Retaining women in programs to prevent transmission of HIV from mother to child remains a priority and a challenge for health systems in low- and middle-income countries. When women stop participating in services that provide counseling and treatment to prevent mother to child transmission—or PMTCT—they risk compromising their health because they no longer receive antiretroviral treatment for HIV, and the chances increase for their child to acquire the infection in utero, during labor and delivery, or through breastfeeding.
Through MEASURE Evaluation, funded by the United States Agency for International Development, we conducted qualitative research about gender-related factors affecting women’s participation in PMTCT services. Women, men, health workers, and stakeholders reported many factors that affect women’s participation, which, when examined together, present a complex environment that women with HIV must navigate. It was easy to imagine how various factors influenced each other and we wanted to understand the interplay of those factors and characterize that system. So we went back to our data to map those relationships. We also consulted literature and hypothesized some connections to create causal loop diagrams.
A causal loop diagram includes factors that influence an outcome over time. Variables are represented by nouns or noun phrases. We linked variables with an arrow when a change in one causes a change in another. The diagrams and full description can be found in our publication in PLoS One. Causal loop analysis is used for analyzing complex systems in which there are many variables and feedback loops. It can identify key factors, by their influence on other factors or, vice versa, on the number and kind of factors that influence them. This type of analysis can identify where service providers or policymakers could intervene to maximize impact.
We characterized three groups of factors affecting women’s participation as “poverty,” “gender,” and “health system” (health service provision and design). We found that psychosocial health—which we defined as related to self-esteem, self-efficacy, and social support—has a major influence on women’s participation in PMTCT services. It is affected by many factors associated with poverty, gender, and the health system, in addition to factors that we did not diagram, such as personal health history.
The primacy of psychosocial health in women’s participation in PMTCT services is a logical finding, but it surprised me. It’s easy to get overwhelmed when you’re analyzing a problem with many variables in entwined relationships and feedback loops. It was somewhat satisfying to emerge with a concrete finding that could strengthen intervention efforts.
Psychosocial health is a difficult variable to modify effectively because it is affected by so many factors. Thus, an intervention that addresses only one or two of those root cause factors will likely not succeed in sustaining a change in psychosocial health because other factors will continue to influence it.
We also found several so-called “balancing” loops indicating circumstances in which an action that would seem positive—for example, working outside of the home due to improved health resulting from program participation—could later restrict participation in PMTCT programs due to lack of time. Program implementers and policy makers need to consider how to address such negative balancing loops so that they can better support women’s continued participation.
The main message here is that women’s participation in PMTCT services is influenced by a complex interaction of factors associated with poverty, gender, and the health system. Thus, we need flexible interventions that can work effectively in a complex environment.
Jennifer Yourkavitch, MPH, PhD, IBCLC, is a Senior Technical Specialist at MEASURE Evaluation, ICF. This article originally appeared in ScienceSpeaks: Global ID News.