This article was downloaded by: [George Washington University] On: 02 February 2015, At: 00:58 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
AIDS Care: Psychological and Socio-medical Aspects of AIDS/HIV Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/caic20
Disclosure of HIV serostatus among pregnant and postpartum women in sub-Saharan Africa: a systematic review a
b
Melanie Tam , Anouk Amzel & B. Ryan Phelps
b
a
Department of Global Health and Population, Harvard School of Public Health, Boston, MA, USA b
United States Agency for International Development, Washington, DC, USA Published online: 30 Jan 2015.
Click for updates To cite this article: Melanie Tam, Anouk Amzel & B. Ryan Phelps (2015) Disclosure of HIV serostatus among pregnant and postpartum women in sub-Saharan Africa: a systematic review, AIDS Care: Psychological and Socio-medical Aspects of AIDS/ HIV, 27:4, 436-450, DOI: 10.1080/09540121.2014.997662 To link to this article: http://dx.doi.org/10.1080/09540121.2014.997662
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
AIDS Care, 2015 Vol. 27, No. 4, 436–450, http://dx.doi.org/10.1080/09540121.2014.997662
Disclosure of HIV serostatus among pregnant and postpartum women in sub-Saharan Africa: a systematic review Melanie Tama*, Anouk Amzelb and B. Ryan Phelpsb a
Department of Global Health and Population, Harvard School of Public Health, Boston, MA, USA; bUnited States Agency for International Development, Washington, DC, USA
Downloaded by [George Washington University] at 00:58 02 February 2015
(Received 8 October 2014; accepted 8 December 2014) Disclosure of one’s HIV status can help to improve uptake and retention in prevention of mother-to-child transmission of HIV services; yet, it remains a challenge for many women. This systematic review evaluates disclosure rates among pregnant and postpartum women in sub-Saharan Africa, timing of disclosure, and factors affecting decisions to disclose. PubMed and EMBASE databases were searched to identify relevant studies published between January 2000 and April 2014. Rates of HIV serostatus disclosure to any person ranged from 5.0% to 96.7% (pooled estimate: 67.0%, 95% CI: 55.7%–78.3%). Women who chose to disclose their status did so more often to their partners (pooled estimate: 63.9%; 95% CI: 56.7%–71.1%) than to family members (pooled estimate: 40.1; 95% CI: 26.2%–54.0%), friends (pooled estimate: 6.4%; 95% CI: 3.0%–9.8%), or religious leaders (pooled estimate: 7.1%; 95% CI: 4.3%–9.8%). Most women disclosed prior to delivery. Decisions to disclose were associated with factors related to the woman herself (younger age, first pregnancies, knowing someone with HIV, lower levels of internalized stigma, and lower levels of avoidant coping), the partner (prior history of HIV testing and higher levels of educational attainment), their partnership (no history of domestic violence and financial independence), and the household (higher quality of housing and residing without cospouses or extended family members). Interventions to encourage and support women in safely disclosing their status are needed. Keywords: HIV/AIDS; pregnant women; disclosure; PMTCT; sub-Saharan Africa; systematic review
Introduction In 2013, there were 3.2 million children living with HIV worldwide, with 91% of them in sub-Saharan Africa. There were 240,000 children who became newly infected, in part because 30% of pregnant HIV-infected women did not receive antiretroviral (ARV) drugs for prevention of mother-to-child transmission of HIV (PMTCT) (UNAIDS, 2014). The 2013 World Health Organization (WHO) consolidated guidelines on the use of ARV drugs recommended treatment for all pregnant and breast-feeding women with HIV, regardless of CD4 count or clinical stage (WHO, 2013). However, there are important barriers to initiating and retaining all eligible women on antiretroviral therapy (ART) that must be overcome. Uptake and retention in PMTCT services is often dependent upon disclosure of HIV status to partners, relatives, and others in the community (Chinkonde, Sundby, & Martinson, 2009; Nassali et al., 2009; Olagbuji et al., 2011). Yet, disclosure remains a challenge for many pregnant and postpartum women (Hardon et al., 2012; Nassali et al., 2009). Fear of intimate partner violence, abandonment, and blame for bringing the HIV infection into the family or for being unfaithful prevents many
women from sharing their HIV-infected status (Bwirire et al., 2008; Crankshaw et al., 2014; Stinson & Myer, 2012). Many have also observed other HIV-infected individuals being discriminated against and fear that they will face the same stigma if they choose to disclose (Bwirire et al., 2008; Hardon et al., 2012; Kasenga, Hurtig, & Emmelin, 2010; Madiba & Letsoalo, 2013). Studies have shown that nondisclosure has many implications on the ability of pregnant women to adhere to PMTCT recommendations. Women face suspicions and questioning about frequent visits to the clinic (Shroufi, Mafara, Saint-Sauveur, Taziwa, & Vinoles, 2013), contributing to loss-to-follow-up (Bwirire et al., 2008). Unintended disclosure is a major concern as well, especially when ART or PMTCT clinics are near one’s home or segregated from the rest of a health facility (Chinkonde et al., 2009; Lubega et al., 2013; Shroufi et al., 2013). Adherence to ARV drugs is difficult in the setting of nondisclosure, as women do not want to be seen taking the drugs (Madiba & Letsoalo, 2013). To prevent stigma that may accompany disclosure to healthcare providers, some women opt to deliver at home (Awiti Ujiji et al., 2011). Among those who choose not to disclose to home-based birth attendants, postdelivery infant prophylaxis and feeding counseling is unlikely to
*Corresponding author. Email: [email protected] The work of Anouk Amzel and B. Ryan Phelps was authored as part of their official duties as an Employee of the United States Government and is therefore a work of the United States Government. In accordance with 17 USC. 105, no copyright protection is available for such works under US Law. Melanie Tam hereby waives her right to assert copyright, but not her right to be named as co-authors in the article.
