power (Crehan 1997; Poewe 1980; 1981) and relationship power (Conroy 2013; Jewkes and Morrell 2012). My goal is to provide a more nuanced and fluid understanding o f the gender inequalities facing the Malawian women who participated in my study. The discussion focuses on a description of gender constructions in marriage in Malawi, out 102 management of a positive HIV diagnosis. To understand how gender constructions contribute to or mitigate against gender inequalities in marriage, I first discuss how household responsibilities and privileges are assigned to men and women in Malawi. Sch olars of African feminism argue that Western feminist assumptions of universal male onto the African social context (Oyewumi 2005, 100; Mohanty 2003; Steady 2005). These scholars suggest that western approaches assess gender relations through dualistic frameworks which map binary categories such as private/public and dominant/subordinate onto more nuanced cultural forms. Importantly, they point to the tendency to fra me all African women as -powerful African male. Such representations of a homogenous group of oppressed African women with the same interests and experiences fail to ars further argue that in the African context, the experience of women must recognize the larger North/South political economic power structures that frame the micro -level power relations between men and women (Mohanty 2003). According to normative ideals reported by HIV -positive peri -urban women, gender roles family. Both male and female participants in my study assert that the husband has greater economic power w ithin the household. Resources provided by the head of the household include basic necessities such as building the home in the village or paying rent in town, feeding the family, and paying school fees. In short, community participants reported that the h usband is - 103 secure control over household resources, leaving many wom en limited space in which to leverage economic power. influenced by larger political economic changes in Malawian society which have limited many economic support for their families. Such changes may illustrate how macro -level political economic forces impact micro -level social relations. Historical and contemporary political economic changes resulting in poverty and urbanization have widened the ga experience. In the 1990s, structural adjustment policies and food insecurity fostered labor migration as heads of families struggled to secure employment. Men faced with few options for work migrate in sear ch of ganyu or day labor, or to work on tea estates to provide for their families. These political economic processes constrain normative expectations of the amfumu, leaving many men unable to provide for the economic needs of their family (Bryceson 2006). (See Chapter 2 for a full discussion of the literature review). When discussing the household responsibilities of men, one community member the lived realities of many low income men struggling to maintain their identities as husbands household responsibilities, the informant talks about his responsibility to provide for the fami ly: But the way things are in this country affect my family a lot. Without money a family cannot go on well, these days, money cements the family. (Male community member, Blantyre) 104 While these gendered divisions in decision -making and economic power stru cture these power relations which offer women opportunities to leverage their decision -making autonomy related to household management. Role of the Matrilineage As noted in Chapter 2, although the south is recognized as a matrilineal area, many of these matrilineal practices have declined and no longer function in the highly urbanized setting of my study. The matrilineal kinship system further shapes power differen tials between men and women, demonstrating the fluidity of gender inequalities and the dynamic nature of gender and power relations between men and women in the Southern African context. In Malawi, the historical context of gender and kinship relations rev eals a complex and contested relationship between husbands and wives within marriage, and within the institution of marriage in relation to the larger matrilineal system. It appears that larger political, economic, and social factors historically shaped so cial relations within the matrilineage, including the relationship between husbands and wives and their relation to the wider kinship system. Specifically, Western notions of marriage and family structures were mapped onto African family structures at a ti me when political and economic forces favored these new marital forms. In Malawi, Christian conceptions of the ideal family (Phiri 1983) clashed with matrilineal notions of the relationship between husband and wife (Vail 1991). Marriage has been increasing ly restructured as a relationship between the husband and wife rather than a relationship between respective families (Phiri 1983). 105 While husbands often have greater decision -making power within the household, members of the kinship network may counter thi s decision -making authority, particularly within matrilineal kinship systems. For example, marital support is often solicited from the ankhoswe if problems emerge in the marriage. It is important to note that this is one of the few remaining roles for the ankhoswe . Most other roles are diminished in the favor of the husband. Community members and community leaders such as village chiefs explain that husbands and wives typically address marital problems by firs t attempting to solve the problem by themselves. If the couple is unsuccessful, they then seek counsel from the ankhoswe , or marriage counselor, or from marriage counselors within the church. If the problem remains unresolved, the couple seeks assistance f rom the traditional authority. The ankhoswe is known as the marital advisor within matrilineal and patrilineal kinship systems. These members of the kinship group have significant influence over marital relations. Ankhoswe are initially responsible for con tracting the marriage through the exchange of gifts or chikole . The primary responsibility of the ankhoswe is to assist the couple with marital conflict. There are two marital counselors within a single household banjaone from each side of the family. If a husband is accused of causing problems, the wife will seek support from his ankhoswe . Similarly, if the wife is accused of causing problems, the husband will seek support ankhoswe (Van den Borne 2005). However, while the majority of the m en and women in this study come from matrilineal kinship groups, many of the gender relations described by informants reflect a shift in matrilineal kinship structures, where vestiges of the matrilineal system persist in the context of urbanization which h as altered the practice of matriliny. The majority of the women interviewed in this study came from urban, peri -urban, and semi -rural areas in the Blantyre district of Southern Malawi, 106 where the matrilineal system and residence patterns have been changing over time (Phiri 1983; Desai and Johnson 2005) and matrilineal practices have broken down, thus reshaping kinship relations. My data demonstrate the malleability of matrilineal kinship structures, and illustrate how matriliny is functioning in peri -urban a reas of Blantyre, where men and women have fewer opportunities to earn a living, particularly for those with little education. Consequently, these economic and social s upport. In this study, one of the matrilineal structures that has remained is the support from the had not moved far from their matrilineal home, their mother ofte n served as PMTCT guardian and were relied upon for emotional support. Gender and Household Production Female Gender Roles husband before he leaves for work. She then prepares food for the children and prepares them for school. After c aring for husband and children, she begins her daily chores, or goes to work if she is employed. Daily household chores may include a combination of the following: hauling water, cleaning dishes, sweeping, and gathering firewood, gathering or buying food. For women living -rural villages surrounding Blantyre, a typical day may also include cultivating maize, cassava, ground nuts and other crops in small gardens during the growing 107 season from January to March. Male and female info re responsibilities as young girls. Informants explain that young girls are assigned childcare responsibilities to ease the caregiving burden of mothers with multiple childr en. Most girls learn caregiving and housekeeping responsibilities at a young age from the mother, who models these tasks. In addition to caring for their husbands and their children, women are responsible for providing care for sick members of the extended family banja to maintain linkages between the marital household or banja and the extended family or lineage. rs in rural and urban areas. Informants claim that household responsibilities are more likely to be shared in town, particularly if the wife is employed. In villages have assistance with childcare. Division of Labor To understand the power relations between men and women in marriage in this peri -urban and urban area, it is necessar y to first examine control over production 3 and reproduction within the household. Division of labor within the study communities aligns with the concept of !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!3 I define control over production to include access to and control over resources such as land and labor. 108 sexual parallelism or complementarity in which men and women have separate, but parallel household responsibilities. This division of labor involves an interdependence between men and banja household. For example, in the households in my study, husbands are responsible for clearing fields for cul tivation while women Women typically have control over domes tic responsibilities, maintaining the home with domestic tasks such as cooking, cleaning, fetching firewood and water, and preparing water for baths (Crehan 1997). As reported by PMTCT patients, a typical day consists of prayer, cooking, cleaning, collecti ng and preparing water for baths, none of which is controlled by the husband. The concept of sexual parallelism, therefore, reframes unilineal concepts of power by suggesting that men and women, in the domain of household division of labor, may have differ ent ideologies and interests (Poewe 1981), offering a more nuanced understanding of power relations between husbands and wives in Malawi. Access to Land In matrilineal communities, women typically control key resources such as access to land within a matr ilineage. In Malawi, women of matrilineal descent have rights to land through the mbumba or female members of the family. Women, husbands and children typically cultivate their land with some families hiring outside labor if they are financially able to do so. Women often manage the daily labor of children and outside laborers (Peters 1997). However, when land is scarce, women from matrilineal kinship groups with access to land do not benefit from these kinship rights rather than become more dependent upon members of the family who earn a wage 109 income (Davison 1988 ). This dynamic reflects a shift in political and economic relations in Malawi, in which land tenure reform coupled with the diminished size of landholdings restrict ary land rights have been upheld, creating a climate of ). Gender Roles and Socialization In this peri -urban and urban sample, gendered divisions in labor are often learned early in childhood when fathers typica lly mentor sons and mothers mentor daughters in gender normative attitudes and practices. Girls are expected to help mothers with cooking, fetching water, gathering firewood, and caretaking younger siblings. Boys, in contrast, are expected to learn the trade of their fathers. These normative gender roles are then reproduced during the main rites of passage within Malawian society. In rural areas, girls learn gender ideals for women in initiation ceremonies such as chinamwali . Girls whose families no longer adhere to traditional chinamwali attend Christianized constructions of traditional chinamwali ceremonies within the local church. institutions as the chinamwali . Chinamwali are the rites of passage conducted for young girls at women described a range of values conveyed within the chinamwali ceremony which relate to a tion to womanhood. Sexual purity and personal hygiene were the most common values imparted during chinamwali. hygiene during menstruation. Christian -based ceremonies reinforce the importance of refraining from 110 initiation ceremonies stress the importance of respect for elders, particularly parents, who are to n rituals signifies the historical with the pursuit of education. However, for many of these women, poverty, kinship responsibilities and gender inequalities interacted to prevent their continued education. Community informants, enacting their roles as mothers, shared aspirations to reinforce the importance of education to their daughters. These w omen claimed to downplay early sexual relations when mentoring their young girls. While these women are reproducing the values espoused by the Christian church, they may also be reinventing their own lives that were once redirected due to limited economic opportunities within the family. In urban areas , members of the kinship group and friends celebrate the upcoming wedding of the bride and groom. A central component of the ceremony is the transfer of normative values and ideals for shaping the subjectivities of women as new wives. Women are provided with a range of domestic skills within the househo ld as well as the importance of being an attentive wife. Women are taught how to maintain a clean home, how to provide satisfying meals, and how to are also dis seminated and reproduced within this premarital ceremony. Often conveyed with humor, women are advised to maintain an attractive appearance in the face of potential threats to 111 ability to secure a stable marriage. within society. Chinamwali ceremonies prepare young girls for the next stages of their development, shaping their future roles as mothers and wives. By imparting values of respect for parents and other elders, young women are prepared for navigating future social relations within their own kinship group as well as the kinship group of their husband. Proscriptions for behavior in marriage are again reinforced by marriage counselors within premarital counseling sessions at the church. At a private session prior to the kitchen party, chu rch elders, alangize , or family elders, guide men and women to not stray from the marriage, strategies to mitigate conflict. During this session, the future bride and g room are also instructed how to cultivate harmonious relations with in -laws and the wider kinship group. Gendered Dimensions of Education and Employment In Malawi, access to education is gendered, with boys having greater opportunity to pursue an educatio n. This is particularly relevant in families with low socioeconomic status, where the ability to provide education to all children is limited (Government of Malawi 2011). In ation results in fewer economic opportunities, thus positioning marriage as an opportunity for economic security. education were described by community women unable to finish the ir education. Women who 112 worked in domestic labor in Blantyre report that their education had been cut short due to poverty parents who were unable to provide school fees for their daughters, and uncles unable to provide school fees for their bvumwe, or sis opportunities arose when education was no longer financially attainable. These women lamented their shortened education, but remained confident in their agency to reinvent different life trajectories for their daug hters, envisioning daughters who will complete their education prior to As a mother I have the responsibility of teaching my children about our culture, and good behav ior. For the matured girls, I teach them how to dress well, more especially when they are in periods. I also tell my girls to avoid making friendships with boys because they can impregnate them and make them fail their school. (Female domestic worker, Blan tyre focus group discussion) In contrast to the experience of the PMTCT patients in my study, professional community women and PMTCT nurses describe more fluid gender roles that are structured by class. Many childcare, cooking and cleaning while they are at work. Professional women participating in community focus group discussions claimed that when they return from their professional jobs, the maid may have already prepared the evening meal, whereas the woma rural areas, both male and female informants claim that women spend the majority of time on the farm. 113 Part II: Gender and Power Control over Reproduction Women may also derive power from their positionality as mothers (Long 2009; Walker 1995). This is particularly true for women in matrilineal societies where children become a major source of labor and political power for the matrilineage, and husbands have little control over reproduction. Among both the Bemba and the Luapula, if a couple is unable to have children the woman may divorce the husband and will take the children with her to her matrilineage (Poewe 1981). Poewe notes that the Bemban and Luapulan communities explicitly differentiate between Such opportunities to challenge the source of infertility were absent among the PMTCT patients I interviewed in Malawi, where f can be diminished if she is unable to reproduce. The notion that infertility may be located in the reproductive potential of the husband was largely absent (see Chapter 4, Part II: Infertility as Gendere d Embodiment ). These women lack the power to demand divorce and find another husband with whom to have children. Moreover, some women may be replaced by another wife or a fiisi (male surrogate) will be solicited if they are unable to have a child. If it is believed that a woman cannot provide a child for her husband, he will seek another wife, or sometimes negotiate to obtain the reproductive capabilities of her sister. For many of the PMTCT patients in my study, husbands often decide where a woman will giv e birth at home or at the hospital because they live far from their mothers. Husbands also strongly influence the number of children in the family. For those women who disclose their HIV -positive status to husbands, husbands decide whether or not their wif e will attend PMTCT programs to access ARVs. When asked why they discussed their reproductive health care 114 decisions with husbands, many patients claimed that husbands provided the money needed for them to pay for transportation and occasional hospital cost s. These gendered divisions in discussed. is further mitigated by kinship relations within patrilineal kinship systems. In patrilineal societies, husbands and mothers -in-law have greater decision -making power in regards to reproduction. Mothers -in-law within patrilineal societies may co -opt a woma daughters -in-law. They may have direct influence over the number of children a woman has, and may decide where a woman will deliver. Political Power In many African soci eties, women derive power from leadership positions such as religious leaders, or village chiefs (Crehan 1997; Poewe 1980; 1981). Women also have access Chinamwali ceremonies are conducted by women as they alone are believed to have the cultural knowledge related to womanhood that must be imparted to future generations of young women. In addition, in communities where traditional birth attendants (TBAs) rem ain revered, women in these positions are able to leverage status within the wider community not only in their role as healer, but also as spiritual leaders. Within many African societies, traditional African religious systems historically assigned politic al and spiritual power to TBAs and other healers due to their symbolic role in connecting the larger community with the ancestral world 115 (Livingston 2005). In recent years, TBAs have lost much of their cultural power as the Malawian state has censored their maternal mortality (Bisika 2008). Relationship Power Relationship power is another key domain within gender relations from which I analyze gender inequalities. Situated at the inte rsection of gender, sexuality, poverty, and decision - pronounced and most relevant to an analysis of how gender inequalities interfere with the health seeking of HIV -positive mothers. relationships in Malawi. I analyze relationship power across the three dimensions within Co -making Economic Dependence ich This dimension of relationship power demonstrates how gender inequities are produced by unequal access to education and employment, resulting in limited economic p ower within the kinship and class shape gender inequalities within marriage. While earlier in the chapter I discuss how within the families of PMTCT patients, men oc cupy separate, interdependent, and 116 complementary positions of power within the household division of labor, it is clear that for -making authority is rooted in his control over access to economic resources, as movem ent to urban areas distances men and women from their kin. Many PMTCT patients claim they often rely upon husbands for access to money, linking household decision -making with economic power. Specifically, the majority of husbands provided transport money. For those women who disclose their HIV status to husbands, they reported that husbands withheld transport money because they were angry that their wife was HIV -positive. Other husbands withheld transport money because the couple was discordant. For those who do not disclose their status, they must find alternative means to access cash. Husbands typically control access to cash in the household. Thus, the decision -making au thority and rtunities. Among women participating in this study, economic dependence differed by social class, particularly differences in education and income level . Interviews with community members clarified the gendered dynamics of economic power within marriage. W omen who work in income generating positions have more negotiating power in their relationship, and may be less fearful of divorce. For example, while discussing decision -making within the household, community members from the professional sector, who enjo yed greater economic freedom than leverage decision -making power and satisfy their material needs within their relationships. These women report using charm to co -opt the 117 strategies were professionals who reported having greater autonomy in their ability to make decisions wit hin the marriage and their ability to demand divorce. Some of these women demanded divorces from husbands who physically abused them. As they displayed their scars to me, they recounted agency, their independent decision to leave the marriage and, eventual ly, secure new husbands whom they happily married. When asked who usually makes the decisions in marriage, these women answered: It depends... mostly the husband, according to our culture the husband. Nowadays because of democracy... the woman is in charg can make decisions. When I say I want my husband to buy a suit, it is just because of that charm that it is working. (Female teacher, focus group discussion) These linkages between money and marriage were looked down upon by other community women who clearly disapproved of this behavior, and who constructed these wives persuasion to secure luxuries in marriage, take from their future hu sbands the money or status to promote their own personal mobility. Sometimes in town people get married not because they love the person, but because of what the person has. There are such kinds of marriages where the person is just there to steal. (Femal e teacher, focus group discussion) The relationship between intimate relations and money has been widely discussed in the literature on intimate relations among African men and women (Cole and Thomas 2009; Cornwall 2002; Jewkes and Morrell 2012). Women en gage in affairs to access cash for basic unable to obtain from their husband. Transactional sex has become increasingly normalized for 118 women unable to find sustainable work in an economy damaged by economic decline. In fact, transactional sex is one means by which some women can leverage economic power and social mobility (Watkins 2004; Jewkes and Morrell 2012) despite the challenges of the local political economy. This economic leverage becomes a way for women to gain immediate access to money without having to negotiate the macro -level barriers of low employment and micro -level barriers to household decision -making authority. In Malawi, the differential acce ss to education and income has contributed to a tition for husbands (Peters 2013 ). In the previous chapter I argued that community -based stigma from female peers may reflect an economic strategy by which women seek to protect their marriages from pot ential threats. Non -HIV -positive women may be deploying stigma to their own advantage by preserving their position within the social hierarchy. The fear of losing intimate relat ions (Cornwall 2002). This ambivalence may reflect a tension between idealized values and beliefs within the institution of marriage and their lived realities. This tension may also reflect the changing structure of marriage in difficult economic times as both men and women seek alternative economic means to ensure the stability of the family, whether it is through migrant labor, ganyu , or transactional sex. The blurring of boundaries between marriage and money again emerges in a recent study of the use of charm and other discursive strategies as leveraging mechanisms within marriage. In 119 men co -opt the love of their husbands through persuasive discourse that may or may not reflect true emotion, illustrating the performative nature of love when money is at stake. The notion that some women challenge the power inequities in relationships by deploying agency in intimate relationships (Jewkes and Morrell 2012). Jewkes and Morrell describe a nuanced understanding of power inequities in intimate relations in which young women leverage control by managing multiple boyfriends and carefully selecting those relationships that procure their material needs and facilitate greater social mobility. These young women carefully leveraged relationship power by assessin g their male partners on the potential for social mobility by the gifts they received and continued or discontinued their relationships accordingly. Decision -Making Dominance productive resources confers decision -making power to women within the household (Crehan 1997; Poewe 1981). In Malawi, this decision making power increases with education level and wealth. Joint decision -making is more common among urban women living in ur ban areas, who have a secondary education, and women in higher income brackets. In this study, such decision -making power was more common to professional community women rather than PMTCT patients. According to the Malawi 2010 Demographic and Health Survey , men who report that wives should participate in decision -making are most likely to be older, with higher levels of education, and have higher income levels. Decision -making is also influenced by education. 120 Women without any education are least likely to be the primary decision maker in the household (31%). However, 48% of women with more than a secondary education are the main decision makers in the household (Government of Malawi 2011). -making power is also impacted by income level. For women in the lowest socioeconomic bracket, 53% report that husbands are the primary decision makers. In contrast, only 24% of women in the highest quintile claim that husbands make the majority of decisions. In terms of how income is spent, 40% of women cl aim that husbands decide how their earnings are spent, whereas 68% of women whose husbands have cash income claim that the husband decides how his earnings will be spent. Regarding decisions related to healthcare, 44% of women claim that husbands make thei r health care decisions (Government of Malawi 2011). In a recent study of Malawian women living in a rural region of Southern Malawi, 85% of In this next section, I discuss the social structure -making Connell 2012). PMTCT patients in this study describe the husband, bambo , as the amfumu, or who makes these decisions, PMTCT patients and community members constructed men as the primary decision - there to listen to what the me rural marriages where most patient informants held little economic power when compared to that of their husband. 121 Several informants legitimize the decision -making authority of the husband by invoking the Bible, employing Christian discourse on the institution of marriage. Representations of the offered by both male and female informants. For the PMTCT patients in this study, husbands are responsible for most household Of PMTCT patients who partici pated in individual interviews, over 40% reported that husbands made the majority of decisions within the household. It's like in our culture, a man is taken as an overseer of everything. The wife has to consult the man about what is taking place. Every d ecision is to be by a man. (Male community member, Blantyre) Because here in Malawi, we women are supposed to respect our husbands. So the husband is the head of the family and the decision maker and the breadmaker. Whatever we do, we have to respect them . So even in the form of taking the medication we have to ask him first. If they say no then we are not taking them. (Female health care provider, Blantyre) Some PMTCT patients redefined the decision -making authority of husbands, asserting that the husban amfumu as the sole decision -maker, highlighting the potential gap between normative gen der spaces in which they may assert agency within their relationships. Decision -making dominance is particularly useful in examining where power inequities inte disclosure to a spouse as a participant in PMTCT programs. This social structure helps explain 122 why many poor, peri -urban patients fear HIV disclosure, perceiving few options for social and material security outside the marital relationship. Negotiating safe sex is widely discussed in the literature on gender and HIV/AIDs (Bwirire et al. 2008; Hunter 2007; Iliffe 2006; Jewkes and Morrell 2012; Schatz 2005; Susser 2009). While I did not include questions about safe sex in my interview guide, I noticed that the issue was of concern to some women who participated in the study. While conducting PMTCT patient focus groups, I noticed a pattern of questions from the women who particip ated. After focus group discussions concluded, women were invited to ask questions of the nurse husband. While this question was not a part of my interview guide, it points to an inability for HIV -positive poor urban women to negotiate safe sex. to her husband, reflecting the absence of decision -making power with which to mainta in control over the relationship. Both social contexts, in fact, require a husband and wife to discuss the diminished in terms of her ability to communicate needs because of the potential for conflict, particularly divorce. When I asked PMTCT patients whether or not their husbands were HIV -positive, many (42%) did not know the HIV status of their husbands. Some of these women reported asking their husband to g et an HIV test, but were unsuccessful. Several informants noted the gendered consequences of divorce, highlighting economic disclose an AIDS status because they fear be 123 identity as a wife. The economic effects are most salient, as many PMTCT patients rely upon their husbands for economic sec urity. Faced with financial hardship, many women have little means to fulfill their mothering roles. As a result, many leverage economic security for themselves and their children by positioning themselves within economically stable marriages. Health care providers and community members cited poverty and the need for material support to care for children as the number one reason women may discontinue with PMTCT treatment, ey are Divorce here means the husband says I have nothing to do with you, you are not my wife. So the husband goes the other way and the wife goes the other way. The man is free to marry. The lady is free to marry. Usually the man can get another wife but the lady usually cannot have other children. She has other children, so no man would be happy to most of them it's hard to get another husband. It's very, very rare. (Female health care provider, Blantyre) The thing is, the fact that most men are breadwinners and those ladies are just will not have food for m women. (Female health care provider, Blantyre) Women bear not only the economic burden of divorce, but also the social one. In rural areas, women who are divorced have a more difficult time ge tting remarried, as many men choose not to assume economic responsibility for the children of another man in their second marriage may be assigned a lower status, as that of the second or third wife. 124 The potential to remarry for men, power to support a family rather than to a history of infidelity. Men who have the financial means to support a new wife are deemed marriageable because of their economic power, potentially fulfilling loc al understandings of client -patron relationships. According to norms in many rural communities, men with economic power are often obligated to share their wealth construc Thus, the consequences of divorce may be greater for women living in peri -urban areas who have few economic resources in which to maintain the health and well -being of their family, without the economic suppor t of their husband. This is particularly true in peri -urban areas where many women are more economically dependent upon their husband. As a result, women may lose a husband upon HIV status disclosure while husbands remarry with relative ease. The sexual fr power to remain in the marriage when asked to leave, to demand fidelity when th e husband has strayed, and to secure social and economic support to manage AIDS within PMTCT. Social Norms relationship power (Connell 1987). Social norms related to s exuality help explain the power discussion between husbands and wives of outside partners and questions of fidelity within the marriage. 125 Fidelity is a normative ideal e spoused by most individuals and reproduced within major social institutions within society. There is a gap, however, between this cultural ideal and its everyday practice (Watkins 2004). The problem of fidelity within marriage emerged in several discussion s with community members responsible for disseminating and reproducing ideals within marriage. When asked to comment on the most common concerns addressed in marriage oncern expressed by villagers and church - - nd engage in multiple partnerships as disruptive to healthy marital relations. Similarly, women are advised not to In their discussions of marriage, community informants highlighted the normative value of fidelity between husband and wife. While fidelity is an aspiration for both men and women in likely to be sanctioned. The ankhoswes say , if a man has cheated, you must have done something wrong. Maybe provider) W to gendered understandings of fidelity. Health care providers and community members consistently claim that having affairs is condoned for husbands. Yet women who have affa irs 126 ity to secure multiple partners is a marker of masculinity (Kaler 2003). The notion of kawerekawere appears to have a double standard for many women. Informants offered that there is a double standard wh bambo and keep it a the family: For example, if this one is my husband and he finds another girl and I know it, that he is provider) But it is our culture for women to love their husbands more, and to keep their secrets to have a good family. If you have problems in the house When the man is okay, (for) this lady, it's the end of the marriage. On the part of the husband the lady can take care of the husband. On the part of the wife, it is the end of the marriage. They say this one is a prostitute while maybe she has caught this disease while taking care of other patients. But th ey say this one is a prostitute. (Female community member, Blantyre) secrec 127 stability of marriage to secure the welfare of their children by mainta ining a family structure that rategic preservation of marriage procures economic stability for her children, indexing the social leverage gained as women pursue identities associated with motherhood that are reinforced by moral dis course on infidelity ( Jewkes and Morrell 2012). These w omen strategically recognize the pivotal social and economic role of their husband, and thus their identity as wife, encouraging women to avoid divorce at all costs. Fidelity Intersections with Local Masculinities by community members and some patients as a reason husbands would not want their wife to attend the PMTCT clinic at the hospital. One did not continue with PMTCT. Claiming that her husband believed that she was meeting another man at the clinic, this patient became one of thirteen women defaulters. The assumption that some wives are not to be trusted emerged within a focus group discussion of community men w It depends on the beauty and the way you trust your wife, you can let her go alone if you doubt her behavior . Some men escort their wives for fear of other men proposing l ove to them. Sometimes one needs to escort his wife depending on the end you want to achieve after attending antenatal services together with his wife. (Male domestic worker, Blantyre) 128 While local norms of masculinity grant husbands more freedom to pursue multiple partnerships, in certain social contexts, having multiple partners may be socially rewarded. In her study of cultural perceptions towards