Approaches to the development of human rights indicators for women’s health

We propose that human rights indicators related to women’s health be developed. First, we present two published examples from Burma and Afghanistan that illustrate how specific violations of human rights have been related to women’s health outcomes in the literature. We then review the state of current human rights measures relevant to women’s health, propose a process through which new indicators can be developed, and discuss potential challenges. An instrument that can be incorporated into household epidemiological surveys could be used to identify human rights violations pertaining to the health of women, with the ultimate goal of improving health and realizing rights for these populations. 

The framework of rights has an extensive history in women’s health. Recent interest in human rights issues particularly affecting women is reflected in Millennium Development Goals and the passage of Resolution 11 on preventable maternal mortality and morbidity (Table 1). Given the range of health and rights interactions for women and their vulnerable status, we propose that human rights indicators related to women’s health be developed. Such indicators could generate concrete data to shape policy from a rights-based perspective and galvanize efforts to address rights abuses with negative consequences. We highlight examples of the current use of human rights indicators in this context, discuss challenges in indicator development, and recommend future directions.

Current examples of human rights approaches to women’s health issues
 In Burma and Afghanistan, researchers have attempted to generate indicators of women’s health and rights. The relationship between individual-level exposure to human rights violations and access to basic women’s health services was examined in four eastern Burmese states marked by long-standing conflict.1 In these areas, forced relocation of ethnic minority communities has been a commonly reported rights abuse.1,2 Forcibly displaced women were more likely to have an unmet need for modern contraceptives (OR=1.7, 95% CI: 1.2-2.5) and were substantially less likely (OR=0.2 [0.1-0.5]) to receive basic antenatal care interventions such as malaria screening, blood and urine tests, iron supplementation, or insecticide-treated bed nets. In these analyses a discrete and context-relevant experience (forced displacement) was used as a human rights indicator for women’s health outcomes.
In a second example, prevalence of current major depression was estimated among Afghan women from Taliban-controlled areas (78%) and non-Taliban areas (28%) during the period of Taliban rule.3 Rates of suicidal ideation (65% vs. 18%) and attempted suicide (16% vs. 9%) were also substantially higher where the Taliban were in control. Furthermore, women in these areas were more likely to report that the Taliban had a large/extreme effect on their depressive symptoms (65% vs. 30%) or on suicidal ideation and/or attempt (22% vs. 2%). Here, the human rights indicator is not discrete, but rather a summary measure of violations experienced by women living under the Taliban’s policy of enforced gender inequality.
These cases illustrate approaches to quantifying relationships between rights violations and women’s health. Analyses were conducted to assess 1) a fairly specific human rights indicator, namely, the role of forced displacement, and 2) a multifaceted indicator of Taliban control. The specifics of context and the challenges of data collection (reaching women in conflict zones in Burma, or women under enforced societal segregation in Afghanistan) make local adaptation and innovation essential to the development and use of simple indicators.
Human rights in relation to women can also be examined by using health indicators as prima facie evidence of “right to health” violations. The maternal mortality ratio is described as the “indicator with the greatest disparity between rich and poor countries,” and is often followed by an argument that improved women’s health is an unrealized “right.” Although this usage is important and enhances the visibility of human rights, new data must be generated that can further bolster accountability to these rights.
Status of human rights indicators

Human rights measures in general are still under development. Many researchers have used national or area-level measures (for example, state protection against sex discrimination) as rights indicators. However, given that state-level indicators (such as laws) are implemented to varying degrees, there is a potential disconnect between their ideal impact and the way these rights actually affect individuals’ lives. This explains why analyses of relationships between ratification of human rights and health outcomes have shown no consistent associations.4 Given this mismatch, a human rights approach to women’s health measures would benefit from measurement that extends to individual-level experience.

There have been recent advances in creating indicators of the right to the “highest attainable standard of health” (as laid out in Article 12 from the International Covenant on Economic, Social and Cultural Rights). One guide for identifying health indictors could be based on the Millennium Development Goals. We can also build on Backman et al., who published a list of 72 features of a country’s health system reflecting the right to health in 2008. 5 Some of the indicators directly relating to women included: 1) treaty-bound state protection against sex discrimination, 2) centralized collection and public availability of information on maternal deaths, 3) state law requiring sexual and reproductive health education for youth, and 4) national-level statistics on the prevalence of violence against women, the proportion of women accessing prenatal care, and maternal mortality.