Downloaded by [George Washington University] at 00:58 02 February 2015
AIDS Care occur (Kasenga et al., 2010). Adherence to feeding options is also difficult when family and community members are not aware of the woman’s HIV status, as early discontinuation of breast-feeding or offering breast milk substitutes is often discouraged (Chinkonde et al., 2009; Kasenga et al., 2010; Madiba & Letsoalo, 2013). A systematic review was performed by Medley, Garcia-Moreno, McGill, and Maman (2004) in 2004 to assess disclosure rates, barriers to disclosure, and outcomes of disclosure among women in developing countries. Findings revealed that 16.7–86% of women disclosed their status to their sexual partners. Although fear of negative outcomes was identified as one of the main barriers to disclosure, the review found that the majority of women experienced positive reactions to revealing their status. Since 2004, a number of additional studies have been published on HIV serostatus disclosure. This systematic review aims to incorporate these new studies and determine rates of disclosure to partners, family members, friends, and religious leaders specifically among HIV-infected pregnant and postpartum women in sub-Saharan Africa. Furthermore, timing of disclosure and factors affecting pregnant women’s decisions to share their HIV status with others will be examined as well. Methods Search strategy A systematic search of published literature in PubMed and EMBASE was conducted. The following search strategy was used in PubMed: (“HIV”[Mesh] OR “Acquired Immunodeficiency Syndrome”[Mesh] OR “HIV”[tiab] OR “AIDS” [tiab]) AND (“Disclosure” [Mesh] OR “Self Disclosure”[Mesh] OR disclos*[tiab]) AND (“Infectious Disease Transmission, Vertical” [Mesh] OR “PMTCT” [tiab] OR “mother to child transmission” [tiab] OR “Pregnant Women”[Mesh] OR “Pregnancy”[Mesh] OR pregnan*[tiab]). In EMBASE, three separate searches were conducted: (1) “Human immunodeficiency virus”/exp OR “acquired immune deficiency syndrome”/exp OR HIV:ti,ab OR AIDS:ti, ab, (2) “interpersonal communication”/exp OR “self disclosure”/exp OR disclos*:ti,ab, and (3) “vertical transmission”/exp OR “pregnant woman”/exp OR “pregnancy”/exp OR PMTCT:ti,ab OR “mother to child transmission”:ti,ab OR pregnan*:ti,ab. Articles retrieved from all three searches combined were considered.
437
Studies must have also been conducted in sub-Saharan Africa, published between January 2000 and April 2014, and written in English. Studies that did not include HIVinfected pregnant or postpartum women in their study samples were excluded. Randomized controlled trials, cohort studies, case-control studies, and cross-sectional studies were all considered for this review. Reviews, editorials, letters, meeting abstracts, dissertations, studies with convenience or purposive sampling, and randomized controlled trials with only postintervention disclosure rates available were excluded. Studies that described barriers and facilitators of disclosure, but did not assess statistical significance, were excluded as well. In cases where multiple studies were based on the same cohort of women, the study with the most complete information available about disclosure rates, timing of disclosure, and predictors of disclosure was used. Data extraction Articles obtained through the search strategies described above were reviewed on the basis of their titles and abstracts. Those that satisfied the inclusion and exclusion criteria were assessed in their entirety. The articles that satisfied all criteria were then selected for this review. Characteristics of each study including location, study design, study sample, disclosure rates, timing of disclosure, and statistically significant predictors of disclosure or nondisclosure were extracted. Studies differed in the way that they presented disclosure rates. Therefore, to improve comparability, rates were recalculated and standardized for certain studies, so that all numerators were women who disclosed their HIV status, and all denominators were total number of HIV-infected women assessed for disclosure in each study. Statistical analysis Pooled proportions of women who disclosed their HIV status were obtained using Stata version 13 with random effects models to account for within- and between-study variation. Weights used were based on the DerSimonian– Laird method. Pooled estimates were calculated for rates of disclosure to any person (among studies that did not specify the person disclosed to), partners, family members, friends, and religious leaders.
Results Study characteristics
Inclusion and exclusion criteria For inclusion in this review, studies must have examined rates of HIV status disclosure or statistically significant factors associated with disclosure or nondisclosure among HIV-infected pregnant or postpartum women.
The initial literature search yielded 1042 articles. One hundred and sixty-four duplicates were removed and the remaining 878 abstracts were screened against the inclusion and exclusion criteria. Eight hundred and ten articles were excluded in this step, leaving 68 full-text
438
M. Tam et al.
articles to be assessed for eligibility. Forty-seven articles met all search criteria for this systematic review (Figure 1). The studies spanned 14 countries in sub-Saharan Africa. Studies were published between 2001 and 2014, with 3 published between 2001 and 2004, 18 published between 2005 and 2009, and 26 published in 2010 or later.
Downloaded by [George Washington University] at 00:58 02 February 2015
Disclosure rates Forty-three studies reported on disclosure status among HIV-infected pregnant and postpartum women (Table 1). Twenty-two of these studies reported rates of HIV serostatus disclosure to any person (but did not specify who the person was), ranging from 5.0% to 96.7% (pooled estimate: 67.0%; 95% CI: 55.7%–78.3%; Figure 2). Thirty-eight studies looked at rates of disclosure to husbands or partners, which ranged from 30.2% to 93.3% (pooled estimate: 63.9%; 95% CI: 56.7%–71.1%; Figure 3). Seventeen studies reported rates of disclosure to family members. Of these, nine studies considered family members as a broad category and reported disclosure rates ranging from 20.6% to 65.6% (pooled estimate: 40.1; 95% CI: 26.2%–54.0%; Figure 4). The other eight studies disaggregated disclosure rates by specific family members or combined disclosure to family members with disclosure to partners, friends, or others. Rates of disclosure to friends were reported by seven studies. Excluding one study that grouped disclosure to friends with disclosure to family members, the rates ranged from 1.7% to 15.1% (pooled estimate: 6.4%; 95% CI: 3.0%–9.8%; Figure 4). Rates of
Figure 1. Flow diagram of study selection process.