condom use, Kaler (2003) found that young men in rural Malawi make claims to being HIV -positive to their pee rs to assert their masculinity. These men gain currency with their peers when claiming to be HIV -positive, even if they have not been tested. Such claims point to the social pressure placed upon men at a young age to have multiple partners to enact masculi ne identities. informants. Both in discussions related to spousal participation in PMTCT programs and in -seeking practices, informants asserted that men are expected to associated with physical strength. Info rmants claim local masculinities link sickness with attend antenatal clinics with wives seeking reproductive health care such as PMTCT. Most of the cultures re gard the woman as a little weaker than the man. If a man says he is sick, people will say you are strong even if he is feeling pain. (Male health care provider) expected to be may lead some men to demand divorce when a wife has been unfaithful. A study conducted by Watkins (2004) on transactional sex in rural Malawi found that men who are awar infidelity are expected to leave the marriage and request a divorce. When asked why a husband 129 would be angry if his wife announced that she was going to the PMTCT clinic, one man responded: Some think that the wife was unfaithful to him or e lse if the man was unfaithful to his wife he could think that he is the one who brought the virus in the family, but if you are mind is that his wife was promiscuous bec ause extramarital unprotected sex is the major way to transmit the virus. But this is not to underscore the fact that there are other ways of contracting the HIV virus. (Community men, Blantyre focus group discussion) While men were more commonly cast as the marital partner more likely to engage in class admitted greater freedom when discussing their sexuality as compared to PMTCT patients. Informants rarely discuss ed women as engaging in outside affairs, but when probed further some who fail to adhere to normative gender ideals which ensure the preservation of the family. Watkins (2004) argues that the disconnect between the cultural ideal of fidelity and the practice of multiple partnerships points to the cultural practice of ties of depe ndence, which involves reciprocal exchanges in which an individual with economic power provides material benefits to an individual in economic need. These relationships are locally understood as forms of social security, providing individuals with the soci al capital that may help them in times of need (Watkins 2004). unlike when a wom an does go out and has an extramarital affair, that woman is not very serious. (Female health care provider, Blantyre) 130 Accommodation I have tried to demonstrate the shifting nature of gend er and power relations by deconstructing gender inequalities across various dimensions of power: economic, reproductive, political, and relational. I have shown that power inequities for the women in this study differ across class lines, and emerge primari ly in the social structure of relationship power in the context of a changing political economy. I have explored how these gender inequities at play in marriage help explain the attitudes and behaviors of the PMTCT patients interviewed in this study. Resis tance to idealized norms may be expressed in subtle ways. In this study, contact with women was primarily through the clinic, after which I was unable to contact them due to the sensitivity of their HIV/AIDs status. However, I did have an opportunity to in terview domestic workers and professionals in the community through focus group discussions. While the majority of HIV -positive women I spoke with struggled with the power inequities in their lives, I encountered a handful of women who enacted agency by le veraging relationship power within 231; Foucault 1977) occurred infrequently among PMTCT patients and more frequently among the small group of professional women I i ncluded in my focus group discussions. Below I Higher -status professional community women with greater economic and decision -making power were able to more directly challenge potential power inequities within their relationships by deploying power discursively. These women divorced husbands when the relationship was not meeting their needs, employed discursive strategies to get their social and 131 economic needs met in the relatio nship, and exhibited greater freedom with discursive strategies, joking about husbands as we chatted before and after the taped focus group discussions. The higher po sition within the social hierarchy as compared to PMTCT patients I interviewed. Conclusion HIV/AIDs management tell us about the daily lives of Malawian women? 2) How do th e barriers es? To answer these questions, I reviewed the normative gender roles for men and women within Malawian society. I discussed how these constructions contribute to or mitigate against gendered inequalities within the institution of marriage. I argued that po wer differentials within marriage and the decision -making authority conferred to husbands lead many women to remain in marriages regardless of questions of infidelity. Yet the social and economic consequences of her marriage, and preserve her identity as wife. caregiving role central to her maternal identity. This chapter considers the tension that emerges of motherhood and wifehood. Many women conceptualize attendance within PMTCT programs as central to fulfilling the mothering role, yet fear being identified with a program associated with an AIDs identity. Consequently, PMTCT participation has a social cos t in which women may risk losing their marriage an institution which procures social and 132 who represent the majority of those in the study population. 133 Chapte r 6: Local Understanding of HIV/AIDS and Community -Based Stigma programs, it is important to look at how HIV/AIDs is understood within the study community, and how these c and wives. In this chapter, I examine how community -based stigma, the second -most common of sexuality. -related stigma, I explore how . PMTCT patients and community members report that HIV -positive women hesitate to disclose their HIV status to the community because of the stigma they experience. This stigma, manifested in gossip about these women, is fueled by local understandings of HI V/AIDS that construct women with AIDS as immoral or promiscuous. Exposing an HIV -positive identity within the public sphere may bring unwanted attention to the problems and conflicts within the household. HIV disclosure, ability to enact the sometimes conflicting identities of wife and mother, placing them in a double bind. be used as a lens to understand how gender roles and relations affe ? This chapter is divided into two sections: 134 periences with HIV/AIDS -related stigma in the chapter, I STDs and contemporary experie nces with stigmatization and HIV/AIDS. I discuss the relationship between gender inequities and local understandings of HIV/AIDS as they intersect es with HIV/AIDs stigma and their responses to stigmatizing attitudes and practices encountered in the household, kinship group, and larger community. Part I: Defining Stigma when a member of a social group who embodies an attribute perceived to be different from the norm falls outside of culturally -constructed categories of normality (Goffman 1963). Goffman argued that attributes that are incompatible with culturally -defined categories of normality become so urces of discrimination Social scientists have criticized this classic sociological understanding of stigma for its essentialist focus on individualized traits of the stigmatized and individual practi ces of those who engage in stigma production. This essentialist perspective problematizes individual actions isolated from their larger social contexts, thereby erasing the processes by which stigma is produced. Defining stigma more broadly, social scienti sts grounded in the theoretical constructs of structure and power argue that stigma must be recognized not as an entity, but as a social process in which certain social groups construct difference, thereby reproducing moral values within 135 society (Parker an d Aggleton 2002; Sontag 1989). Drawing upon Foucault, Parker and Aggleton (2002) note that stigma production involves the leveraging of power against particular social groups with histories of marginalization. For persons with HIV/AIDS, associations with sexuality evoke a moral discourse directed towards individuals represented as sexually promiscuous. Within regions of the Global North, these narratives of blame have implicated homosexual populations for causing the HIV/AIDS epidemic (Parker and Aggleton 2 002), while in the Global South such narratives are often gendered, centering on the supposed promiscuity of women (Nyblade et al. 2003). -Based Stigma old level interact in which women are more likely than men to be at risk of being stigmatized, experience stigma differently and experience greater consequences of stigma, thus reproducing the inequalities women face in their daily lives ( Van Hollen 2013) . Women are more likely than men to be at risk for being stigmatized for several reasons. This is in part due to gender relations, power differentials within marr reproduction and sexuality . HIV/AIDS discourse in rural Malawi. With its roots in sexual practice, an HIV/AIDS diagnosi s reveals family secrets either the sexual promiscuity of the husband or wife. Participants in my which an AIDS diagnosis makes public the secret relationships co nducted outside the intimate 136 as a wife. Women are more likely to be blamed for bringing AIDs into the family due to the different standards of sexual morality for men and women. As discussed in Chapter 2, these gendered standards may in part be due to historical constructions of the family rooted in Christian discourse on the nuclear family that compelled women to uphold the moral representations of their husband ( Vaughan 1991). Several participants invoked Christian discourse, associating an HIV/AIDS identity with a husband or wife who has sinned. As one community participant informants related g integrity of the family. The feeling that people has is that if somebody is in this state, he or she is a sinner. She may have contracted that through sexual intercourse. So the feeling that people have way to other places so people cannot know their status. (Female community member) Stigma and the Structure of PMTCT Programs Women are also more likely to be blamed for bringing AIDS into the family because of the structure of PMTCT programs. Integration of PMTCT services within Reproductive Health HIV during the initial prenatal care visit, leading most pregnant women to automatically receive 137 participating in PMTCT programs. This immediate provision of HIV/AIDS testing and counseling (HTC) at the prenatal clinic consequently provides women with a diagnosis often well before the husband has had an opportunity to be tested himself. Husbands are actively encouraged to accompany their wives to the clinic. However, few men initiate HTC. This was confirmed by over 40% of the participants in this study who did not know the s tatus of their husband. Encouraging husbands to accompany wives to the public prenatal clinic is reported by nursing and clinical staff to be an ongoing consistent challenge for many health facilities. As a result, most women who attend the prenatal clinic for health services become aware of their many women vulnerable to blame and stigmatization as the only spouse within the marriage with a positive status. Several informants reported that some m en who claim to not know their status actually do know it, but do not disclose it to their wives. These men are believed to seek treatment at private health facilities that protect them from unwanted public disclosure. Thus men able to afford care at priva te facilities have a greater ability to manage the public exposure that contributes to stigma production, reinforcing the gendering of stigma. That is the main problem in Malawi, they hide it. Most of the men go in private while we ladies go to Queens. Me n go to private hospitals. (Female community member) The majority of PMTCT participants express concern that attendance at the hospital means their private lives will be revealed. In their own words, several current PMTCT patients participating in a focus Women view the hospital as the site at which their HIV/AIDS status is first disclosed to the community. As a public facility, there are many opportunities for others to learn of their posi tive 138 status. Some women meet their neighbors when waiting in line at the ART clinic. Others attend Moreover, the rooms that have been set aside for private one -on-one counseling are known by claim other pregnant mothers within the prenatal clinic gossip about why PMTCT mothers enter a different room to receive medication, claiming status will be disclosed to the local community. Clinical observations confirmed the allegation ely to be disclosed in the clinical setting. Despite integration of PMTCT into reproductive health services, some services dedicated to AIDS treatment remain separate from other clinical services, taking place on specific days, for example, or within a unique room dedicated to AIDS patients. As a result, PMTCT patients maintain that their identities are easily revealed. Enacting the caretaking responsibilities of future mothers places women at risk of being stigmatized for their positive HIV status: Yeah o f course we are coming to the antenatal clinic. They just take you from the room and they say come to that room. As for us, we don't feel good. Those people who are staying there, (they say) why do they go into that room? Because we Malawians, we don't rev - themselves why are they going into that room. It needs privacy. (Current PMTCT patient) The gendered experience of HIV/AIDS stigma is supported by experiences within PMTCT programs in other countries within the Global South (Van Hollen 2013; Nyblade et al. 2003; Ogden and Nyblade 2005). Van Hollen (2013) found that rural Indian 139 women who were HIV -positive mothers experienced signi ficantly greater stigma than their HIV -positive husbands. These women were committed to a shared goal of avoiding the social and economic costs of divorce. They employed a number of strategies to hide their positive status to preserve marriage and protect themselves from disapproving kin mothers -in-law, in particular who sought to protect the moral representations of the lineage and exercise newly -acquired status. In India, efforts to dismantle the gendering of stigma have become institutionalized. Health o fficials have changed the name of this global program from Prevention of Mother to Child Transmission (PMTCT) to Prevention of Parent to Child Transmission (PPTCT) to hold husbands accountable for their role in HIV transmission (Van Hollen 2013). These cha nges highlighting the global nature of this problem. A study of HIV/AIDS stigma in Zambia (Nyblade et al. 2003) found that community members point to HIV -positive women are accused of violating normative structures of women as wives and mothers by failing to enact their caretaking responsibilities and requiring care for themselves. Family members who are no longer take care of wives and mothers with HIV/AIDS. The study found that households experiencing poverty expressed the greatest stigma towards family members with HIV/AIDS (Ny blade et al. 2003). This study and others illustrate how stigma is produced when members of a kinship group are unable to meet their social obligations (Kleinman and Hall -Clifford 2008). In this study, some participants spoke to the stigma experienced by w omen in patrilineal kinship systems, arguing that they were more likely to suffer from discrimination from mothers -140 in-law who ill -treated them for bringing HIV/AIDS into the lineage and to their future offspring. Relationships with mothers -in-law are even more contested when women are unable to bear children. Several women in my study report knowing friends who disclosed their HIV status to that power differentia ls within relationships shift the blame from husbands to wives. As one health care provider states: Usually, HIV is linked to behavior in the family, so they think that if they tell the husband they are positive, they will think the wife is the one who is HIV -positive and is the one who infected the man. So instead of (disclosing) they just remain quiet. If she ar of disclosure may be influenced by the likelihood that they will be ability to enact wifehood, as she is likely to be blamed for bringing AIDS to the marriage. Accor dingly, some women assume this blame without complaint, manifesting the gender inequalities that often led to the initial HIV exposure. This process of internalization becomes an embodiment of the inequalities women experience in their daily lives, as many are often not the partner responsible for contracting the virus. When discussing the fact that many women are blamed for the HIV status within the couple, even if the husband introduced the virus to the relationship, a focus group made up of community men appeared shocked, and clearly did not identify with a husband who would blame 141 Gendered Embodiment and t he Clinical Phenomenon of Discordance The clinical phenomenon of discordance, in which one partner is tested positive and another negative, is an additional site upon which the gendering of stigma is reproduced. Informants report that discordance creates s ignificant confusion and marital conflict around who has the virus and who does not, and more specifically who will assume blame for the virus. Health care providers and patients maintain that a woman positioned within a discordant couple is likely to be b lamed for bringing the virus into the family if the wife has been found positive and the husband has not yet seroconverted, meaning that he has not developed antibodies to the HIV virus which allows for its detection. Discordance often evokes fears of infi delity when both partners are not found positive. While conducting participant observation at a rural ART clinic, I met one couple who had received their HIV test results. The wife had brought the husband in for testing because she had questioned his fidel ity. When the couple received their results, the husband tested negative and the wife positive, yet the husband had been the partner accused of infidelity. The wife appeared confused by the results, declaring they were not possible. When the health care pr ovider unconvinced. The husband, looking surprised himself, proudly co -opted his test results with the biomedical knowledge that temporarily exempted him from further acc usations of infidelity. And also there is a problem with the discordant couple the mother positive, the husband ive. Why is this a problem? (Interviewer) negative? Maybe she is seeing somebody. So, to start using those condoms and the like because they are advised