WHO’s Department of Reproductive Health and Research (RHR) has also made strides in the development of indicators to measure women’s and children’s health. As of 2001, WHO had selected 17 indicators for the global monitoring of reproductive health.6 These indicators were compiled in a toolkit for use by countries at the regional and national levels. The RHR has used this set of indicators to monitor international progress towards the MDGs and other global goals, as well as to estimate the global burden of reproductive morbidity and related mortality. In accordance with a rights-based approach, the RHR has also used data on reproductive health indicators, such as maternal mortality and coverage of maternity care, to track how well countries are upholding human rights norms and obligations.7

While Backman et al.’s and WHO’s indicators provide a basis on which to build,  they also pose several challenges. WHO has noted the need for additional indicators measuring prevention and quality of reproductive health care.6 Being closely interconnected, some factors act simultaneously as outcomes and risk factors, and are difficult to assess given lack of standardized data collection approaches. Because the majority of the proposed quantitative indicators relate to area-level indicators (such as the proportion of women experiencing a condition or protected by treaties or laws), they are not easily integrated into studies that collect data at the individual or household level, where so many rights violations occur. Creating a set of indicators that could be incorporated into individual and/or household surveys could bridge the gap between what women actually experience in daily life and the structural conditions and ideals embodied in the law.

We have outlined a process through which such indicators could be developed. This process begins with establishing partnerships at the local level and is followed by a qualitative investigation in which both rights violations and health outcomes relevant to that setting are identified. On the basis of these efforts, individual or household survey instruments are developed and implemented using standard methods to explicitly document both human rights measures and health indicators. Analysis of these factors leads to quantification of the association(s) between rights violations and health outcomes, and pathways through which any observed relationship might arise can be posited. The process can help identify points of intervention and, through mobilization of community partners, context-appropriate strategies to address these issues can be developed.

A concrete example of this process is illustrated in Figure 1; we initially undertook the activities listed in the separate steps of the figure as part of our study of human rights violations related to maternal and obstetric care on the Thailand-Burma border. In this case, we had already established relationships (step 1) with a range of community-based organizations and health leadership committees representing the health interests of ethnic populations in eastern Burma (Mon, Karen, Karenni, Shan); our role included providing technical assistance on health programming and evaluations, including population-based surveys to estimate health indicators. After our partners expressed an interest in expanding the set of domains being measured within their target populations to include human rights, we jointly conducted a series of qualitative assessments (step 2) with key stakeholders, providers, and community members to identify the types of rights commonly violated. In this case, these included various types of attacks, arbitrary arrest/detention, forced labor, forced relocation, rape, etc. Eventually, this process led to a context-appropriate quantitative instrument (step 3a) that allowed for measuring individual or household exposure to rights violations and characterizing these in terms of frequency, timing, and duration of exposure. When this module was combined with other more standard modules assessing coverage of maternal health interventions and outcomes (step 3b), standard epidemiological methods were then applied to population-based data. These data were collected using the combined instruments to analyze and quantify associations between rights violations and critical health outcomes of interest (step 4). As we described above, examples of such quantified associations include our estimate that among forcibly displaced women, the odds of unmet need for contraception were 1.7 times higher and the odds of receiving no antenatal care services in the prior pregnancy were 5.9 times higher than women not reporting this rights violation.1 We utilized information about the influence of human rights violations on health indicators to work with partners to develop a context-specific approach to service delivery that would help overcome the impact of these ongoing violations (step 5a). Specifically in this case, we posited that a mobile, three-tiered network of community-based providers, with maximum flexibility in both task-shifting and mobility, would achieve the main aim of bringing care directly to forcibly displaced communities, thereby mitigating to some extent the coverage issues. We then designed and implemented this very model over three years and closely monitored the changes in health outcomes (step 6) that were achieved using this human-rights based approach. In this particular case, we observed a 35% reduction in unmet need, substantially increased access to essential maternal health interventions, and a tenfold increase in the presence of delivery attendants able to provide emergency obstetric care.