Our findings show that partner disclosure rates in times of treatment scale-up are high in Kenya, Uganda and Burkina Faso, where the majority (respectively 92.7%, 77.5% and 71.4%) of respondents reported having disclosed their HIV-positive status to their partners. The disclosure rate is much lower in Malawi, where only one-third of our respondents reported having disclosed to their partners. Disclosure was positively associated in univarite analysis with being older, living in urban areas, higher education levels, and being male, while being negatively associated with membership to support groups and presence of symptoms requiring treatment. The multivariate analysis shows age, presence of symptoms requiring treatment, membership to support groups and country of recruitment to be independently associated to the prevalence of disclosure.
Analysis of the disclosure stories revealed varying reasons to disclose to partners in all four countries. In some cases, HIV-positive individuals wanted to protect their partners from infection (the reason mentioned most by women). In other cases, healthcare needs drove the disclosure process (mentioned more often by men). Both men and women also referred to upholding the integrity of their relationship as a reason to disclose to their partner. The differences by gender in reasons to disclose may be related to the fact that men more often test late, when they are already sick. Women are tested more often in PMTCT, when they are not yet sick.
One in four survey respondents scored positively on our measures for self-stigma. But in our multivariate analysis, self-stigma was not associated with disclosure. Self-stigma was rarely referred to in the open-ended responses. However, the disclosure narratives suggest that fear of enacted stigma is an important reason for non-disclosure for both men and women. Given the often negative reactions HIV-positive individuals report when they disclose to their partners, the fear of enacted stigma seems justified. Women reported more negative consequences than men, though the stories also revealed changes in response where negative reactions from partners became more supportive over time.
These findings have implications for PMTCT programs which involve routine HIV testing and early treatment initiation in antenatal care. Health workers routinely encourage HIV-positive pregnant women to refer their partners for testing, though in reality this seems easier said than done. As we have argued elsewhere, home-based testing may be a more gender-neutral way of scaling up HIV testing, as both men and women can be tested simultaneously at home .
Our interviews and focus group discussions with members of support groups suggested that support groups advocate caution when disclosing to partners. HIV-positives should disclose when they are ‘ready’ to face the consequences. In light of the negative partner responses reported by many respondents, this makes sense. A key finding of our multivariate analysis is that membership in support groups is associated with lower levels of disclosure to partners. This finding is independent of the country of recruitment, gender, and education.
The current analysis has certain limitations. Its qualitative findings cannot be generalized. Small numbers of HIV positive who answered the open ended questions made it impossible to analyse subtle differences in disclosure dynamics by country. The gender dynamics that we reported need to be further analysed in larger scale studies that can account for the differential effects of HCT practices on partner disclosure. Women who test in PMTCT are given different advice than men who test in provider initiated testing and counselling (PITC) programs in general health clinics. How do these different health care constellations affect the observed differences in motivations to disclose?
The role of support groups in disclosure processes needs to be further examined. Is the association observed between support group membership and non-disclosure caused by a tendency of people who fear enacted stigma to attend these groups? A study in five African countries found that individuals participating in regular support groups reported experiencing greater stigma (the study treated enacted and feared stigma as one). The study hypothesized experienced stigma as a reason why people participate in support groups in the first place .
Finally, our mixed-methods study was unable to explain the differences observed in partner disclosure rates between countries due to our inability to stratify the analysis (small sample size). Why are they lower in Malawi than in the other three countries? We suspect that differences in HTC – both in policy and in how services are delivered – might be important part of the story. Counsellors in Malawi possibly place less emphasis on partner disclosure, or gender relations may be different in the country. Future mixed-methods studies should combine measurement of disclosure rates with observation of counselling practices on the ground, with more attention for gender relations and kinship dynamics.
We also recommend that future studies follow longitudinal designs to better grasp the dynamics that affect disclosure to partners. New models are needed to understand how (fear of) enacted stigma is associated with, becoming members of support groups and partner disclosure. Our qualitative findings reveal that women often face negative responses when they disclose and that their fear of enacted stigma might be justified, in their individual stories. Further questions include - what role do support groups play? And how they protect people from the risks of enacted stigma?