somebody is negative just be sleeping together. (Female health care provider) 142 Forms of Stigma Participants in my study report multiple forms of discrimination toward women with HIV/AIDS. Common practices include ridicule and accusations t hat are often made public by members of the community. Several patients cited these forms of stigma as reasons to fear the public and private spaces where they may experience discrimination. These practices included mocking, name -calling, and pointing fing ers. Other informants suggested that persons with living with AIDS (PLWA) experience various forms of social exclusion that deprive them of public services or benefits. These included exclusion from receiving fertilizer coupons, and exclusion from receivin g loans to start small businesses. Some women experienced social exclusion structured by fears of contagion. Women reported that some community members continue to fear contagion, forcing some PLWA to eat alone, and as observed within rural communities, so me lose their homes when they are rejected by their families. While conducting participant observation within positive living support groups for PLWA, two women were seeking housing from anyone in the community. Family members tended family because of their positive status. Less commonly expressed, one woman reported that family members with which she lived would not address her medical needs, and even withheld food. Informants described the social exclusion associated with stig and no longer considered as members of the community. Below a twenty -year -old informant al process in which she is moved to the margins of society and an altered identity in that she is no longer perceived to be a member of the kinship group and, at times, is treated as less than human: 143 I love my mother so much because when I am sick she doe she receives a message from the church. She has a responsibility at the church. She tells oes not even care to come to see how I spent the day. The younger ones are the same. When my mother is away, I can spend three days without a bath, they eat their nsima right here without sharing with me, so I see that the person who loves me is my mother. (Female patient) women discussed above (Van Hollen 2013). As discussed in Chapter 5, local conceptualizations of masculinity allow men to engage with multiple partners. An HIV status for women, however, contradicts the normative moral representations they are expected to uphold. While exploring contemporary constructions of HIV/AIDS, I learned that a double standard of morality censors rs, yet condones that of the husband. I found that the nature of HIV/AIDS as a sexually transmitted disease evokes ideas about sexuality, fidelity, morality, and relations between husbands and wives concepts that are central to how marriage is understood a nd practiced within Malawian society. the PMTCT clinic signifies an HIV/AIDS identity to the community and subjects women to allegations of promiscuity from community members who perceive an HIV status as a violation of sociocultural norms. According to health care providers and PMTCT patients, good mothers are compliant patients they attend the clinic to collect their medicati on and treat themselves and their child. Yet these mothers must manage community -based stigma when their fidelity and morality are questioned upon HIV status disclosure, causing tension in their roles as mothers and wives. -based s tigma is well -founded, given local understandings of AIDS 144 -positive are labe mothers. The experience of stigma not only varies by gender, but also by class. Wealthier members of the community may have greater agency in their ability to hide their p ositive status from the larger community. Some informants discussed health seeking by professional men whose practices diverged considerably from the dominant demographic in this study women with - oy greater privacy in their treatment seeking. One informant described two church members who traveled to South Africa to access medicine privately. Similarly, a health care provider described a patient who arranged for special services to attend the clini c during off -clinic hours to collect his ARVs. One dominant strategy that took place across class lines was the practice employed by all patients accessing PMTCT or ART services at multiple clinic sites far from home to hide an HIV/AIDs identity. Position ality, AIDS Gossip and Performing the Good Wife PMTCT participants, health care providers, and community members stress the potential for community members to spread rumors or gossip about their AIDS status. PMTCT participants carefully defined those who w ere to be entrusted with knowledge of their status, and those who were not. The majority of women interviewed clearly articulated distrust of female community members mos t likely to engage in stigma production are other women. While holding a focus group discussion with current PMTCT participants, I asked women whether they told friends or members of the community of their HIV/AIDS status. A 145 chorus of all the previously qu questioned further, women claimed other women discriminated against, stigmatized, and -positive identity did not mock them and instead were sources of support. These women are understood to be a potential source of social alliance not only in their capacity to not stigmatize, but in their shared identity as HIV -positive mothers. I witnessed this during the focus group discussion. Women who met each other for the first time at the focus group developed friendships, shared advice, and sometimes returned to the PMTCT clinic together. PMTCT participants stressed their fear of being the target of stories discussed by The only friends who know are only those that are also receiving medications too, because we k others will just talk about you. (Current PMTCT patient) No, we are unable to tell friends. You know Malawians, this is the way they feel. If you are shaking hands, (they say) she is HIV -positive, she can transmit it to us, so we shouldn't chat with her. Most Malawians do not reveal what has happened to them when we are talking of HIV/AIDS. That's what we Malawians do. (Current PMTCT patient) Another participant described her shift ing positionality on the boundaries of AIDS stigma production, admitting she once gossiped about other HIV -positive women until she was found HIV - would never tell frie 146 Manyazi There are so many things, even with those that you chat with in your neighborhood, the chatting starts to change, and o Spreading rumors about the sexual practices of other women may reflect the construction of HIV - udy of HIV -positive women living in urban Nigeria, Cornwall (2002) found a similar pattern of social commentary on sexual practices of other women. These women would expr ess dismay, problematizing and reinforce a social identity of the good wife whose subjectivity is further realized when compared (Goffman 1963). As an individual and collective pro cess, the identity of women as mothers and 147 clinic (Cornwall 2002). This intersubjective process allows certain identities, such as the unfaithful wife, to gain tract ion when witnessed by other mothers and wives. Part II: Health Care Provider Attitudes within PMTCT Programs Just as women fear being constructed as promiscuous by friends and members of the community, they fear accusations by the health care providers who serve them. It appears that social class interacts with stigma by positioning women living in poverty as the recipients of stigmatizing processes at the clinic. Poor attitudes and worker burn -out lead to interactions between health workers in positions of power and HIV -positive new mothers living in poverty that result in further marginalization of HIV -positive women. This mistreatment, while not rooted in stigmatizing attitude s about HIV/AIDs, ultimately contributes to stigma production, thereby reproducing inequalities. This mistreatment illustrates the double bind when health care providers expect women to comply with treatment, yet punish them for being compliant when seekin g health care at the clinic. Unlike the wider community, health care providers do not blame their PMTCT patients for being HIV -positive. Many health care providers maintain that women obtained the HIV virus from their husbands. Further, many appear devoted to their work, employing multiple strategies in which to assist their patients, going so far as to help women disclose their status to their husbands. These providers go to great lengths to assist women in maintaining confidentiality related to their stat us. Some nurses enact new identities when conducting outreach, disguising stigmatize the woman in the community and disclose her status to the husband. Howeve r, the 148 women experience when facing stigma in their communities. Several women report negative interactions with some of the health care providers, claiming that n had been received by providers. Women report being shamed by nurses in front of other women, focus group discussion, some participants reported that he alth workers were stigmatizing them, - In a similar vein, patients expressed disappointment in the attitude of health workers and how patients were treated when attempting to access ARVs. Several women maintained that workers we patient load. One community member not only heard of such mistreatment, but witnessed it himself at a nearby clinic, noting that men are not immune to the stigmatization of AIDS patients. Below a teacher describes a conversation with a health care provider friend: Nowadays it's just an assumption that sometimes these women stop taking these drugs o are publicized. So sometimes even men, not only women, they stop taking the drugs becaus e of the way they are handled by the medical person here. (Male teacher) experiences with STDs during the development of the biomedical health care system in Southern Africa. He alth -seeking for the treatment of sexually transmitted diseases was a humiliating and 149 shameful process in which both men and women were ridiculed by colonial administrators for can men and ltivating shame in patients was 148). As discussed in Chapter 2, many of these attitudes persist today in the gendering of stigma experienced by PMTCT patients as m others and wives living with HIV/AIDS. L ocal the history of STDs in colonial Malawi (Vaughan 1991). I argue that manyazi (shame) has been reproduced discursively a nd is well alive today in the treatment -seeking of these women living with HIV/AIDs. primary context: when the patient attends the clinic late or at a time outside of designat ed clinic hours. Informants claim that health workers sometimes mistreat PMTCT patients when they do not show up at the clinic at the designated time. Observations at one facility confirmed that patients were scolded by health care providers when arriving late to the clinic. With few resources available, clinics are often held in rooms with limited space. As a result, private conversations are a challenge. Patients waiting to be seen by health care providers often fell silent when other patients were scolde d for clinical missteps, listening carefully to the nature of the accusations, perhaps to avoid future public shame. 150 Mistreatment of patients may signal the broader issue of resource limitations within government -funded health facilities. While many health care workers did not reveal the causes of their work stress, one theme that continued to emerge from interviews and clinical observations was the work burden faced by many health care providers within the government -run health care system. Often health ca re providers worked in circumstances in which staff numbers were lower than required. PMTCT nurses work on the forefront of care for HIV -positive pregnant and lactating e of the -positive status. These nurses play and psychological health. As agents within PMTCT programs, nurs es are responsible for imparting all information related to AIDS management in HIV -positive women and their exposed children. PMTCT nurses are powerful socializing agents. In addition to the health education provided to mothers, women are counseled on fami participation in PMTCT. PMTCT nurses guide women through many of the major stages of motherhood: proscribing maternal practices during pregnancy, delivery, and breastfeeding. In practice as mothers, nurses guide women through particularly sensitive issues within marriage infidelity. Claiming that many women refuse to share their HIV/AIDS status with their husband, nurses o ften guide women through the social and emotional challenges of communicating a positive status to their husband. For women who are afraid to disclose their status to their 151 husbands, nurses often offer to intervene on behalf of the mother and to tell the h usband of his this status. For some women, the PMTCT nurs e may be the only person that is aware of her condition. This sensitive role requires considerable trust in the health care provider and is often private worlds, aware o knowledge secret. While some PMTCT nurses evoke fear in patients, they also work diligently to promote the health of women and their infants. Many of these women have dedicated decades of thei r lives to a profession they deeply respect. Efforts to promote patient adherence are rooted in a professional and personal dedication to serve the health of HIV -positive women and infants. One nurse captures this sentiment in the following discussion of t he struggle to serve HIV/AIDS patients in 2005 a time prior to the provision of free ARV medication. Her comment sheds light on provider awareness of the failures of transnational development processes such as the provision of pharmaceuticals, whose sustai nability remains uncertain: For most of us, there is poverty here in Malawi. (Before free ARVs) only those who were us these drugs to continue. If they continue, we wi ll thank them. Do you think it will continue? Because if it stops, Malawians will die. (Female health care provider) surrounds the experience of p atients with HIV/AIDS (Van Hollen 2013). Efforts to avoid stigma 152 included traveling to distant clinics where patients did not expect to encounter relatives, friends, or members of the community. Unique to this group of women was the extent to which they st rategized to keep their positive status from the watchful eyes of husbands, neighbors, and the kinship group. PMTCT patients in this study used a number of creative strategies to keep their positive status secret, d their positive identity. For many of these women, the primary strategy employed to remain within PMTCT was to creatively hide a positive status and health seeking practices from husbands and their community, so that they could access treatment within PMT CT. Such strategies included hiding their ARVs in the flour bin or in holes in the ground outside the home, buying blank Health Passports (medical record books), and not selecting PMTCT guardians to accompany them to the clinic. The following practice was consistently reported by PMTCT patients and community members: taking the drugs. I have heard of people traveling from here to maybe (distant town) just to take the medi (Male community member) The most common strategy leveraged by women to avoid experiencing stigma and discrimination is the practice of hiding an HIV/AIDS status. Not disclosing an HIV status provides them with increased agency to negotiate the challenges of their daily lives. Manyazi or hy would a woman not Manyazi may not so much reflect a fear of stigma, but a fear of losing immediately, ensures her economic survival. 153 Fears of unwanted disclosure were confirmed when conducting participant observation. At the beginning of the study period, PMTCT participants were provided with a 2kg bag of flour (ufa) for their participation. After consulting with local colleagues it was beli eved that a bag that was any smaller than 1kg would be insufficient as a token of appreciation. When participants were offered the 2kg bag, all refused the flour, despite the low socioeconomic status of many of the women. Only when the 2kg bag of flour was replaced with a smaller 1kg bag which would not attract attention, did women willingly accept the flour. Several women who had initially refused the flour returned over the next month to collect it. PMTCT participants also remarked on the potential for st igmatization when carrying a Health Passport, patient medical files which bear the stamp of a positive status. Health care providers and community members claim that many patients buy an extra file one contains information about the HIV -positive diagnosis while the other does not. Having two Health Passports allows women to control who has access to her HIV/AIDs diagnosis. When asked by clinic personnel why information is missing, some women claim to have lost their file. A patient remarks below on how the Master Card used at the clinic reveals her AIDS status: these, they know we are here to receive medications. Just like how it is at the ART clinic, when they see you lined u how this can change. (Current PMTCT patient) Several health care providers commented on the confusion created when a patient presented a Health Passport (medical file) without their HIV statu sa file that had been bought for the sake of the current clinical visit. When I asked how women hide their status at the clinic, nurses and doctors claimed that patients buy multiple Health Passports, gaining some control 154 over their status disclosure. As While stigma production as an obstacle to PMTCT participation remains formidable, women engage in a number of creative c oping strategies to navigate barriers to PMTCT treatment. For some women, support from mothers, husbands, or church counselors provided the necessary support to ensure success with health seeking. For those women who disclosed their HIV status to their mot her, the mother served as a key means of social support. A few women maintained that mothers served as their PMTCT guardians, a buddy system in which PMTCT patients are asked to bring a family member, adherence. Women shared their HIV status and their health -seeking at the PMTCT clinic with few individuals. In this study, patients cited they needed support afte r receiving their diagnosis and initiating treatment. Several current participants cited husbands as a source of economic support. This was most relevant for women living in the rural region who self - those fortunate to have s upportive relationships, husbands were a source of emotional support. One current patient was unique in her claims to have been transformed by her HIV diagnosis. Moving from marital infidelity to marital harmony and religious atheism to religious devotion, she remains one of the very few informants strengthened by her HIV/AIDS diagnosis, and one of the few who expressed having a supportive spousal relationship upon disclosure. Citing support from a devoted husband, her relationship was transformed from one of marital infidelity to marital harmony after her husband confessed to having an extramarital affair. The 155 couple sought support from a local church and received marital counseling that strengthened their marriage and converted both husband and wife to rel igious practice. Attending positive living support groups is another means by which PMTCT patients could cultivate support to counter stigmatizing practices in the community. While visiting one Positive Living support group 4, I met a second patient who suc cessfully remained within PMTCT. When asked about how her life was different now that she was on ARVs, the woman recalled her initial failed health, and a corresponding inability to cultivate her home garden because of her ill -health and immobility. Shortl y after her husband left her, she joined a national -positive man who soon became her husband. While describing her recent good health and well -being since starting ARVs, she demonstrated a vitality and gratitude for health through a celebration dance in front of her fellow PLWA. While women appear to experience the majority of stigmatizing practices from other women, they seem to derive considerable emotional support from other HIV -positive mothers who share a collective identity as mothers and HIV -positive patients. Positive living support groups have emerged throughout Malawi to promote the principles of self -governance of PLWA by PLWA. Some of these groups provide, in a ddition to counseling by other PMTCT mothers, a forum in which HIV -positive new mothers can reconstruct shared experiences of gender inequalities, stigmatization, and other forms of HIV -related social marginalization. Focus group !