disclosure to religious leaders, as reported by three studies, ranged from 6.0% to 10.9% (pooled estimate: 7.1%; 95% CI: 4.3%–9.8%; Figure 4). Timing of disclosure Twelve studies analyzed the timing of disclosure and changes in the proportion of women who disclosed their HIV status over the course of the studies (Table 1). The majority of HIV-infected pregnant women disclosed to their partner prior to delivery (Bedell, van Lettow, & Landes, 2014; Brou et al., 2007). According to three studies conducted in Tanzania, Kenya, and Zimbabwe, 60–80% of women who disclosed their HIV statuses did so within a few days after testing (Kiula, Damian, & Msuya, 2013; Mucheto et al., 2011; Roxby et al., 2013), while 35.4–57% of those who had not disclosed did not intend to do so (Kiula et al., 2013; Mucheto et al., 2011; Roxby et al., 2013). Those who have disclosed after delivery have been found to do so around the period of weaning or resuming sexual activity (Brou et al., 2007). Eight studies that measured disclosure rates at different intervals reported increased rates over time. While Sendo et al. found that 47.7% of women disclosed their HIV status to their sexual partners less than one month after being diagnosed, 10 additional women disclosed 1–2 months after diagnosis, 11 disclosed 3–4 months after diagnosis, and 6 women disclosed after four months (Sendo, Cherie, & Erku, 2013). In another study, 35% of women who had not disclosed to their partners and 39% of women who had not disclosed to others at baseline did disclose their statuses by follow-up three months after delivery (Makin et al., 2008). Among women who were counseled individually in Zambia, 49.3% of those who had not disclosed to their partner at enrollment did so by the six month postpartum visit (Semrau et al., 2005). In South Africa, Mundell et al. studied the effects of a 10-session psychosocial support group intervention. At baseline, 71% of women in the intervention group and 60% of women in the control group had disclosed to someone. By the eight-month follow-up visit, 97% of women in the intervention group and 82% of women in the comparison group had disclosed to someone (Mundell et al., 2011). The rate of nondisclosure to partners has been observed to decrease from 13.5% in the first six months since diagnosis to 5.4% after one year postdiagnosis (Sagay et al., 2006). Another study found that 21.5% of women who had not disclosed at enrollment did so after one year. Two years after delivery, the percentage of women who disclosed their status to partners increased to 71.9%, from 53.4% immediately after diagnosis (Irungu et al., 2012). In a study by Antelman et al., (2001) 40% of women disclosed to their partners four
Downloaded by [George Washington University] at 00:58 02 February 2015
Table 1. Summary of studies: disclosure rates and timing of disclosure. Proportion of women who discloseda (%)
Study design
Person disclosed to
Bedell et al. (2014) Lawani, Onyebuchi, Iyoke, Onoh, and Nkwo (2014) Onono, Cohen, Jerop, Bukusi, and Turan (2014) Bucagu, Bizimana Jde, Muganda, and Humblet (2013) Kiula et al. (2013)
Malawi Nigeria
Cross-sectional Cross-sectional
Partner Overall
89.0 89.2
Kenya
Prospective cohort
Overall
68.1
Rwanda
Prospective cohort
Partner
81.1
Tanzania
Cross-sectional
Overall Partner Family member
60 40.8 45.6
Mnyani and McIntyre (2013)
South Africa
Cross-sectional
Ndondoki et al. (2013) Odongkara et al. (2013) Roxby et al. (2013)
Côte d’Ivoire Uganda
Cross-sectional Cross-sectional
Partner Parent Sibling Friend Other Partner Partner
90.9 34.9 24.0 17.1 2.9 57.0 83.1
Kenya
Prospective cohort
Partner
48.6
Sendo et al. (2013)
Ethiopia
Cross-sectional
Partner Family members
72.9 40.2
Udigwe et al. (2013) Ekama et al. (2012)
Nigeria Nigeria
Cross-sectional Cross-sectional
Hardon et al. (2012)
Burkina Faso Kenya Malawi Uganda
Cross-sectional
Partner Overall Partner Overall Partner Family member Friend
90.5 86.5 84.1 82.5 30.2 50.8 7.9
Timing of disclosure 87% disclosed prior to delivery.
Of those who disclosed to a partner, 80% did so within seven days of receiving results. Forty-three percent of those who did not disclose to their partner planned to do so in the future; 57% did not plan to disclose.
Of those reporting timing of disclosure to partners, 60% disclosed within three days of testing; 79% disclosed within 30 days of positive test. Of the original group of women who had not disclosed, 21.5% did so after one year, 61.5% still had not, and 35.4% did not intend to. Time since diagnosis: 4 months: 7.7%
439
Country
AIDS Care
Study
440
Proportion of women who discloseda (%)
Study
Country
Study design
Person disclosed to
Irungu et al. (2012)
Burkina Faso Kenya
Prospective cohort
Overall Partner Family member Friend
64.4 53.4 20.6 3.0
Muluye, Woldeyohannes, Gizachew, and Tiruneh (2012) Torpey, Mandala, et al. (2012) Watson-Jones, Balira, Ross, Weiss, and Mabey (2012) Kirsten et al. (2011)
Ethiopia
Cross-sectional
Partner Family member
87.6 65.6
Zambia
Retrospective cohort Prospective cohort
Partner
93.3
Overall
36.5
Overall
93.0
Overall Partner Overall
89.4 70.8 55.2
Tanzania
Tanzania
Mepham et al. (2011)
South Africa
Prospective cohort RCT
Mucheto et al. (2011)
Zimbabwe
Cross-sectional
Mundell et al. (2011)
South Africa
Quasiexperimental
Overall
65.2
Olagbuji et al. (2011)
Nigeria
Cross-sectional
Tejiokem et al. (2011)
Cameroon
88 88 63.9 2.4 6 81.0
Kuonza et al. (2010) Nacro et al. (2010)
Zimbabwe Burkina Faso
Prospective cohort Cross-sectional Prospective cohort
Overall Partner Family member Person outside family Religious leader Partner Partner Partner Any family member (including partners) Sibling
48.1 40.7 48.6 11.4
Timing of disclosure Disclosure to Partner: 2 weeks after delivery: 59.4%; 3 months after delivery: 65.0%; 6 months after delivery: 65.4%; 12 months after delivery: 69.5%; 18 months after delivery: 72.3%; 24 months after delivery: 71.9%
Of all who disclosed, 79.5% did so a few days after testing, 1.8% when deciding on delivery method, and 19.8% were tested with their partner/spouse. Of those who did not disclose, 38% do not intend to do so. At two-month follow-up, 94% (intervention group) and 78% (comparison group) disclosed to someone. At the eight-month follow-up, 97% (intervention group) and 82% (comparison group) had disclosed.