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!4 Like other surrounding countries in southern Africa and across the global South, rights -based advocacy organizations (brought by NGOs) have emerged to challenge and mitigate the stigma experienced by many patients with HIV/AIDS. These organizations deploy community -wide campaigns to sensitize, mobilize and transform social perceptions towards HIV/AIDS. One of the underlying incentives of the programs is to encourage the community to test for HIV as well as to discourage stigmatizing perceptions toward the illness and more importantly towards persons living with HIV/AIDs. 156 discussions and conversati ons with these women were lively and often punctuated by women sharing information about employment opportunities and education opportunities. Conclusion In Malawi, as in other parts of the Global South, HIV/AIDS stigma has been a persistent social proble m since the onset of the HIV/AIDS epidemic. Early representations of HIV/AIDS often linked individuals with marginalized groups such as sex workers, who were often blamed for perpetuating the epidemic. I have found that local understandings of HIV/AIDS con tinue to be linked with stereotypes of prostitution and unfaithfulness. These stigmatizing notions of AIDs may persist because of the historical context and discursive power of biomedical discourse towards the treatment of sexually transmitted disease as a condition associated with promiscuity. These historical processes with the history of STDS contrast with historical and contemporary understandings of motherhood and wifehood, challenging these social identities. Consequently, the tension between these identities reinforce unequal power relations and community -based stigma toward mothers living with HIV/AIDS. In this study, I found that PMTCT patients relied upon more subtle forms of resis tance to the power relations that structured their PMTCT health seeking, deploying secrecy to evade marital conflict and community -based stigma. Such strategies enabled these women to preserve marriage and attend PMTCT to meet their maternal needs to prote ct the health and well -being of their child. Situated in the public sphere, it is clear that the infrastructure of the PMTCT clinic -based stigma. While deploying secrecy ma y be understood as a passive reaction to power inequities, I 157 argue that it is a performative act that allows women to resist the power relations that threaten their ability to retain their marriage and protect their child. Buying blank medical files, seeki ng care at distant clinics, asking strangers to serve as guardians, and providing false addresses to health workers who visit them in their communities to ensure ARV adherence allows patients to creatively resist the biomedical and gendered dimensions of p ower that shape their social relations. 158 Chapter 7: Dissertation Conclusions While conducting clinical observations, I attended a PMTCT clinic in Lilongwe. I noticed a patient reject an intrusive gaze by a few health care workers as she interacted with th e clinician. A well - boundaries between public and private sphere became blurred as the private world of the patient was met with the eyes of three health care workers who sto pped their tasks to listen to the the rhythm of a routine medical history -taking. After some silence, the woman offered an Ndataya pansi mankwa la, interaction between clinician and patient expanded to include a much wider audience as the eyes and ears of the greater community seemed to wait for the response of a woman who effectively resisted tra reconstruction of self as subject, the clinician accepted the subtle challenge to her authority. She outine medical history. The curious health care providers and community members who had turned to listen to the exchange returned to their tasks and conversation. Stories of such acts emerged while conducting fieldwork in PMTCT clinics and various communit y settings in Malawi. Patients, community members, and health care providers reported a number of strategies women use to manage their social relations as HIV -positive pregnant women. As discussed in Chapter 6, patients and health care providers report tha t some women hide ARV medication from husbands, family, and the community. At home, some women hid 159 ARVs in the flour bin or in the ground outside the home. Nurses reported that after PMTCT clinics, they would find ARVs in the toilet and under pillows after women had delivered. Other women strategically bought new medical files that would not contain their HIV status so they could control who had access to this private information. Still others traveled to distant clinics to avoid the potential for stigma an d marginalization that challenged their social identities and constrained their access to care. I found that women employed secrecy as a strategy to manipulate the gender and power structures that challenge their social identities and health seeking practi ces within PMTCT meet their social and clinical needs. Secrecy is a coping strategy that allows women to leverage power and agency to negotiate the competing iden tities of mother, wife, and patient by minimizing the stigma and discrimination that challenges these identities. These coping strategies further allow women to meet their clinical needs to access treatment within PMTCT programs so that they can protect th eir HIV -exposed child, which is their ultimate priority. This dissertation examines a research question that has concerned public health practitioners, clinicians, and social scientists in recent years why do HIV -positive pregnant women drop out of PMTCT p rograms (Bwirire et al. 2008; Chinkonde et al. 2009; Van Hollen 2013; Van der Straten et al. 2014; Painter et al. 2004; Manzi et al. 2005; Weigel et al. 2012)? To better understand why HIV -positive women drop out of PMTCT programs, I examined the following specific research questions: 1) What strategies for AIDS treatment and management do pregnant and postpartum women employ? 2) How do broader social and gender relations affect in the program -160 structured interviews, focus group discussions, and clinical observations with PMTCT patients, the health care providers who care for them, and var ious community members in the Blantyre district of Southern Malawi. I began the dissertation with an account of the experiences of a young HIV -positive patient I met during the final stages of my study when I was interviewing women who had dropped out of P MTCT treatment. In the short time that I came to know Chitsulo, I learned that HIV -positive poor peri -urban women enjoy both supports and constraints in their daily lives as they negotiate social relations in marriage, the extended family, and the wider co mmunity as HIV - multiple social, cultural, and economic forces that shape the HIV/AIDS management of poor peri - erience, in particular, sheds light on the social forces produced in the interaction between gender and power inequities and local understandings of HIV/AIDS. My study of the reasons why some women drop out of PMTCT programs reveals the multiple gender and My objective has been to show relations in marriage and the community: challenging their identities as wives and mot hers, and constraining their health seeking practice as patients in PMTCT programs. Community -Based Stigma This study shows that the HIV -positive women in this study experience multiple forms of stigma, discrimination and marginalization from spouses, fam ily, and the larger community. Participants suggest local understandings of HIV/AIDS construct HIV -positive women as 161 immoral or promiscuous, situating them on the margins of being good mothers and wives. In t his study community, an HIV/AIDS diagnosis means that someone in the relationship the husband or wife has been promiscuous. However, wives are more likely to be blamed because of a gendering of stigma that is produced by gender roles and relations and their intersection with social norms related to wome Despite continued efforts to provide confidential HIV services within the clinic, the lihood of meeting family and community members who do not know their positive status is enhanced simply by seeking care at the public PMTCT clinic. As a result, health -seeking at the clinic exposes an HIV -positive identity, bringing attention to the proble identities as mothers and wives. Women are more likely to be blamed for being HIV -positive because of different standards of morality for men and women. Furthermore, as a sexually trans mitted disease (STD), HIV/AIDS evokes ideas about morality that have been reproduced discursively in local understandings of HIV/AIDS. In colonial biomedical treatment of STDs and contemporary exp eriences with stigmatization. I discuss how manyazi was often experienced by men and women seeking STD treatment because inculcating shame in those seeking treatment for STDs was viewed by colonial officials as a means to instill morality in Africans (Vaug han 1991). Local Understandings of Reproduction and the Construction of Motherhood Just as women are stigmatized for being HIV -positive, they are stigmatized for their 162 infertility if they are unable to reproduce. I included a discussion of infertility in this dissertation to highlight the parallels between two health conditions that illuminate the gender and power structures that shape failing to conform to ideals of motherhood and wifehood , fostering a tension between these competing social identities. Infertile women and women living with HIV/AIDS share a common experience of gendered stigmatization rooted in the social obligation to reproduce. In signifying e gendering of infertility and gendering of stigma demonstrate In addition, both the gendering of infertility and the gendering of stigma reveal how women are constructed as the cause of the ir health condition. Yet, in each of these conditions, women have little control over having additional children, or negotiating safe sex. I have attempted to show that many women in these clinics experience a double bind in which they are blamed when they are unable to reproduce, and also blamed for deciding to reproduce when HIV -positive. These constructions of HIV -positive mothers contrast with the attitudes and practices of women I met and interviewed who prioritized the health of their children and con extend their lives to enact the role of mother. I demonstrate, however, that this enactment of moth erhood places women in a double bind where this critical social identity is co ntested by her identity as a wife. Some health workers experiencing burn -out express negative attitudes towards PMTCT are pregnant while knowing that they are HIV -positive. Others reproduce class -based power differentials in negative interactions with PMTCT patients. Other health care providers 163 stigmatize women for having additional pregnancies while HIV -positive. As discussed in Chapter 6, these health care pro viders redefine the social categories of motherhood, redefining who is eligible to be a mother and who is not. These attitudes demonstrate how biomedical knowledge of the perceived efficacy of ARVs transforms what it means to be HIV -positive and pregnant a t a time when use of ARVs is expected to reduce mother -to-child transmission to less than 5% ( Fasawe et al. 2013). Gender Inequities in Marriage At the same time that women are stigmatized for being HIV -positive by community members and health workers at the clinic, they experience challenges to their marriage when barrier to P MTCT participation reveals larger structural issues of gender ideals, relations a nd inequalities, and the double entities as mothers and HIV/AIDS patients. AIDS management because HIV disclosure may lead to conflict between husbands and wives over sexuality and fidelity in marriage. Applying the theory of gender and power, I argued that gender inequities most often emerge in PMTCT patient s relationship power , decision -making authority in relationships, limited economic ability to l eave marriage, and historically - PMTCT patients in this study are poor peri -urban women with limited economic power. The 164 positionality of these women demonstrates how structural vulnerabilities are produced when particular positionalities intersect gender, poverty, and a positive HIV status. Recently, one of the largest HIV clinical trials conducted to test the efficacy of the prophylactic use of ARVs (PREP) was challenged by the women wh o participated (Van der Straten et al. 2014). After several years of expected success in discovering an effective HIV prevention drug, w -adherence led research investigators to end the intervention. Transnational researchers were baffl ed to find that 70 % of the over 5,000 South African, Ugandan, and Zimbabwean women who participated had not taken the medication, yet told health c are providers they had adhered enacting adherence while secret ly disposing of the medication, as Donald McNeil noted in a February, 2015 New York Times article . The narrative surrounding the ethics of this intervention were called into question , as many wo men were paid stipends that were disproportionate to local costs of living in Kenya. Most relevant to the present study, however, is that these women were able to collectively resist a health intervention with which they did not agree. Such acts reflect th e performative nature of power. Again, subtle forms 2013, 82), however, the above case demonstrates the collective power of multiple acts of resistance, and the agency of wome n who choose to not adhere to meet their own needs. Double Bind: Negotiating the Competing Identities of Wife, Mother, and Patient To my knowledge, social scientists have not studied the interaction of HIV -positive d patients in PMTCT programs, and how they may create a double bind for HIV -positive women when seeking HIV/AIDS treatment. Furthermore, 165 social scientists have not explored how these identities are specifically structured by gender and power inequities. Th ial identities to their HIV/AIDS management. I argue that gender and power inequities structure the social identities of HIV -positive pregnant women situating women in a double bind as they negotiate the competing identities of mother, wife and compliant patient. The PMTCT patients in this study clearly prioritized their social role as mothers and shaped their health -seeking decision -making accordingly. When asked why they participate in PMTCT, women expr essed an overwhelming acceptance of a positive status. Women pointed to the marked improvement in morbidity and to care for the exposed child sheds light on her most significant social role to fulfill caregiving responsibilities is central to her maternal identity. PMTCT participation has a social cost in which women may risk losing their marriage an institution whi children, particularly for women living in poverty, who represent the demographic of PMTCT patients in my study population. Employing Secrecy: Mediating Competing Identities Finally, fe w studies have explored the strategies women employ to negotiate health seeking within PMTCT programs. I have shown how in some instances, women respond to these social forces with subtle forms of resistance, employing secrecy to avoid marital conflict, minimize stigma, and address their heal th needs and those of their HIV -exposed child. I found 166 that women employed secrecy as a strategy to manipulate the gender and power structures that challenge their social identities and health seeking practices within P MTCT programs. gender and power relations in the social relations of PMTCT patients, providing an example of disguised and express th emselves through discrete and veiled acts of rebellion (Crehan 1997). Women employ secrecy to preserve their social identities and access PMTCT treatment for their child. Women also continue to have children while being HIV -positive. Some women may drop ou t, or not adhere to PMTCT, not because of the many barriers discussed in this dissertation, but by their own choice. It is possible that some women choose to avoid treatment in PMTCT programs to resist the discursive power embedded within health interventi ons and alter them to meet their own needs. These women may choose to drop out of PMTCT programs because they perceive these programs as another power structure to negotiate as it seeks to control their reproductive health and sexuality. Two such women are included in this study. As take the ARVs provided by the PMTCT program. o meet their social and clinical needs. Secrecy is a coping strategy that allows women to leverage agency to mediate the competing identities of mother, wife, and patient, minimize the stigma and discrimination that challenges these identities, and meet th eir clinical needs to access treatment within PMTCT programs so they can protect their HIV -exposed child, which is their ultimate priority. 167 This study reveals the tension women experience in negotiating the conflicting identities of wife, mother, and patie nt as HIV -positive pregnant women and new mothers. The study S management, and how gender and power inequities structure social identities of HIV -positive pregnant women. I have attempted to show that social and cultural obstacles rooted in gender and economic inequities may pose a far greater challenge than the clinical regimen demanded of patients within the may shed light on how best to implement these p ublic health interventions. Study Limitations There are several limitations to this study. The study relies heavily upon the attitudes and experiences of health care providers who treat HIV -positive women wi thin PMTCT programs. This clinic -based focus occurred because given the sensitivity with talking to HIV -positive women, my primary concern was to not disclose the status of women who participated in the t forced disclosure, I was unable to follow- responses to interviews conducted only once and within the clinic setting. Interviews with community men and women were conducted to provide greater depth to the interviews conducted with PMTCT patients in the hospital. Second, this study focuses on the experiences of HIV -positive women. Given the important role of gender relations and inequalities uncovered in the analysis, the perspec tive of menspecifically the husbands and partners of PMTCT patients would provide a more balanced perspective on gender roles and relations. 168 Finally, interviews with women who had dropped out of the program were much less informative as compared to interv iews and focus groups with women who remained in the PMTCT program. This is likely because these women were most at risk of negative consequences i ef Gender and HIV/AID S Research As discussed in Chapter 2, the majority of the anthropological and public health literature on gende r and HIV/AIDS focuses on prevention specifically how gender inequalities may relations once they have begun treatment to understand why gender inequities may interfere with et al. 2003; Bw irire et al. 2008, Chinkonde et al. 2009; Kuonza et al. 2010; Van der Straten et al. 2014). For example, research conduc ted in Southern Africa (Nyblade et al. 2003) found that women were more likely to be stigmatized for being HIV -positive because they are believed to fail to fulfill their roles as caregivers to members of the extended family bec ause of their illness from HIV. Still, few studies exa mine the role that gender plays (Van Hollen 2013; Chinkonde et al. grams. Van Hollen (201 3) is one of the first treatment. In her study of HIV -positive mothers in India, Van Hollen draws attention to the fact that an HIV/AIDS diagn osis disrupts family and kinship relations , which may manifest in stigmatizing attitudes and practices toward women. Within the patrilineal context of her study, Van Hollen (2013) found that women were stigmatized by mothers -in-law who blamed them for 169 bringing HIV/AIDS into the marriage, thus threatening the moral integrity and continuity o f the lineage . motherhood. As discussed in Chapter 4, social scientists have examined the subjectivity of mothers with HIV/AIDS (Long 2009; Yeatman and Trinitapoli 2013) and the fertility desires of mothers with HIV/AIDS (Yeatman and Trinitapoli 2013 ). Social scientists have explored , from a phenomenological perspective, the subjective experien ce of HIV -positive mothers at a time when ARVs were not available to women who could not afford them. Long (2009) argues that -positive motherhood. Similarly, Van Hol len (2013 responsibilities for women who were unable to access ARV treatment. Van Hollen found that to take on the mo thering role after her expected death. In addition, while literature has drawn attention to the role of stigma in men and HIV/AIDS wi th gender relations (Van Hollen 2013 ), and how it may affect the relationship power of HIV -positive pregnant women . This dissertation found that women are more likely to be blamed for their HIV/AIDS status rather than their husbands. Because the women in this study are more likely to be blamed for bringing HIV/AID S into marriage and may be labeled ge, they are placed in a double bind with few options for accessing treatment other than secrecy or non -adherence. 170 Suggestions for Future Research This stu dy raises several questions that may be explored to address the challenges faced by PMTCT patients. First, how can PMTCT programs engage the support of spouses without reproducing the gender and power inequities HIV -positive women may experience in marriag e? In recent years, spousal involvement has been identified as a potential strategy for encouraging these efforts. For many women in this study, spousal involvement may discourage their participation in PMTCT programs due to fears of marital problems and economic instability. Addressing Gender I nequities: Increasing the R elationship Power of HIV-Positive Women A robust strategy would involve increasing the relationsh ip power of HIV -positive women -making authority and challenging the power and decision -making authority in the relationship, gender mainstreaming may provide one into development interventions to involve men and women equally in economic strategies to minimize power differentials between men and women in the household ( USAID 2012). Such strategies may provide women with the economic security to leave their marriage if forced to do so. Challenging social norms that focus on the sexual practices of women , and deemphasizing the tendency to blame the husband or the wife may promote a more balanced perspective on HIV/AID S as a disease that is the joint responsibility of husbands and wives. 171 Providing Social Support from Other HIV -Positive W omen Second, how c an social support be provided to HIV -positive pregnant women and new mothers outside the clinic setting to help women negotiate the social, cultural, and economic barriers they encounter? In Southern Africa, support groups for HIV -positive mothers have bee n implemented to provide social support to HIV -positive women from other women living with HIV ( Foster et al. 2014 ). By involving women with shared identities, these programs may provide social support and allow women to problem -solve with other women expe riencing similar social, cultural, and clinical challenges. Such programs could be adopted to help women negotiate the social effects of PMTCT participation, such as marital distress and community -based stigma. Minimizing HIV/AIDS Stigma through S ervice D elivery Third, how can stigma and discrimination toward women living with HIV be reduced? Models of integrated care may provide confidentiality for women who avoid PMTCT clinics for fear of public disclosure by providing all clinical services at a single m edical visit. For HIV -positive new mothers, such confidentiality would recognize the privacy valued by women living with HIV. This may promote treatment adherence and minimize the effects of gender inequalities and community -based stigma that preven t low -income women in resource -poor settings from fully accessing these programs. One such program has been tested in Kenya. Known as the -related services, including HIV /AIDS care , during one clinical visit. This structure would allow both men and women to seek treatment for HIV/AIDS privately while addressing other health needs ( Andrews 2011 ). 172 Finally, a better understanding of the perspectives of patients involved in PM TCT programs may help to identify strategies to create PMTCT programs more responsive to PMTCT patients themselves may provide answers to some of these questions. Furt her, given the centering on the attitudes and experiences of husbands of PMTCT patients would provide a more balanced perspective on how gender and power structures sh ape the social relations and health seeking of HIV -positive pregnant women. 173 APPENDICES 174 Appendix A: Survival Analysis of Plan B+ PMTCT Patient Adherence 175 Appendix A: Survival Analysis of Plan B+ PMTCT Patient Adherence Quantitative Methods Below I descr ibe a survival analysis , conducted at Queen Elizabeth Central Hospital, to unable to adhere to ARV treatment during the study period. Data were retroactively collected from PMTCT registers at QECH to determine number of PMTCT appointments attended and appointments missed within the first 12 months of Plan B+, between October 3, 2011 and October 3, 2012. Data were entered into Excel and transferred to SPSS for t he analysis. The quantitative analysis investigates two research questions: 1) What proportion of women within the study population (for the quantitative study) default within the study period; and 2) Does the likelihood of defaulting depend on age as a co variate? Quantitative Analysis Data was retrospectively collected from PMTCT registers at QECH to determine number of PMTCT appointments attended and appointments missed within the first 12 months of Plan B+, between October 3, 2011 and October 3, 2012. D ata was entered into Excel and transferred to SPSS for the analysis. Patients included in the analysis represent the first patient cohort receiving the 5A treatment regimen under the current national PMTCT policy, Plan B+. This new treatment regimen was im plemented by the Malawian Ministry of Health (MOH) on October 3, 2011. Patient medical files were reviewed retrospectively to determine eligibility. All PMTCT patients registered between October 3, 2011 and October 3, 2012 were included in the analysis. PMTCT programs last for a period of two years, or until the end of the breastfeeding period, as recommended under Plan B+ (Ministry of Health 2011). This analysis, which was conducted between October 3, 2011 and October 3, 2012, covers at most the first year of the program. Patients registered in the national PMTCT program are scheduled to return monthly for medication refills according to Plan B+ guidelines. Those patients who fail to return for refills two or more months after pill completion are defined as Survival Analysis Methods Survival analysis assesses the proportion of a population that survive until an event of interest occurs. Literature on the use of survival analysis methods suggests this statistical method is mos t commonly used to assess patient survival after receiving medical treatment, or to measure time to disease relapse following the application of a new treatment regimen (Singh and Mukhopadhyay 2011). Survival data analysis has been increasingly used for ot her studies, such as retention within drug and alcohol treatment programs (Kelly et al 2011). Following this trend, I employed survival analysis to model the amount of time patients remain on ARV treatment within PMTCT programs until the event of interest, treatment in the presence of censored data. Censored data occur when the event of interest, in ant during our specified time period. This is the case for many of the participants in this study. Results for the 183 patients registered within the PMTCT program at QECH demonstrate ng the study period, and 176 118 patients did not. Data were right -censored, indicating that the event of interest, defaulting, may still occur beyond the study period of one year or after transferring from the study. In other words, time to default for a cens ored patient is > (number of days observed). Censored data in this study include transfers, or patients who transferred to other PMTCT clinics in the region to finish treatment as well as those who successfully completed treatment through the end of the study (October 3, 2012). For these patients, we know that their time to default is greater than the number of days they were followed in the study. Ninety -nine patients were still active at the end of the first year the program, and nineteen patients transfe rred during the study period, or chose to obtain treatment at a PMTCT clinic other than the district hospital. Consequently a total of A survival analysis was conducted using SPSS Version 21 (CITE) to calculate the length of time patients remain on treatment prior to defaulting. Survival analysis methodology was statistic as compared to a simple prob ability ratio. A simple probability is more commonly used in the literature (Nachega et al. 2013) to assess adherence within PMTCT programs. These statistics calculate the probability of defaulting by including in the numerator the total number of patients who default, dividing the number by the total number of patients started on treatment. For example, in this study, 65 patients defaulted out of a total 183 patients who were initiated on treatment. This simple frequency states that 35.5% of patients in th e program defaulted, or failed to adhere to treatment. A simple extension of this logic may lead someone to the incorrect conclusion that 64.5%, or the remaining 118 patients, were still active or successfully engaged in the treatment at the end of the stu the same as still active. The calculation of 64.5% includes in the numerator patients who were active at the end of the study and patients who transferred out of the study before it ended. A survival analysis, in contr ast, does not assume that transfers were still active at the end of the study period. Transfers instead are treated as censored observations and removed from the daily survival calculations illustrated in Table 1, a table of Kaplan Meier survival estimates (Clark et al. 2003).The Kaplan Meier Life Table (Table 1) shows the proportion of patients remaining within the program at each daily time interval, the number of patients who defaulted at d. This table represents a daily summary of the full SPSS output . Number exposed to risk number entering the program minus half of those patients who withdrew during the time interval. The number of terminal events defaulted in the given time interval. The number of censored events represents the number of censored patients at each daily interval. These include patients who transferr ed or patients who successfully completed the first year of treatment. The daily proportion defaulting represents the proportion of patients who default during the given time interval. The ratio is calculated by dividing the number of defaulting patients b y the number of patients at risk of defaulting during the given time interval. The daily proportion not defaulting represents the proportion of patients participating within the given time interval, and is calculated by subtracting the proportion of patien ts terminating by the given time interval from one. The cumulative proportion not defaulting represents the cumulative proportion of patients not defaulting up to the given time interval. 177 For example, on Day 0, 29 patients defaulted, 18 patients were censo red, leaving 136 calculations remove all censored data from the calculation, including both transfers and patients still participating at the end of the study. Therefo re, the Kaplan Meier survival estimate provides a more accurate estimate of the cumulative proportion not defaulting that continually adjusts the denominator for all censored observations, both transfers and those still active at the end of the study. As a result, the cumulative proportion of patients who did not default (.527) using survival techniques is lower than one might have inferred from the simple probability estimate (.645). The survival probability is therefore the preferred calculation because i t accounts for the uncertainty of patient behavior, such as whether or not patients who transferred out of the program ultimately defaulted. Because of the nature of censored data, it appears that survival probability is the most statistically robust asses sment of patient adherence (Prinja et al. 2010). Table 1: Kaplan Meier Life Table Daily Summary Day Number entering or exposed to risk Number Defaults Censored Daily proportion defaulting Daily proportion not defaulting Survival estimate, cumulative proportion not defaulting Standard Error 0 183 29 18 0.158 0.842 0.842 0.027 1 136 4 1 0.029 0.971 0.817 0.029 2 131 4 2 0.031 0.969 0.792 0.031 7 125 2 0 0.016 0.984 0.779 0.031 8 123 0 3 0.000 1.000 0.779 0.031 17 120 0 1 0.000 1.000 0.779 0.031 25 119 0 2 0.000 1.000 0.779 0.031 28 117 0 1 0.000 1.000 0.779 0.031 29 116 0 4 0.000 1.000 0.779 0.031 30 112 0 1 0.000 1.000 0.779 0.031 37 111 4 0 0.036 0.964 0.751 0.033 44 107 0 1 0.000 1.000 0.751 0.033 49 106 2 0 0.019 0.981 0.737 0.034 51 104 1 0 0.010 0.990 0.730 0.034 52 103 0 2 0.000 1.000 0.730 0.034 55 101 1 0 0.010 0.990 0.723 0.035 56 100 1 1 0.010 0.990 0.715 0.035 57 98 0 1 0.000 1.000 0.715 0.035 59 97 0 1 0.000 1.000 0.715 0.035 62 96 0 1 0.000 1.000 0.715 0.035 65 95 0 1 0.000 1.000 0.715 0.035 71 94 0 1 0.000 1.000 0.715 0.035 78 93 0 2 0.000 1.000 0.715 0.035 178 79 91 0 1 0.000 1.000 0.715 0.035 86 90 1 0 0.011 0.989 0.707 0.036 89 89 0 1 0.000 1.000 0.707 0.036 90 88 0 2 0.000 1.000 0.707 0.036 93 86 0 1 0.000 1.000 0.707 0.036 94 85 0 2 0.000 1.000 0.707 0.036 95 83 0 1 0.000 1.000 0.707 0.036 98 82 0 1 0.000 1.000 0.707 0.036 99 81 1 1 0.012 0.988 0.699 0.036 100 79 0 1 0.000 1.000 0.699 0.036 106 78 1 0 0.013 0.987 0.690 0.037 107 77 0 2 0.000 1.000 0.690 0.037 111 75 0 3 0.000 1.000 0.690 0.037 113 72 1 0 0.014 0.986 0.680 0.038 114 71 2 0 0.028 0.972 0.661 0.039 115 69 0 1 0.000 1.000 0.661 0.039 117 68 1 0 0.015 0.985 0.651 0.040 119 67 0 3 0.000 1.000 0.651 0.040 120 64 1 1 0.016 0.984 0.641 0.040 124 62 0 1 0.000 1.000 0.641 0.040 127 61 1 2 0.016 0.984 0.631 0.041 129 58 0 1 0.000 1.000 0.631 0.041 131 57 0 1 0.000 1.000 0.631 0.041 134 56 2 1 0.036 0.964 0.608 0.043 142 53 0 1 0.000 1.000 0.608 0.043 146 52 0 1 0.000 1.000 0.608 0.043 148 51 1 0 0.020 0.980 0.596 0.043 150 50 0 1 0.000 1.000 0.596 0.043 153 49 0 1 0.000 1.000 0.596 0.043 154 48 2 3 0.042 0.958 0.571 0.045 157 43 0 1 0.000 1.000 0.571 0.045 161 42 0 1 0.000 1.000 0.571 0.045 162 41 2 0 0.049 0.951 0.543 0.047 163 39 0 1 0.000 1.000 0.543 0.047 168 38 0 1 0.000 1.000 0.543 0.047 170 37 0 1 0.000 1.000 0.543 0.047 175 36 0 1 0.000 1.000 0.543 0.047 178 35 0 1 0.000 1.000 0.543 0.047 197 34 1 0 0.029 0.971 0.527 0.048 204-359 Varies 0 33 0.000 1.000 0.527 0.048 179 Patterns in Defaulting Another benefit to using Kaplan Meier survival estimates is that they reflect the pattern of defaulting and censoring as it occurs over time. The cumulative survival estimates are plotted over time in the cumulative survi val function shown in Figure 1.The first vertical drop in the curve represents the event of interest, defaulting, occurring at the highest rate in the first ten days of the program. The rate is relatively higher again in the second month, as shown by the d ecrease in cumulative survival in the second month of the program, followed by a steady increase in the number of defaults (or a steady decrease in the survival function) from three to six months. The median estimated time to default is 218 days. There wer e no patient defaulters after 197 days. This left 52.7% of patients still active at the end of the study. Figure 1: Survival Function *Time 1=Days Cumulative Hazard Function The hazard function illustrates the relative risk that patients will default w ithin the given time interval (Figure 2). While survivor functions emphasize the probability of not having the event of interest or defaulting occur, the hazard function emphasizes the probability of defaulting at the given time, and is based upon survival probability up to a specific point in time (Clark et 180 al. 2003). The shape of the hazard function also suggests the time of greatest risk of defaulting is in the first 10, and subsequently, 30 days of the program. Figure 2: Hazard Function *Time 1=D ays Discussion Patients may default early in the program for a number of clinical, programmatic, and sociocultural reasons. Patients may not have experienced the clinical benefits of the medication, or have experienced initial side effects from treatment. Dep ending on the period in pregnancy, women may be forced to leave early on in the program. For example, women who begin treatment prior to delivery may choose to deliver at home, or through the use of a traditional birth attendant (TBA), which may lead some women to fear returning to the hospital for treatment. Further, women who begin treatment at delivery may delay return to the program because of the demands of the initial weeks of motherhood. Importantly, because of the nature of censored data, it is unkn own whether patients who appear to default have transferred to another facility without receiving an official transfer or have died. As stated in the qualitative component of the study, regardless of the point in the treatment trajectory, many patients may be unable to successfully adhere to treatment within PMTCT for a variety of sociocultural reasons, including lack of spousal and community -based support and the nature of HIV/AIDS stigma, as discussed in more detail in Chapters 5 and 6. 