M. Tam et al.
Downloaded by [George Washington University] at 00:58 02 February 2015
Table 1. (Continued)
Downloaded by [George Washington University] at 00:58 02 February 2015
Table 1. (Continued) Proportion of women who discloseda (%)
Country
Study design
Person disclosed to
Peltzer and Mlambo (2010) Ezechi et al. (2009) Ezegwui et al. (2009)
South Africa
Cross-sectional
Overall
74.1
Nigeria Nigeria
Cross-sectional Cross-sectional
Delvaux et al. (2009)
Rwanda
Case-control
Desgrees-du-Lou et al. (2009) Maru et al. (2009)
Côte d’Ivoire
Prospective cohort Cross-sectional
Partner Overall Partner Mother Father Sister Brother Mother-in-law Sister-in-law Friend Priest Partner Other Partner
90.3 96.7 91.3 14.1 10.9 19.6 10.9 2.2 3.3 3.3 10.9 83.9 72.9 42.8
Peltzer, Chao, and Dana (2009)
South Africa
Prospective cohort
W. E. Sadoh and A. E. Sadoh (2009)
Nigeria
Cross-sectional
Tonwe-Gold et al. (2009) Adejuyigbe, Orji, Onayade, Makinde, and Anyabolu (2008) Bii et al. (2008) Makin et al. (2008)
Côte d’Ivoire
Prospective cohort Prospective cohort
Partner Family/others Overall Partner Mother Sister Friend Overall Partner Parent Sister Pastor Partner
30.8 42.9 59.5 51.7 20.7 16.4 1.7 91.9 72.6 8.1 4.8 6.5 52.8
Partner Family/friends
52.9 16.3
Nigeria
Nigeria
Kenya South Africa
Partner Overall Partner Parents Other family member Friend
51.5 59.0 47.7 of those who had partners 11.9
Three months postdelivery: Overall: 81%; Partners: 67% of those who had partners; Others: 59%. Thirty-five percent of those who had not disclosed to their partner at baseline did so, and 39% of those who had not disclosed to others at baseline did so.
441
Cross-sectional Prospective cohort
Timing of disclosure
AIDS Care
Study
442
Table 1. (Continued)
Downloaded by [George Washington University] at 00:58 02 February 2015
Country
Study design
Person disclosed to
Proportion of women who discloseda (%)
Brou et al. (2007)
Côte d’Ivoire
Prospective cohort
Partner
14.7 10.2 46.2
Creek et al. (2007)
Botswana
Doherty et al. (2007)
South Africa
Overall Partner Family member Overall
84.9 63.0 34.2 52.5
Albrecht et al. (2006)
Zambia
Partner
52.8
Sagay et al. (2006)
Nigeria
Prospective cohort (intervention) Prospective cohort Prospective cohort Cross-sectional
Partner Other
89.3 63.3
Moth, Ayayo, and Kaseje (2005) Semrau et al. (2005)
Kenya
Cross-sectional
Overall
5.0
Zambia
Prospective cohort
Partner
67.4b
Etiebet, Fransman, Forsyth, Coetzee, and Hussey (2004) Kilewo et al. (2001)
South Africa
Cross-sectional
Partner
60.4
Tanzania
Prospective cohort
Antelman et al. (2001)
Tanzania
RCT
Overall Partner Family member Overall Partner Female relative
22.2 16.7 5.6 51.0 40 36
a
Defined as number of women who disclosed divided by total number of HIV-infected women in each study. Among individually counseled women.
b
Timing of disclosure
Most disclosed to their partner prior to delivery. Of women who breastfed, 55.7% disclosed before delivery, 17% between delivery and resumption of sexual activity, and 19% between delivery and weaning. Of women who did not breast-feed, 65.8% disclosed before delivery and 8% between delivery and resumption of sexual activity. Disclosure after delivery peaked around weaning or resuming sexual activity.
Disclosure to partner: 2 years (AOR = 12.28; 95% CI: 2.53– 59.52), smooth relationship before test (AOR: 6.76; 95% CI: 2.14–12.31), discussion about voluntary counseling and testing before HIV test (AOR: 12.28; 95% CI: 2.53,59.52). Mucheto Nondisclosure: Believing disclosure would cause divorce (AOR = 7.82, p = 0.03), living with extended family et al. (2011) (AOR = 10.3, p = 0.01), and needing spousal approval of HIV testing (AOR = 0.11, p < 0.001). Nondisclosure: Nulliparous (pregnant for the first time; p = 0.024) and not married (p = 0.026) Olagbuji et al. (2011) Makin et al. (2008) Disclosure: Disclosure to partner prior to enrollment is associated with being married compared with not being married (AOR = 2.32; 95% CI: 1.20–4.47), prior discussion about testing (AOR = 4.19; 95% CI: 2.34–7.49), and having a partner with tertiary education (AOR = 2.76; 95% CI: 1.29–5.88). Early disclosure to others associated with better housing (AOR = 1.26; 95%CI: 1.06–1.49) and knowing someone with HIV (AOR = 2.13; 95% CI:1.20–3.76). Postenrollment disclosure to partner higher amongmarried women compared with women who are not married (AOR = 5.31, 95% CI:1.25–22.58). Nondisclosure: Likelihood of disclosure to partners was lower with past experience of violence (AOR = 0.48;95% CI: 0.24–0.97) and higher levels of internalized stigma (AOR =0.91; 95% CI: 0.84–0.98). Likelihood of disclosure to others waslower with financial dependence on partners (AOR = 0.46; 95% CI: 0.25–0.85), older age (AOR = 0.91; 95% CI: 0.84–0.97), and increased levels of avoidant coping (AOR = 0.84; 95% CI: 0.72–0.97). Brou et al. (2007) Disclosure: Formula feeding (AOR = 1.54; 95% CI: 1.04–2.27). Nondisclosure: Lesslikely to disclose when living with their own family but without theirpartner than living with their partner only (AOR = 0.29, 95% CI: 0.17–0.50) and when they had a co-spouse rather than being the only wife (AOR = 0.51,95% CI: 0.31–0.83). Antelman Disclosure: Knowing someone with HIV/AIDS (RR: 2.28; 95% CI: 1.36–3.82). Compared with having ≥6 et al. (2001) lifetime sexual partners, more likely to disclose if had one partner (RR: 2.41, 95% CI: 1.48–3.92) or 2–5 lifetime partners (RR: 1.54, 95% CI: 1.48–3.92). Nondisclosure: Compared to married ≥2 years, less likely to disclose if married
AIDS Care: Psychological and Socio-medical Aspects of AIDS/HIV Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/caic20
Disclosure of HIV serostatus among pregnant and postpartum women in sub-Saharan Africa: a systematic review a
b
Melanie Tam , Anouk Amzel & B. Ryan Phelps
b
a
Department of Global Health and Population, Harvard School of Public Health, Boston, MA, USA b
United States Agency for International Development, Washington, DC, USA Published online: 30 Jan 2015.