181 Covariate Analysi s Further analysis of the data suggested that age may be a predictive factor for likelihood of defaulting. We used the cox regression method of analysis to assess the effects of age on patient Age wa s categorized into the following subgroups: 1) Patients over the age of 18 and under the age of 30 (n=112) and 2) Patients over the age of 31 (n=71). The results of the cox regression are summarized in Table 2. We found a significant difference in default rates between the 2 groups at a 5% level of significance (p=0.016). The median survival time for older women in group was 330 days whereas the median survival for younger women was 158 days. Figure 3: Cox Regression Analysis *Time 1=D ays Table 2: Cox R egression Analysis Results B SE Wald statistic df Sig. Exp(B) Age category .661 .275 5.790 1 .016 1.936 182 The above cox regression analysis shows that older women (ages over 31 -45) (upper, green curve) have almost twice the likelihood of staying in the p rogram as compared to women under the age of 30 ( lower, blue curve). Similarly, women in the younger group have almost twice (1.936) the relative risk of not remaining within the program. 183 Appendix B: PMTCT Core Services and Patient Profiles 184 Appendix B : PMTCT Core Services and Patient Profiles Table 3: PMTCT Core Services PMTCT Core Services Step 1 Utilization of Antenatal Care Step 2 Receive HIV pretest counseling Step 3 Acceptance of HIV test Step 4 Receive HIV test results and post -test counsel ing Step 5 Use of ARV prophylaxis for mother and/or infant (for seropositive mothers) Step 6 Use of labor and delivery services which include PMTCT interventions Step 7 Postnatal follow -up with mother and infant, with HIV test for exposed infant and timely access to treatment, care and support Table information sourced from F igure 1 in Marcos et al 2012. 185 Table 4 : Patient Profile District Hospital (QECH) Health Centre (Ndirande) Health Centre (Mpemba) Total 18-25 12 0 1 5 3 26-40 33 4 7 6 50 NA 3 3 Total 45 4 8 14 71 Education None 0 0 2 0 2 Primary 3 3 6 9 21 Some Secondary 17 1 0 3 21 MSCE 20 0 0 0 20 College 1 0 0 0 1 NA 4 0 0 2 6 Total 45 4 8 14 71 Employment Farming 0 0 5 3 8 Business 13 0 1 6 20 Housewife 18 4 2 2 26 Professional 9 0 0 0 9 NA 5 0 0 3 8 Total 45 4 8 14 71 Husband HIV Positive 18 0 1 1 20 Negative 3 1 0 2 6 Not tested 9 3 7 8 27 NA 15 0 0 3 18 Died 0 0 0 0 0 Total 45 4 8 14 71 Marital Status Married 39 4 8 8 59 Single 2 0 0 1 3 Divorced 1 0 0 3 4 Widowed 0 0 0 0 0 NA 3 0 0 2 5 Total 45 4 8 14 71 186 Appendix C: Interviews and Instruments Used 187 Appendix C: Interviews and Instruments Used Table 5: Instruments Used Semi -structured Interviews Focus Group Discussions Current PMTCT Patients QE District Hospital 14 3 Ndirande Health Centre 4 Mpemba Health Centre 0 1 14 Total 32 5 Health Care Providers QE District Hospital 11 Ndirande Health Centre 3 Mpemba Health Centre 4 Total 18 Community Leaders Teachers 2 Church leaders 2 1 Village Chiefs 2 Birth Attendants 2 Total 6 3 188 Patient Interview Questions (Chichewa) Mphatikiri B: Chisanzo chamafunso ofunsidwa pokambirana ndi odwala: Ena mwamafunso okhudzana ndi odwala ok amba za zopinga zamatenda ndi njira zopezela chithandizo 1. ntchito? Kodi muli ndi ana angati? 2. Kodi mungalongosole bwanji za thanzi lanu tsopano? 3. Kodi mungandiuze zimene munachita pamene munamva koyamba kuti muli ndi matenda a HIV/AIDS? Kodi adokotala anakuuzani chiyani? Kodi inu munachita chiyani pamene munazindikira kuti muli ndi matendawo? 4. Kodi mubanja lanu, ndi ndani amene amakhala ndi mphamvu ya ulamuliro pokhuzana ndi chisamaliro cha moyo wanu? Umoyo wa mwana wanu? ali mu chilinganizo cha PMTCT 1. Kodi mwakhala mukutenga nawo mbali muchilinganizo cha PMTCT kwa nthawi yayitali bwanji ku QECH/Kawale? 2. 3. Kodi zinthu zikukuyendarani bwanji? Kodi zinthu zimene mukuona kuti zimagwira ntchito ndi chani, kapena zinthu zimene sizimagwira ntchito ndi chiyani? 4. Kodi munayamba mwatenga nawo mbali muzilinganizo zina za PMTCT? Kodi ndi chifukwa chiyani munaganiza kusiya? 5. Kodi pali munthu wina amene amakutha ndizani kapena kukupingani potenga nawo mbali? 6. Kodi mungachipititse bwanji patsogolo chilinganizo chimenechi? 7. Kodi munaberekera mwana wanu kuti? Kodi ndi chifukwa chiyani munasankha kuti kuti muberekere mwana ku QECH/kwina? Kodi pamene munaberekera mwana k unja kwa chipatala, zinakupangitsani mavuto otani kuti mulowe kapena musalowe nawo muchilinganizo cha PMTCT? 1. Kodi anthu amadziwa kuti mumapita ku PMTCT? 2. Kodi ndi ndani amene amadziwa kuti mumapita ku PMTCT? 3. Kodi iwo amakulimbikitsani kapena kukugwetsani mphwayi? Ndi chifukwa chiyani mukuganiza choncho? Kodi mungakonze bwanji vuto limeneli? 4. Kodi mumabwera ndi ndani ku chipatala? 5. Kodi ndi ndani amene munamuuza po zakudwala kapena kusadwala kwanu kwa matenda a HIV/AIDS? Anzanu? Abale anu? kapena mu umoyo wapamudzi: 1. Kodi ndi chiyani chimene chingapangitse chipatala cha PMTCT kukhala c hosavuta komanso 189 2. Ngati mutakhala ndi mwayi wosintha chinthu china pakalandilidwe kamankhwala mchilinganizo cha PMTCT , chingakhale chani? 3. Kodi ndi zifukwa zina ziti zimene zingamupangitse munthu kuti asabwererenso ku chilinganizo cha PMTCT? 4. Kodi munayamba mwalephelapo kupita kuchipatala panthawi imene munayenera kupita kumeneko? Kodi munalephera chifukwa chani? 5. Kodi mukudziwapo mayi amene anasiyilatu kubwera ku chipatala cha PMTCT? Popanda kundidziwitsa dzina lake, kodi mukudziwa zifukwa zake zimene zinamupangitsa kulephera kubwerera kuchipatala cha PMTCT? Kodi mukuganiza kuti anachoka chifukwa chiyani? Chinachitika ndi c hiyani kuti apange chitsimikizo kuti achokedi? 6. Kodi alipo anthu ena amene amaoneka kuti sangamalize nawo chilinganizo cha PMTCT? Kodi mukuganiza kuti izi zili choncho chifukwa chani? Kodi munamvapo chani zokhudzana ndi nkhani imeneyi? 7. Kodi ndi chifukwa ch ani amayi amatenga nawo mbali mchilinganizo cha PMTCT? Kodi ndi chifukwa chani iwo amasankha kuti amalize nawo chilinganizo chimenechi? 8. Kodi munamvapo zakusankhidwa kapena kusalidwa kumene kunapangidwa kwa munthu amene ali ndi matenda a HIV/AIDS kapena kwa banja lake? 9. Kodi mukuganiza kuti chingachitike ndi chani popeza njira yothetsa vuto losankha odwala ngati likupitilirabe masiku ano? 10. Kodi ndi ziti zimene munamvapo kuchokera kumaphunziro a za umoyo zimene zakhala zofunika kwambiri pamoyo wanu? Izi zili ch oncho chifukwa chani? Kodi zilipo zina zimene munamvapo zimene ndizosakuthandizani? 11. Kodi patakhala chinthu china chosiyana chimene chingakuchepetseni mavuto kuti muthe kudzisamale nokha ndi chiyani? Kapena kuti muthe kusamala mwana wanu amene ali ndi kachi lombo ka HIV? Mphatikiri C: Chinsanzo chamafunso ochokera kwa gulu la odwala: 1. Kodi ndi zifukwa ziti amayi amatenga nawo mbali mchilinganizo cha PMTCT? 2. Kodi ndi zifukwa ziti amayi amasiya kutenga nawo mbali mchilinganizo cha PMTCT? 3. Kodi ndi zifukwa ziti zi mene zingapangitse amayi kuti amalize bwino potenga nawo mbali mchilinganizo cha PMTCT? Mphatikiri D: Chinsanzo mafunso ochokera kwa anthu apamudzi: 1. Kodi ndani mayi wachisanzo chabwino? Kodi ndi zinthu ziti zimene zimaonetsa kuti mayi uyu ndi wabwino? Kod i ndi zinthu ziti zimene zimaonetsa kuti mayi uyu ndi woyipa? 2. 3. Thandizo lazofunika mu umoyo w amunthu? 4. kupangira zinthu limodzi? 5. Kodi malo abwino kuti mayi angaberekere mwana ndi kuti? 6. Kodi ndi zinthu ziti zimene zipatsa mayi nkhawa kwa mayi ndi chiyani? 7. Kodi muli ndi chiyembekezo chotani pa umoyo wanu komanso umoyo wa mwana wanu? 190 Patient Interview Questions (English) 1) Please describe your typical day to me. Do you work? How many children do you have? 2) How would you describe your current state of health? 3) Can you tell me when you first found out that you had HIV/AIDS? What were you told? What did you do when you found out? 4) Who in your fam ily usually has the final say on decisions related to your health care? Your Patient perceptions of PMTCT program: 1. How long have you been attending the PMTCT p rogram at QECH/Kawale? 2. What are your responsibilities as a PMTCT participant? 3. 4. Have you participated in other PMTCT programs? Why did you decide to leave? 5. Does anyone help or hinder your part icipation? 6. How would you improve the program? 7. Where did you deliver your child? Probe: Why did you choose to deliver at QECH/other? Has your delivery outside of the hospital affect your ability to participate in PMTCT? Patient perceptions of supports with in clinical, household, or community setting: 1. Do people know that you attend PMTCT? 2. Who knows that you attend PMTCT? 3. Do they support or discourage you? Why do you think so? How could you fix this? 4. Whom do you usually bring with you to the clinic? 5. With whom have you shared your HIV/AIDS status with? With whom might a woman typically disclose her HIV/AIDS status? Friends? Family members? Patient perceptions of challenges, within clinical, household, or community setting: 1. What, if anything, would make the PMT CT clinic easier/better? Probe: What aspects of PMTCT do you like? What aspects of PMTCT do you dislike? 2. If you could change any aspect of your treatment responsibilities within PMTCT, what would that be? 3. What are some of the reasons someone would not retu rn to PMTCT? 4. Have you ever missed an appointment? Why did you miss this appointment? 5. Can you think of a woman who has not returned to the PMTCT clinic? Without telling me who this woman is, do you know the reason why she was unable to return? Why do you th ink that is the case? What happened to cause her decision to leave? 6. Are certain people less likely to complete PMTCT? Why do you think that might be so? What have you heard about that? 7. Why do women participate in PMTCT? Why would they choose to complete t he program? 8. Do you personally know of any examples of women who have begun a PMTCT program and were unable to continue? Why do you think that is the case? What happened to cause her 191 decision to leave? Do you know anyone else who has been unable to complete PMTCT? 9. Have you ever heard of a case of stigma or discrimination directed toward a person living with HIV/AIDS or their family members? 10. What do you think can be done to address the stigma that persists today? 11. Whic h information from the health education session has been most valuable to you? Why is that so? Has any information been not helpful? 12. What, if anything, could be different that would make it easier for you to care for yourself? To care for your (HIV -positiv e) child? Sample Patient Focus Group Questions : 1. What are the reasons women participate in PMTCT? 2. What are the reasons women drop out of PMTCT? 3. What are some factors that would make women better able to finish PMTCT? 4. What is an ideal mother? (Probe: What a re the characteristics of a good mother? What are the characteristics of a bad mother?) 5. What is an ideal period of time for a mother to breastfeed her child? 6. Who in the family is most likely to provide women social support? Material support? 7. Who in the com munity is most likely to provide women social support? 8. Where is the best place for a woman to give birth? Why? 9. What are the main things that worry mothers? 10. What are your hopes for your future and that of your child? 192 Health Care Provider Interviews Gener al Research Questions 1. What strategies for AIDS treatment and management do women employ? 2. program completion? 3. gies and program completion? 4. How do community concepts of motherhood and ideas about breastfeeding practices affect 5. completion ? Sample Provider Interview Questions 1. Tell me about your professional background and training. How long have you been working at this facility? What are your main responsibilities? Could you describe a typical day at the PMTCT clinic for me? 2. What are the general characteristics of your patient population? How many doctors/nurses/ midwives currently work within the PMTCT program? 3. What are the main components of your PMTCT program? What is the typical routine for a woman and infant who participate in the pro gram? 4. How do you enroll women to participate in PMTCT? 5. What are the greatest barriers to providing PMTCT to women and their infants? 6. Why do you believe some women do not complete PMTCT? Why do you think that is so? 7. Why do you believe some women complete PM TCT? 8. At what point in the PMTCT process are women most likely to not return? 9. Please describe efforts made by PMTCT staff to track patients who are lost to follow -up. 10. How would you improve the program? Sample Focus Group Questions 1. What are some reasons a w oman would attend PMTCT? 2. What are some reasons a woman would leave PMTCT? 3. Where do most women typically give birth? 4. Can a woman deliver at home and still attend PMTCT? 5. How long should a woman breastfeed? 6. With whom does a woman share her HIV/AIDS status wit h? With whom might a woman typically disclose her HIV/AIDS status? Spouse? Family members? Friends within the community? 7. Who typically makes the decisions within a household? 193 REFERENCES 194 REFERENCES ry of the Global Movement 1947 British Medical Bulletin 67(1): 13 -25. Agrawal, Arun. 2005. Environmentality: Technologies of Government and the Making of Subjects . Durham, NC: Duke University Press. : a One - Frontlines USAID Online Edition: Global Health/Iraq. April/May. Accessed March 1, 2015 at http://www.usaid.gov/news -information/frontlines/global -healthiraq/health -services -wheels -one-stop-shop Ashforth, Adam. 2005. Witchcraft, Violenc e, and Democracy in South Africa . Chicago, IL: University of Chicago Press. Auerbach, Carl and Louise Silverstein. 2003. Qualitative Data: Introduction to Coding and Analysis . New York, NY: New York University Press. Balcha, T. T., S. S. Lecerof and A. R Journal of the International Association of Providers of AIDS Care 10(3): 187 -192. Bateson, Gregory. 1972. Steps to an Ecology of Mind: Collected Essays in Anthropology, Psychiatry, Evolution, and Epistemology . Chicago, IL: University of Chicago Press. In AIDS, Sexuality, and Gender in Africa: Collective Strategies and Struggles in Tanzania and Zambia, edited by Carolyn Baylies and Janet Bujra, 1 -24. London and New York: Routledge. Bernard, H. Russell. 2000. Social Research Methods: Qualitative and Quantitative Approaches . Thousand Oaks, CA: Sage Publications. Biehl, Jo‰o . 2005. Vita: Life in a Zone of Social Abandonment . Berkeley, CA: University of California Press. 195 Birbeck G. L., E. Chomba, M. Kvalsund, R. Bradbury, C. Mang'ombe, K. Malama, T. Kaile, P. Antiretroviral Adherence in Rural Zambia: The First Year of Treatm American Journal of Tropical Medicine and Hygiene 80(4): 669 -674. The Effectiveness of the TBA Programme in Reducing Maternal Mortality and East African Journal of Public Health 5(2): 103 -10. Boo th, Karen. 2004. Local Women/Global Science: Fighting AIDS in Kenya . Bloomington, IN: Indiana University Press. Bornstein, Erica. 2003. The Spirit of Development: Protestant NGOs, Morality and Economics in Zimbabwe . New York, NY: Routledge. Bourdieu, Pie rre and Jean - Reproduction in Education, Society and Culture , 1-68. London: Sage Publications. t II: Multivariate Data Analysis British Journal of Cancer 89: 431-436. The Journal of Modern African Studie s 44: 173-202. World Development 34(9): 1654 -1666. Burke, Timothy. 1996. Lifebuoy Men, Lux Women: Commodification, Consumption, a nd Cleanliness in Modern Zimbabwe . Durham, NC: Duke University Press. Bwirire, L. D., M. Fitzgerald, R. Zachariah, V. Chikafa, M. Massaquoi, M. Moens, K. Kamoto Reasons for Loss to Follow -up among Mothers Registered in a Preventi on of Mother -to- Transactions of the Royal Society of Tropical Medicine and Hygiene 102(12): 1195-1200. 196 -Related Stigma: From Anthropologic American Journal of Public Health 95(1): 53 -59. Changole, J., C. Bandawe, B. Makanani, K. Nkanaunena, F. Taulo, E. Malunga and G. Malawi Medical Journal 22(1): 5 -9. Chapman, Rachel. 2012. Family Secrets: Risking Reproduction in Central Mozambique. Nashville, TN: Vanderbilt University Press. Chilopora, G ., C. Pereira, F. Kamwendo, A. Chimbiri, E. Malunga and S. Bergstrım. 2007. Postoperative Outcome of Caesarean Sections and Other Major Emergency Obstetric Human Resources for Health 5:17. Chi -to-Child HIV Reproductive Health Matters 17(33): 143 -151. Clark, T. G., M. J. Bradburn, S. B. Love and D British Journal of Cancer 89: 232-238. Cole, Jennifer and Lynn Thomas. 2009. Love in Africa . Chicago, IL: University of Chicago Press. Comaroff, Jean and John Comaroff. 1993 . Modernity and its Malcontents: Ritual and Power in Postcolonial Africa. Chicago, IL: University of Chicago Press. Gender, Health and Theory: Conceptualizing the Issue, in Local and World Perspective. Social Science & Medicine 74(11): 1675 -83. Connell, Robert. 1987. Gender and Power. Society, the Person, and Sexual Politics . Stanford, CA: Stanford University Press. 197 Journal of Interpersonal Violence 29(5): 866 -888. Relations in Ado - American Ethnologist 29(4): 963 -980. Crehan, Kate. 1997. The Fractured Community: Landscapes of Power and Gender in Rural Zambia . Berkeley, CA: University of California Press. Das, Veena. 2007. Life and Words: Violence and the Descent into the Ordinary . Berkeley, CA: University of California Press. Davison, Jean. 1998. Agriculture, W omen and Land: The African Experience . Boulder, CO: Westview Press. . 