Click for updates To cite this article: Melanie Tam, Anouk Amzel & B. Ryan Phelps (2015) Disclosure of HIV serostatus among pregnant and postpartum women in sub-Saharan Africa: a systematic review, AIDS Care: Psychological and Socio-medical Aspects of AIDS/ HIV, 27:4, 436-450, DOI: 10.1080/09540121.2014.997662 To link to this article: http://dx.doi.org/10.1080/09540121.2014.997662
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
AIDS Care, 2015 Vol. 27, No. 4, 436–450, http://dx.doi.org/10.1080/09540121.2014.997662
Disclosure of HIV serostatus among pregnant and postpartum women in sub-Saharan Africa: a systematic review Melanie Tama*, Anouk Amzelb and B. Ryan Phelpsb a
Department of Global Health and Population, Harvard School of Public Health, Boston, MA, USA; bUnited States Agency for International Development, Washington, DC, USA
Downloaded by [George Washington University] at 00:58 02 February 2015
(Received 8 October 2014; accepted 8 December 2014) Disclosure of one’s HIV status can help to improve uptake and retention in prevention of mother-to-child transmission of HIV services; yet, it remains a challenge for many women. This systematic review evaluates disclosure rates among pregnant and postpartum women in sub-Saharan Africa, timing of disclosure, and factors affecting decisions to disclose. PubMed and EMBASE databases were searched to identify relevant studies published between January 2000 and April 2014. Rates of HIV serostatus disclosure to any person ranged from 5.0% to 96.7% (pooled estimate: 67.0%, 95% CI: 55.7%–78.3%). Women who chose to disclose their status did so more often to their partners (pooled estimate: 63.9%; 95% CI: 56.7%–71.1%) than to family members (pooled estimate: 40.1; 95% CI: 26.2%–54.0%), friends (pooled estimate: 6.4%; 95% CI: 3.0%–9.8%), or religious leaders (pooled estimate: 7.1%; 95% CI: 4.3%–9.8%). Most women disclosed prior to delivery. Decisions to disclose were associated with factors related to the woman herself (younger age, first pregnancies, knowing someone with HIV, lower levels of internalized stigma, and lower levels of avoidant coping), the partner (prior history of HIV testing and higher levels of educational attainment), their partnership (no history of domestic violence and financial independence), and the household (higher quality of housing and residing without cospouses or extended family members). Interventions to encourage and support women in safely disclosing their status are needed. Keywords: HIV/AIDS; pregnant women; disclosure; PMTCT; sub-Saharan Africa; systematic review
Introduction In 2013, there were 3.2 million children living with HIV worldwide, with 91% of them in sub-Saharan Africa. There were 240,000 children who became newly infected, in part because 30% of pregnant HIV-infected women did not receive antiretroviral (ARV) drugs for prevention of mother-to-child transmission of HIV (PMTCT) (UNAIDS, 2014). The 2013 World Health Organization (WHO) consolidated guidelines on the use of ARV drugs recommended treatment for all pregnant and breast-feeding women with HIV, regardless of CD4 count or clinical stage (WHO, 2013). However, there are important barriers to initiating and retaining all eligible women on antiretroviral therapy (ART) that must be overcome. Uptake and retention in PMTCT services is often dependent upon disclosure of HIV status to partners, relatives, and others in the community (Chinkonde, Sundby, & Martinson, 2009; Nassali et al., 2009; Olagbuji et al., 2011). Yet, disclosure remains a challenge for many pregnant and postpartum women (Hardon et al., 2012; Nassali et al., 2009). Fear of intimate partner violence, abandonment, and blame for bringing the HIV infection into the family or for being unfaithful prevents many
women from sharing their HIV-infected status (Bwirire et al., 2008; Crankshaw et al., 2014; Stinson & Myer, 2012). Many have also observed other HIV-infected individuals being discriminated against and fear that they will face the same stigma if they choose to disclose (Bwirire et al., 2008; Hardon et al., 2012; Kasenga, Hurtig, & Emmelin, 2010; Madiba & Letsoalo, 2013). Studies have shown that nondisclosure has many implications on the ability of pregnant women to adhere to PMTCT recommendations. Women face suspicions and questioning about frequent visits to the clinic (Shroufi, Mafara, Saint-Sauveur, Taziwa, & Vinoles, 2013), contributing to loss-to-follow-up (Bwirire et al., 2008). Unintended disclosure is a major concern as well, especially when ART or PMTCT clinics are near one’s home or segregated from the rest of a health facility (Chinkonde et al., 2009; Lubega et al., 2013; Shroufi et al., 2013). Adherence to ARV drugs is difficult in the setting of nondisclosure, as women do not want to be seen taking the drugs (Madiba & Letsoalo, 2013). To prevent stigma that may accompany disclosure to healthcare providers, some women opt to deliver at home (Awiti Ujiji et al., 2011). Among those who choose not to disclose to home-based birth attendants, postdelivery infant prophylaxis and feeding counseling is unlikely to
*Corresponding author. Email: [email protected] The work of Anouk Amzel and B. Ryan Phelps was authored as part of their official duties as an Employee of the United States Government and is therefore a work of the United States Government. In accordance with 17 USC. 105, no copyright protection is available for such works under US Law. Melanie Tam hereby waives her right to assert copyright, but not her right to be named as co-authors in the article.
Downloaded by [George Washington University] at 00:58 02 February 2015
AIDS Care occur (Kasenga et al., 2010). Adherence to feeding options is also difficult when family and community members are not aware of the woman’s HIV status, as early discontinuation of breast-feeding or offering breast milk substitutes is often discouraged (Chinkonde et al., 2009; Kasenga et al., 2010; Madiba & Letsoalo, 2013). A systematic review was performed by Medley, Garcia-Moreno, McGill, and Maman (2004) in 2004 to assess disclosure rates, barriers to disclosure, and outcomes of disclosure among women in developing countries. Findings revealed that 16.7–86% of women disclosed their status to their sexual partners. Although fear of negative outcomes was identified as one of the main barriers to disclosure, the review found that the majority of women experienced positive reactions to revealing their status. Since 2004, a number of additional studies have been published on HIV serostatus disclosure. This systematic review aims to incorporate these new studies and determine rates of disclosure to partners, family members, friends, and religious leaders specifically among HIV-infected pregnant and postpartum women in sub-Saharan Africa. Furthermore, timing of disclosure and factors affecting pregnant women’s decisions to share their HIV status with others will be examined as well. Methods Search strategy A systematic search of published literature in PubMed and EMBASE was conducted. The following search strategy was used in PubMed: (“HIV”[Mesh] OR “Acquired Immunodeficiency Syndrome”[Mesh] OR “HIV”[tiab] OR “AIDS” [tiab]) AND (“Disclosure” [Mesh] OR “Self Disclosure”[Mesh] OR disclos*[tiab]) AND (“Infectious Disease Transmission, Vertical” [Mesh] OR “PMTCT” [tiab] OR “mother to child transmission” [tiab] OR “Pregnant Women”[Mesh] OR “Pregnancy”[Mesh] OR pregnan*[tiab]). In EMBASE, three separate searches were conducted: (1) “Human immunodeficiency virus”/exp OR “acquired immune deficiency syndrome”/exp OR HIV:ti,ab OR AIDS:ti, ab, (2) “interpersonal communication”/exp OR “self disclosure”/exp OR disclos*:ti,ab, and (3) “vertical transmission”/exp OR “pregnant woman”/exp OR “pregnancy”/exp OR PMTCT:ti,ab OR “mother to child transmission”:ti,ab OR pregnan*:ti,ab. Articles retrieved from all three searches combined were considered.
437
Studies must have also been conducted in sub-Saharan Africa, published between January 2000 and April 2014, and written in English. Studies that did not include HIVinfected pregnant or postpartum women in their study samples were excluded. Randomized controlled trials, cohort studies, case-control studies, and cross-sectional studies were all considered for this review. Reviews, editorials, letters, meeting abstracts, dissertations, studies with convenience or purposive sampling, and randomized controlled trials with only postintervention disclosure rates available were excluded. Studies that described barriers and facilitators of disclosure, but did not assess statistical significance, were excluded as well. In cases where multiple studies were based on the same cohort of women, the study with the most complete information available about disclosure rates, timing of disclosure, and predictors of disclosure was used. Data extraction Articles obtained through the search strategies described above were reviewed on the basis of their titles and abstracts. Those that satisfied the inclusion and exclusion criteria were assessed in their entirety. The articles that satisfied all criteria were then selected for this review. Characteristics of each study including location, study design, study sample, disclosure rates, timing of disclosure, and statistically significant predictors of disclosure or nondisclosure were extracted. Studies differed in the way that they presented disclosure rates. Therefore, to improve comparability, rates were recalculated and standardized for certain studies, so that all numerators were women who disclosed their HIV status, and all denominators were total number of HIV-infected women assessed for disclosure in each study. Statistical analysis Pooled proportions of women who disclosed their HIV status were obtained using Stata version 13 with random effects models to account for within- and between-study variation. Weights used were based on the DerSimonian– Laird method. Pooled estimates were calculated for rates of disclosure to any person (among studies that did not specify the person disclosed to), partners, family members, friends, and religious leaders.
Results Study characteristics
Inclusion and exclusion criteria For inclusion in this review, studies must have examined rates of HIV status disclosure or statistically significant factors associated with disclosure or nondisclosure among HIV-infected pregnant or postpartum women.
The initial literature search yielded 1042 articles. One hundred and sixty-four duplicates were removed and the remaining 878 abstracts were screened against the inclusion and exclusion criteria. Eight hundred and ten articles were excluded in this step, leaving 68 full-text
438
M. Tam et al.
articles to be assessed for eligibility. Forty-seven articles met all search criteria for this systematic review (Figure 1). The studies spanned 14 countries in sub-Saharan Africa. Studies were published between 2001 and 2014, with 3 published between 2001 and 2004, 18 published between 2005 and 2009, and 26 published in 2010 or later.
Downloaded by [George Washington University] at 00:58 02 February 2015
Disclosure rates Forty-three studies reported on disclosure status among HIV-infected pregnant and postpartum women (Table 1). Twenty-two of these studies reported rates of HIV serostatus disclosure to any person (but did not specify who the person was), ranging from 5.0% to 96.7% (pooled estimate: 67.0%; 95% CI: 55.7%–78.3%; Figure 2). Thirty-eight studies looked at rates of disclosure to husbands or partners, which ranged from 30.2% to 93.3% (pooled estimate: 63.9%; 95% CI: 56.7%–71.1%; Figure 3). Seventeen studies reported rates of disclosure to family members. Of these, nine studies considered family members as a broad category and reported disclosure rates ranging from 20.6% to 65.6% (pooled estimate: 40.1; 95% CI: 26.2%–54.0%; Figure 4). The other eight studies disaggregated disclosure rates by specific family members or combined disclosure to family members with disclosure to partners, friends, or others. Rates of disclosure to friends were reported by seven studies. Excluding one study that grouped disclosure to friends with disclosure to family members, the rates ranged from 1.7% to 15.1% (pooled estimate: 6.4%; 95% CI: 3.0%–9.8%; Figure 4). Rates of
Figure 1. Flow diagram of study selection process.