1997. Gender, Lineage, and Ethnicity in Southern Africa . Boulder, CO: Westview Press. Desai, Sonalde and Kiersten Johnson. Familial a n A Focus on Gender: Collected Papers on Gender Using DHS Data , edited by S. Kishor, 55 -68. Calverton, MD: ORC Macro. Labour Market and Wage Impacts of HIV/AIDs in Review of Agricultural Economics 28: 429-439. Dowding, Keith, ed. 2011. Encyclopedia of Power . Thousand Oaks, CA: SAGE Publications. Drimie, Scott. 2002. The Impact of HIV/AIDS on Rural Households and Land Issues in Southern and Eastern Africa : A Background Paper Prepared for the Food and Agricultural Organization, Sub -Regional Office for Southern and Eastern Africa . South Africa: FAO Integrated Rural & Regional Development. Farmer, Paul. 1992. AIDS and Accusation: Haiti and the Geography of Blame . Berkeley, CA: University of California Press. 198 . 2003. Pathologies of Power: Health, Human Rights, and the New War on the Poor. Berkeley, CA: University of California Press. Daedalus 118(2): 135 -160. Fasawe, Olufunke, Carlos Avila, Nathan Shaffer, Erik Schouten, Frank Chimbwandira, David Hoos, Olive Nakakeeto and Paul De Lay. Cost -Effectiveness Analysis of Option PLoS ONE 8(3): e57778. Ferguson, James. 1994. The Anti -Politics Machine: Development, Depoliticization, and Bureaucratic Power in Lesotho . Minneapolis, MN: University of Minnesota Press. Foster, Geoff, Abigail Kangwende, Vhumani Magezi, Talent Maphosa, Richard Ma shapa, Fadzai Mukora-Mutseyekwa, Angela Mushavi, Simba Rusakaniko, Bridget Shumba and Pemberai Zambezi. Cluster Randomized Trial on the Effect of Mother Support Groups on Retention -in-Care and PMTCT Outcomes in Zimbabwe: Study Design, Challenges, an Journal of Acquired Immune Deficiency Syndromes 67: S145 -S149 Foucault, Michel. 1977. Discipline and Punish: The Birth of the Prison . New York, NY : Random House. . 1978. The History of Sexuality, Volume 1 . New York, NY: Random H ouse. Power/Knowledge: Selected Interviews and Other Writings, 19721977, edited by Colin Gordon, 78 -108. New York, NY: Pantheon. The Foucault Effect: Studies in Governmentality, edited by Graham Burchell, Colin Gordon, and Peter Miller, 87 -104. Chicago, IL: University of Chicago Press. Ginsberg, Faye D. and Rayna Rapp, eds. 1995. Conceiving the New World Order: The Global Politics of Reproduction . Berkeley , CA : University of California Press . 199 Goffman, Erving. 1963. Stigma: Notes on the Management of Spoiled Identity . New York, NY: Simon and Schuster. Facilitating Factors to the Uptake of Antiretroviral Dru gs for Prevention of Mother -to-Child Transmission of HIV in Sub - Journal of the International AIDS Society 16(1). Government of Malawi. 2011. Malawi 2010 Demographic and Health Survey. Zomba: National Statistical Offic e and Calverton, MD: ICF Macro. Human Rights Watch. 2007. Hidden in the Mealie Meal: Gender - HIV Treatment in Zambia 19(18). New York, NY: Human Rights Watch. ica: The Significance Social Science & Medicine 64(3): 689 -700. Iliffe, John. 2006. The African AIDS Epidemic. Athens, OH: Ohio University Press. Jewkes, Rachel and Robert Morrell. 2012 African Teenage Women: Theorising Femininities and Their Connections to HIV Risk Social Science & Medicine 74(11): 1729 -1737. A Yanu - Rur al Malawian Men's Demographic Research: Special Collection 1(11): 349-372. Studies in Family Planning 35(2): 105 -115. Kal ipeni, Ezekial. 2009. Strong Women, Dangerous Times: Gender and HIV/AIDS in Africa . New York, NY: Nova Science Publishers. 200 Kasenga, Fyson, Peter Byass, Maria Emmelin and Anna - of Policy Changes on the Uptake of a PMTCT Programme in Rural Malawi: First Three Global Health Action 2. Kasenga, Fyson, Anna - - Midwifery 26(1): 27 37. Kelly, Sharo n, Kevin O'Grady, Shannon Mitchell, Barry Brown and Robert Schwartz. 2011. Predictors of Methadone Treatment Retention from a Multi -Site Study: A Survival Drug & Alcohol Dependence 117(2-3): 170 -175. Kleinman, Arthur and Rachel Hall -Clifford. Journal of Epidemiology and Community Health 63(6): 418 -419. - Center for Trade and Development. Kritz, Mary, P. Makinwa - Demographic Processes: Moving Beyond Cairo , edited by Harrie t Presser and Gita Sen, 239-260. Oxford: Oxford University Press. Non-Adherence to the Single Dose Nevirapine Regimen for the Prevention of Mother -to-Child Transmission of HIV in Bindura Town, Zimbabwe: A Cross - BioMedCentral Public Health 10: 218. Laher, Fatima, Angela Cescon, Erica Lazarus, Angela Kaida, Matamela Makongoza, Robert S. Con versations with Mothers: Exploring Reasons for Prevention of Mother -to-Child Transmission (PMTCT) Failures in the Era of Programmatic Scale - AIDS and Behavior 16(1): 91 -98. Liamputtong, Pranee, ed. 2013. Women, Motherhood and L iving with HIV/AIDS: A Cross -Cultural Perspective . Dordrecht: Springer. 201 - Women & Health 41(1): 69 -86. Link, Bruce and Jo Phel Annual Review of Sociology 27: 363-385. Livingston, Julie. 2005. Debility and the Moral Imagination in Botswana . Bloomington, IN: Indiana University Press. Lock, Margaret and Vinh -Kim Nguyen. 2010. An Anthropology of B iomedicine . Sussex: Wiley -Blackwell. Long, Carol. 2009. Contradicting Maternity: HIV -Positive Motherhood in South Africa . Johannesburg: Wits University Press. Luedke, Tracy and Harry West, eds. 2006. Borders and Healers: Brokering Therapeutic Resources i n Southeast Africa . Bloomington, IN: Indiana University Press. Lwanda, John Lloyd. 2005. Politics, Culture and Medicine in Malawi: Historical Continuities and Ruptures with Special Reference to HIV/AIDS . Zomba, Malawi: Kachere Series. Mahajan, Anish, Jen nifer Sayles, Vishal Patel, Robert Remien, Daniel Ortiz, Greg Szekeres and Thomas Coates and Recommendations for the Way Forward AIDS 22(2): 67 -79. Malawi Human Rights Commission. 2005. Cultural Practices and their Impact on the Enjoyment of Human Rights, Particularly the Rights of Women and Children in Malawi. Lilongwe: Malawi Human Rights Commission. Manzi, M., R. Zachariah, R. Teck, L. Buhendwa, J. Kazima, E. Bakali, P. Firmenich and P. Humblet. 2005High Acceptability of Voluntary Counseling and HIV -Testing But Unacceptable Loss to Follow -Up in a Prevention of Mother -to-Child HIV Transmission Tropical Medici ne and International Health 10(12): 1242 -1250. 202 Community Strategies That Improve Care and Retention Along the Prevention of Mother -to-Child Transmission Journal of the International AIDS Society 15(2): 2 Marshall, Catherine and Gretchen Rossman. 2006. Designing Qualitative Research, 4th Edition . London: Sage Publications. Massaquoi, M. , R. Zachariah, M. Manzi, O. Pasulani, D. Misindi, B. Mwagomba, A. Bauernfeind and A. Harries . Patient Retention and Attrition on Antiretroviral Treatment at District Level in Rural Malawi Transactions of the Royal Society of Tropical Medicine and Hygiene 103(6) : 594-600. McCall, George and J. L. Simmons. 1978. Identities and Interactions: An Examination of Human Associations in Everyday Life . New York, NY: Free Press. McGuire, Megan , Tamika Munyenyembe, Elisabeth Szumilin, Annette Heinzelmann, Mickael Le Paih, Nenette Bouithy and Mar Pujades -Rodr™guez . 2010. of Pre-ART Tropical Medicine and International Health 19(1): 55 -62. New York Times . February 4. Ac cessed at http:// www.nytimes.com/2015/02/05/health/failed -trial -in-africa -raises -questions -about -how-to-test-hiv-drugs.html Mepham, S., Z. Zondi, A. Mbuyazi, N. Mkhwanazi and M. L. Newell. Challenges in PMTCT Antiretroviral Adherence in Northern Kw aZulu - AIDS Care 23(6): 741 -747. Dying for Growth: Global Inequality and the Health of the Poor , ed ited by Jim Yong Kim , Joyce V. Millen , Alec Irwin and John Gershman, 225-244. Monroe, ME: Common Courage Press. Ministry of Health. 2008. Prevention of Mother to Child Transmission of HIV: A Five Year Scale Up Plan 2008 2013. Lilongwe: Government of Malaw i, Ministry of Health. 203 . 2010. Quarterly HIV Programme Report January -March 2010. Lilongwe: Government of Malawi, Ministry of Health. . 2011. Clinical Management of HIV in Children and Adults: Malawi Integrated Guidelines for Providing HIV Services . Lilongwe: Government of Malawi, Ministry of Health. Matriliny, Patriliny, and Wealth Flows in Rural Malawi African Sociological Review . 6(2): 71 -97. Mohanty, Chandra Talpade and Ann Russo. 1991. Third World Women and the Politics of Feminism . Bloomington, IN: Indiana University Press. Mohanty, Chandra Talpade. 2003. Feminism without Borders: Decolonizing Theory, Practicing Solidarity. Durham, NC: Duke University Press. Moore, Sally Falk. 1994. Anthropology and Africa: Changing Perspectives on a Changing Scene. Charlottesville, VA: University of Virginia Press. Tertiary Hospital, QECH, Blantyre, 2001: A Ca Nursing Ethics 10(2): 208 -214. Adherence to Antiretroviral Therapy for the Success of Emerging Interventions to Prevent HIV Transmission: A Wake up Call Journal of AIDS and Clinical Research 2012(4): 007. Nagar, Richa, Victoria Lawson, Linda McDowell and Susan Hanson Economic Geography 78(3): 257 -284. Nduati, Ruth, Grace Joh n, Dorothy Mbori -Ngacha, Barbra Richardson, Julie Overbaugh, Anthony Mwatha, Jeckoniah Ndinya -Achola, Job Bwayo, Francis E. Onyango, James Hughes and Joan Kreiss. Effect of Breastfeeding and Formula Feeding on Transmission of HIV -1A Randomized Clini Journal of the American Medical Association (JAMA) 283(9): 1167 -1174. 204 Inter Press Service . March 15. Accessed at http://www.ipsnews.net/2011/03/malawi -uncer tainty -over -role -for -traditional -birth-attendants/ Nguyen, Vinh - In Global Assemblages Technology, Politics, and Ethics as Anthropological Problems, edited by Aihwa Ong and St ephen Collier, 124134. Oxford: Wiley -Blackwell. Nkonki, Lungiswa L., Tanya M. Doherty, Zelee Hill, Mickey Chopra, Nikki Schaay and Carl Transmission Programmes: Simpl icity of Nevirapine does not Necessarily Lead to AIDS Research & Therapy 4(27). on for Master of Public Health Program, University of Malawi, College of Medicine. Accessed at http://www.medcol.mw/commhealth/mph/dissertations/juliet nyasulu.pdf. Nyblade, Laura, Rohini Pande, Sanyukta Mathur, Kerry MacQuarrie, Ross Kidd, Hailom Banteye rga, Aklilu Kidanu, Gad Kilonzo, Jessie Mbwambo and Virginia Bond. 2003. Disentangling HIV and AIDS Stigma in Ethiopia, Tanzania and Zambia. Washington, DC: International Center for Research on Women (ICRW). Ogden, Jessica and Laura Nyblade. 2005. Common at its Core: HIV -Related Stigma across Contexts . Washington, DC: International Center for Research on Women (ICRW). Deirdre A ., Lot J. Nyirenda , and Sally J. Theobald. -to-Child Transmission of HIV Infection: Views and Perceptions about Swallowing BioMedCentral Public Health 10: 354. Okoli, James Christian and Gail Elizabeth Lansdown. 2014. "Barriers to Successful Implementation of Prevention -of-Mother -to-Child -Transmission (PMTCT ) of HIV Programmes in Malawi and Nigeria: A Critical Literature Review Study." The Pan African Medical Journal 19: 154. Oyewumi, Oyeronke. 2005. African Gender Studies: A Reader . London: Palgrave Macmillan. 205 Painter, Thomas, Kassamba Diaby, Danielle Mati a, Lillian Lin, Toussaint Sibailly, MoŁse Reasons for Not Participating in Follow up Visits Before Starting Short Course Antiretroviral Prophylaxis for Prevention of Mother to Chi ld Transmission of HIV: British Medical Journal 329(543): 1 -13. Parker, Richard and Peter Aggleton. 2002. HIV/AIDS -related Stigma and Discrimination: A Conceptual Framework and an Agenda for Action . Washington, DC: Population Council. Discourse on the Zambian Copperbelt, 1924 The International Journal of African Historical Studies 27(2): 241 -271. Peacock, Dean 2003. Men as Partners: Activities for People Living with HIV/AIDS . Briefing Paper prepared for the Joint United Nations Programmes on HIV/AIDS Expert Group Meeting: Division for the Advancement of Women and the International Labor Organization. Critique of Anthropology 17(2): 189 -210. . hreats to Customary Tenure in Africa African Affairs 112(449): 543 -562. Petros, George, Collins Airhihenbuwa, Leickness Simbayi, Shandir Ramlagan and Brandon Culture, Health and Sexuality 8(1): 67 -77. Pfeiffer, James and Rachel Chapman . 2010. "Anthropological Perspectives on Structural Adjustment and Public Health." Annual Review of Anthropology 39: 149-165. Pfeiffer, James, Wendy Johnson, Meredith Fort, Aaron Shakow, Amy Hagopian, Steve Gloyd and Kenneth Gimbel - ning Health Systems in Poor Countries: A American Journal of Public Health 98(12): 2134 -2140. 206 Malawi since the Nine Journal of African History 24(2): 257-274. Dialectical Anthropology 5(2): 112 -125. . 1981. Matrilineal Ideology : Male -Female Dynamics in Luapula, Zambia . London and New York: Academic Press. Censoring in Clinical Trials: Review of Survival Analysis Techniques . Indian Journal of Community Medicine 217-221. Quesada, James, Laurie K. Hart and Philippe Bourgois Structural Vulnerability and Medical Anthropology Quarterly 30(4): 339 -62. Culture, Healt h, and Sexuality 7(5): 479 -492. Scheper -Hughes, Nancy. 1993. Death without Weeping: The Violence of Everyday Life in Brazil. Berkeley , CA : University of California Press. Schouten, Erik, Andreas Jahn, Dalitso Midiani, Simon Makombe, Austin Mnthambala, Ze ngani Chirwa, Anthony Harries, Joep van Oosterhout, Tarek Meguid, Anne Ben -Smith, Rony Zachariah, Lutgarde Lynen, Maria Zolfo, Wim Van Damme, Charles Gilks, Rifat Atun, -to-Child Transmission of HIV and the Health -Related Millennium Development Goals: Time for a The Lancet 378(9787): 282 284. Setel, Phillip, Milton Lewis and Maryinez Lyons, eds. 1999. Histories of Sexually Transmitted Diseases and HIV/AIDS in Sub -Saharan Africa . Westport, CT: Greenwood Press. Survival Analysis in Clinical Trials: Basics Perspectives in Clinical Research 4: 145-148. 207 Current Anthropology 40(4): 417 -448. Sontag, Susan. 1989. AIDs and its Metaphors . New York: Farrar, Strauss, and Giroux. Sprague, Courtenay, Matthew F . Chersich and Vivian Black. Constrain Access t o PMTCT and Maternal HIV Services in South Africa: A Qualitative AIDS Research & Therapy 8(10). Steady, Filomena Chioma. 2005. An Investigative Framework for Gender Research in Africa in the New Millennium . CODESRIA (Council for the Development of Social Science Research in Africa). Annual Review of Psychology 55: 519-544. Stinson, Kathryn, Andrew Boulle, David Coetzee, Elaine J. Abrams and Landon Myer . 2010. f Highly Active Antiretroviral Therapy among Pregnant Women in Cape Tropical Medicine & International Health 15(7): 825 -832. Identity Mult iplicity and Intersectionality: Hierarchical Classes Analysis (HICLAS) of Self and Identity 7(1): 89 -111. Stoler, Ann Laura. 1995. History of Sexuality and the Colonial O rder of Things . Durham, NC: Duke University Press. . 2002. Carnal Knowledge and Imperial Power: Race and the Intimate in Colonial Rule . Berkeley, CA: University of California Press. Handbook of Social Psychology Volume 1: Theory and Method , edited by Gardner Lindzey and Elliot Arensen, 311 -378. New York: Random House. Susser, Ida. 2009. AIDS, Sex, and Culture: Global Politics and Survival in Southern Africa . Malden, MA: Wiley -Blackwell. 208 Studies in Family Planning 38: 147-162. Journa l of Southern African Studies 4: 58-66. Theilgaard, Zahra, Terese Katzenstein, Mercy Chiduo, Christiane Pahl, Ib Bygbjerg, Jan Consequences of Stigmatization A Necessary Step Toward s Making HAART AIDS Research & Therapy 8(28). Strategies for Taming a Swahili Husband: Zanzibari Women's Talk About Love in Islamic Agenda : Empowering Women for Gender Equity 27(2): 65 -75. -Related Stigmatization: Ethical and Programmatic Challenges for PMTCT Programs. " Developing World Bioethics 8(1): 43 -50. Torpey, Kwasi , Mushota Kabaso, Prisca Kasonde, Rebecca Dirks, Maxmillian Bweupe, Catherine Thompson and Ya Diul Mukadi Mother -to-Child Transmission of HIV Services in a Resource - BioMedCentral Health Servi ces Research 10: 29. Tranberg Hansen, Karen. (2000). Salaula: The World of Secondhand Clothing and Zambia . Chicago, IL: University of Chicago Press. Tsing, Anna Lowenhaupt. 2005. Friction: An Ethnography of Global Connection . Princeton, NJ: Princeton Uni versity Press. Ujiji, Opondo Awiti, Anna Mia Ekstrım, Festus Ilako, Dorcas Indalo, David Wamalwa and Birgitta Rubenson. Reasoning and Deciding PMTCT -Adherence During Culture, Health & Sexuality 13(7): 829-40. 209 Strengthening PMTCT through Communication: A Review of the Literature . Accessed at http://www.unicef.org/southafrica/SAF_resources_pmtctcommunication.pdf . 2010. Malawi: PMTCT Factsheet . Accesse d at http://www.unicef.org/aids/files/Malawi_PMTCTFactsheet_2010.pdf . 2011. Countdown to Zero: Elimination of Nw HIV Infections among Children by 2015 and Keeping Their Mothers Alive . Accessed at http://www.unicef.org/french/aids/files/hiv_pmtctfacts heetMalawi.pdf USAID (United States Agency for International Development) . 2012. Gender Mainstreaming in ICT for Agriculture: Briefing Paper . Washington, DC: USAID. Vail, Leroy, ed. 1991. The Creation of Tribalism in Southern Africa . Berkeley , CA : Univer sity of California Press. Van den Borne, Francine. 2005. Trying to Survive in Times of Poverty and AIDS: Women and Multiple Partner Sex in Malawi . Amsterdam: Het Spinhuis Publishers. Van der Straten, Ariane, Petina Musara, Juliane Etima, Kubashni Woeber, Elizabeth Montgomery and Miriam Hartmann. 20 14 Results Promotes Open Discourse on Non - AIDS Research and Human Retroviruses 30: A42-3. Van Hollen, Cecilia. 2013. Birth in the Age o f AIDS: Women, Reproduction, and HIV/AIDS in India. Stanford: Stanford University Press. Vaughan, Megan. The Journal of African History 24(2): 275 -283. . 1991. Curing their Ills: Colonial Power and African Illness . Cambridge, UK: Polity Press. 210 Journal of Southern African Studies 21(3): 417 -437. Watkin Population and Development Review 30: 673-705. Weigel, R., M. C. Hosseinipour, C. Feldacker, D. Gareta, H. Tweya, J. Chiwoko, J. Gumulira, M. Kalulu, I. Mofolo, E. Kamanga, G. Mwale, A. Kadzakumanja, E. Jere and S. Phiri. infected Pregnant Women Start Antiretroviral Treatment: An Tropical Medicine & International Health 17(6): 751 -759. Whyte, Susan Reynolds, Michael Wh SAHARA: Journal of Social Aspects of AIDS 1(1): 14 -26. World Health Organization (WHO). 2008. Prevention of Mother -to-Child Transmission of HIV Generic Tr aining Package . Geneva: WHO Press. . 2009. Rapid Advice: Use of Antiretroviral Drugs for Treating Pregnant Women and Preventing HIV Infection in Infants . Geneva: WHO Press. . 2010. PMTCT Strategic Vision 2010 -2015: Preventing Mother -to-Child Transm ission of HIV to Reach the UNGASS and Millennium Development Goals . Geneva: WHO Press. . 2014. Global Health Observatory Data . Accessed at http://www.who.int/gho/en/ - Women, Motherhood, and Living with HIV/AIDS , edited by Pranee Liamputtong, 93 -109. Dordrecht: Springer. Yu, Joseph Kwong -Leung, Solomon Chih -Cheng Chen, Kuo -Yang Wang, Chao -Sung Chang, Simon Makombe, Erik Schout True Outcomes for - Bulletin of the World Health Organization 85(7): 550 -554.