disclosure to religious leaders, as reported by three studies, ranged from 6.0% to 10.9% (pooled estimate: 7.1%; 95% CI: 4.3%–9.8%; Figure 4). Timing of disclosure Twelve studies analyzed the timing of disclosure and changes in the proportion of women who disclosed their HIV status over the course of the studies (Table 1). The majority of HIV-infected pregnant women disclosed to their partner prior to delivery (Bedell, van Lettow, & Landes, 2014; Brou et al., 2007). According to three studies conducted in Tanzania, Kenya, and Zimbabwe, 60–80% of women who disclosed their HIV statuses did so within a few days after testing (Kiula, Damian, & Msuya, 2013; Mucheto et al., 2011; Roxby et al., 2013), while 35.4–57% of those who had not disclosed did not intend to do so (Kiula et al., 2013; Mucheto et al., 2011; Roxby et al., 2013). Those who have disclosed after delivery have been found to do so around the period of weaning or resuming sexual activity (Brou et al., 2007). Eight studies that measured disclosure rates at different intervals reported increased rates over time. While Sendo et al. found that 47.7% of women disclosed their HIV status to their sexual partners less than one month after being diagnosed, 10 additional women disclosed 1–2 months after diagnosis, 11 disclosed 3–4 months after diagnosis, and 6 women disclosed after four months (Sendo, Cherie, & Erku, 2013). In another study, 35% of women who had not disclosed to their partners and 39% of women who had not disclosed to others at baseline did disclose their statuses by follow-up three months after delivery (Makin et al., 2008). Among women who were counseled individually in Zambia, 49.3% of those who had not disclosed to their partner at enrollment did so by the six month postpartum visit (Semrau et al., 2005). In South Africa, Mundell et al. studied the effects of a 10-session psychosocial support group intervention. At baseline, 71% of women in the intervention group and 60% of women in the control group had disclosed to someone. By the eight-month follow-up visit, 97% of women in the intervention group and 82% of women in the comparison group had disclosed to someone (Mundell et al., 2011). The rate of nondisclosure to partners has been observed to decrease from 13.5% in the first six months since diagnosis to 5.4% after one year postdiagnosis (Sagay et al., 2006). Another study found that 21.5% of women who had not disclosed at enrollment did so after one year. Two years after delivery, the percentage of women who disclosed their status to partners increased to 71.9%, from 53.4% immediately after diagnosis (Irungu et al., 2012). In a study by Antelman et al., (2001) 40% of women disclosed to their partners four
Downloaded by [George Washington University] at 00:58 02 February 2015
Table 1. Summary of studies: disclosure rates and timing of disclosure. Proportion of women who discloseda (%)
Study design
Person disclosed to
Bedell et al. (2014) Lawani, Onyebuchi, Iyoke, Onoh, and Nkwo (2014) Onono, Cohen, Jerop, Bukusi, and Turan (2014) Bucagu, Bizimana Jde, Muganda, and Humblet (2013) Kiula et al. (2013)
Malawi Nigeria
Cross-sectional Cross-sectional
Partner Overall
89.0 89.2
Kenya
Prospective cohort
Overall
68.1
Rwanda
Prospective cohort
Partner
81.1
Tanzania
Cross-sectional
Overall Partner Family member
60 40.8 45.6
Mnyani and McIntyre (2013)
South Africa
Cross-sectional
Ndondoki et al. (2013) Odongkara et al. (2013) Roxby et al. (2013)
Côte d’Ivoire Uganda
Cross-sectional Cross-sectional
Partner Parent Sibling Friend Other Partner Partner
90.9 34.9 24.0 17.1 2.9 57.0 83.1
Kenya
Prospective cohort
Partner
48.6
Sendo et al. (2013)
Ethiopia
Cross-sectional
Partner Family members
72.9 40.2
Udigwe et al. (2013) Ekama et al. (2012)
Nigeria Nigeria
Cross-sectional Cross-sectional
Hardon et al. (2012)
Burkina Faso Kenya Malawi Uganda
Cross-sectional
Partner Overall Partner Overall Partner Family member Friend
90.5 86.5 84.1 82.5 30.2 50.8 7.9
Timing of disclosure 87% disclosed prior to delivery.
Of those who disclosed to a partner, 80% did so within seven days of receiving results. Forty-three percent of those who did not disclose to their partner planned to do so in the future; 57% did not plan to disclose.
Of those reporting timing of disclosure to partners, 60% disclosed within three days of testing; 79% disclosed within 30 days of positive test. Of the original group of women who had not disclosed, 21.5% did so after one year, 61.5% still had not, and 35.4% did not intend to. Time since diagnosis: 4 months: 7.7%
439
Country
AIDS Care
Study
440
Proportion of women who discloseda (%)
Study
Country
Study design
Person disclosed to
Irungu et al. (2012)
Burkina Faso Kenya
Prospective cohort
Overall Partner Family member Friend
64.4 53.4 20.6 3.0
Muluye, Woldeyohannes, Gizachew, and Tiruneh (2012) Torpey, Mandala, et al. (2012) Watson-Jones, Balira, Ross, Weiss, and Mabey (2012) Kirsten et al. (2011)
Ethiopia
Cross-sectional
Partner Family member
87.6 65.6
Zambia
Retrospective cohort Prospective cohort
Partner
93.3
Overall
36.5
Overall
93.0
Overall Partner Overall
89.4 70.8 55.2
Tanzania
Tanzania
Mepham et al. (2011)
South Africa
Prospective cohort RCT
Mucheto et al. (2011)
Zimbabwe
Cross-sectional
Mundell et al. (2011)
South Africa
Quasiexperimental
Overall
65.2
Olagbuji et al. (2011)
Nigeria
Cross-sectional
Tejiokem et al. (2011)
Cameroon
88 88 63.9 2.4 6 81.0
Kuonza et al. (2010) Nacro et al. (2010)
Zimbabwe Burkina Faso
Prospective cohort Cross-sectional Prospective cohort
Overall Partner Family member Person outside family Religious leader Partner Partner Partner Any family member (including partners) Sibling
48.1 40.7 48.6 11.4
Timing of disclosure Disclosure to Partner: 2 weeks after delivery: 59.4%; 3 months after delivery: 65.0%; 6 months after delivery: 65.4%; 12 months after delivery: 69.5%; 18 months after delivery: 72.3%; 24 months after delivery: 71.9%
Of all who disclosed, 79.5% did so a few days after testing, 1.8% when deciding on delivery method, and 19.8% were tested with their partner/spouse. Of those who did not disclose, 38% do not intend to do so. At two-month follow-up, 94% (intervention group) and 78% (comparison group) disclosed to someone. At the eight-month follow-up, 97% (intervention group) and 82% (comparison group) had disclosed.
M. Tam et al.
Downloaded by [George Washington University] at 00:58 02 February 2015
Table 1. (Continued)
Downloaded by [George Washington University] at 00:58 02 February 2015
Table 1. (Continued) Proportion of women who discloseda (%)
Country
Study design
Person disclosed to
Peltzer and Mlambo (2010) Ezechi et al. (2009) Ezegwui et al. (2009)
South Africa
Cross-sectional
Overall
74.1
Nigeria Nigeria
Cross-sectional Cross-sectional
Delvaux et al. (2009)
Rwanda
Case-control
Desgrees-du-Lou et al. (2009) Maru et al. (2009)
Côte d’Ivoire
Prospective cohort Cross-sectional
Partner Overall Partner Mother Father Sister Brother Mother-in-law Sister-in-law Friend Priest Partner Other Partner
90.3 96.7 91.3 14.1 10.9 19.6 10.9 2.2 3.3 3.3 10.9 83.9 72.9 42.8
Peltzer, Chao, and Dana (2009)
South Africa
Prospective cohort
W. E. Sadoh and A. E. Sadoh (2009)
Nigeria
Cross-sectional
Tonwe-Gold et al. (2009) Adejuyigbe, Orji, Onayade, Makinde, and Anyabolu (2008) Bii et al. (2008) Makin et al. (2008)
Côte d’Ivoire
Prospective cohort Prospective cohort
Partner Family/others Overall Partner Mother Sister Friend Overall Partner Parent Sister Pastor Partner
30.8 42.9 59.5 51.7 20.7 16.4 1.7 91.9 72.6 8.1 4.8 6.5 52.8
Partner Family/friends
52.9 16.3
Nigeria
Nigeria
Kenya South Africa
Partner Overall Partner Parents Other family member Friend
51.5 59.0 47.7 of those who had partners 11.9
Three months postdelivery: Overall: 81%; Partners: 67% of those who had partners; Others: 59%. Thirty-five percent of those who had not disclosed to their partner at baseline did so, and 39% of those who had not disclosed to others at baseline did so.
441
Cross-sectional Prospective cohort
Timing of disclosure
AIDS Care
Study
442
Table 1. (Continued)
Downloaded by [George Washington University] at 00:58 02 February 2015
Country
Study design
Person disclosed to
Proportion of women who discloseda (%)
Brou et al. (2007)
Côte d’Ivoire
Prospective cohort
Partner
14.7 10.2 46.2
Creek et al. (2007)
Botswana
Doherty et al. (2007)
South Africa
Overall Partner Family member Overall
84.9 63.0 34.2 52.5
Albrecht et al. (2006)
Zambia
Partner
52.8
Sagay et al. (2006)
Nigeria
Prospective cohort (intervention) Prospective cohort Prospective cohort Cross-sectional
Partner Other
89.3 63.3
Moth, Ayayo, and Kaseje (2005) Semrau et al. (2005)
Kenya
Cross-sectional
Overall
5.0
Zambia
Prospective cohort
Partner
67.4b
Etiebet, Fransman, Forsyth, Coetzee, and Hussey (2004) Kilewo et al. (2001)
South Africa
Cross-sectional
Partner
60.4
Tanzania
Prospective cohort
Antelman et al. (2001)
Tanzania
RCT
Overall Partner Family member Overall Partner Female relative
22.2 16.7 5.6 51.0 40 36
a
Defined as number of women who disclosed divided by total number of HIV-infected women in each study. Among individually counseled women.
b
Timing of disclosure
Most disclosed to their partner prior to delivery. Of women who breastfed, 55.7% disclosed before delivery, 17% between delivery and resumption of sexual activity, and 19% between delivery and weaning. Of women who did not breast-feed, 65.8% disclosed before delivery and 8% between delivery and resumption of sexual activity. Disclosure after delivery peaked around weaning or resuming sexual activity.
Disclosure to partner: 2 years (AOR = 12.28; 95% CI: 2.53– 59.52), smooth relationship before test (AOR: 6.76; 95% CI: 2.14–12.31), discussion about voluntary counseling and testing before HIV test (AOR: 12.28; 95% CI: 2.53,59.52). Mucheto Nondisclosure: Believing disclosure would cause divorce (AOR = 7.82, p = 0.03), living with extended family et al. (2011) (AOR = 10.3, p = 0.01), and needing spousal approval of HIV testing (AOR = 0.11, p < 0.001). Nondisclosure: Nulliparous (pregnant for the first time; p = 0.024) and not married (p = 0.026) Olagbuji et al. (2011) Makin et al. (2008) Disclosure: Disclosure to partner prior to enrollment is associated with being married compared with not being married (AOR = 2.32; 95% CI: 1.20–4.47), prior discussion about testing (AOR = 4.19; 95% CI: 2.34–7.49), and having a partner with tertiary education (AOR = 2.76; 95% CI: 1.29–5.88). Early disclosure to others associated with better housing (AOR = 1.26; 95%CI: 1.06–1.49) and knowing someone with HIV (AOR = 2.13; 95% CI:1.20–3.76). Postenrollment disclosure to partner higher amongmarried women compared with women who are not married (AOR = 5.31, 95% CI:1.25–22.58). Nondisclosure: Likelihood of disclosure to partners was lower with past experience of violence (AOR = 0.48;95% CI: 0.24–0.97) and higher levels of internalized stigma (AOR =0.91; 95% CI: 0.84–0.98). Likelihood of disclosure to others waslower with financial dependence on partners (AOR = 0.46; 95% CI: 0.25–0.85), older age (AOR = 0.91; 95% CI: 0.84–0.97), and increased levels of avoidant coping (AOR = 0.84; 95% CI: 0.72–0.97). Brou et al. (2007) Disclosure: Formula feeding (AOR = 1.54; 95% CI: 1.04–2.27). Nondisclosure: Lesslikely to disclose when living with their own family but without theirpartner than living with their partner only (AOR = 0.29, 95% CI: 0.17–0.50) and when they had a co-spouse rather than being the only wife (AOR = 0.51,95% CI: 0.31–0.83). Antelman Disclosure: Knowing someone with HIV/AIDS (RR: 2.28; 95% CI: 1.36–3.82). Compared with having ≥6 et al. (2001) lifetime sexual partners, more likely to disclose if had one partner (RR: 2.41, 95% CI: 1.48–3.92) or 2–5 lifetime partners (RR: 1.54, 95% CI: 1.48–3.92). Nondisclosure: Compared to married ≥2 years, less likely to disclose if married
