United Nations Commission on Information and Accountability for Women’s and Children’s Health

The first CoIA report published in May 2011, Keeping Promises, Measuring Results , outlines a clear and simple framework for accountability comprising three interconnected processes – monitor, review, and act – while at the same time linking accountability for resources to results . This framework emphasizes the need to collect valid and accurate data, analyze and interpret the data to determine whether improvements have been made and commitments kept, and, finally, use the information and the evidence from the review as guidance to make recommendations that are actionable – all in tandem with tracking and reporting results. The reporting of progress is a fundamental component of accountability, and it needs to occur at many levels: health facilities, districts, countries, and regions. The institutions at these levels must be accountable for their results, or lack thereof, to stakeholders in civil society, to national and global partners, and, most important, to citizens, particularly women and their families.

This foundation of accountability will continue to apply to the SDGs and also to the agenda for the new global strategy for 2016–2030. The framework embraces the key principles of accountability contained in the global strategy, including national leadership and ownership of results and the strengthening of each country’s capacity to monitor and evaluate. It is built on the foundation of the right to the highest attainable standard of health and equity in health. In keeping with the post-2015 agenda, it will incorporate a stronger human rights-based approach to accountability, while also continuing to focus on improving measurement, data, and harmonization, but with a greater emphasis on disaggregated data that will make it possible to address the equity concerns of the SDGs and the global strategy – to better identify who is being left behind and why . Finally, there will be a greater emphasis on incorporating people and communities into the accountability dialogue.

The first step in the monitor–review–act cycle is to determine the extent of the problem. Assessing progress toward achievement of the maternal and child health targets requires monitoring how many women are dying from pregnancy-related problems, how many infants and children under five are dying, and how many infants are being born. Countdown 2015, launched in 2003, was a collaboration of more than 40 different institutions and organizations with the Lancet as a central partner. Its task was to provide information on global and national progress toward achieving MDG 4 and 5. Countdown 2015 reports that there were 523,000 maternal deaths in 1990 compared with only 289,000 in 2013 . It is important however to ask how these numbers were obtained and whether or not they are reliable. The answers will become even more important when monitoring the impact of activities and interventions aimed at achieving the 2030 objectives not just at national levels but also at subnational levels and among women and children at greater risk. Accountability cannot exist without data . For maternal and child health, having functional civil registration and vital statistics (CRVS) systems in place is “a fundamental requirement for evidence-based decisions and accountability” .

Civil Registration and Vital Statistics

In its 2011 report, CoIA made three recommendations for obtaining “better information for better results,” the first of which was on the subject of vital events: “By 2015, all countries [should] have taken significant steps to establish a system for registration of births, deaths, and causes of death, and [should] have well-functioning health information systems that combine data from facilities, administrative sources, and surveys” . The UN defines civil registration as the “universal, continuous, permanent, and compulsory recording of the occurrence and characteristics of vital events pertaining to the population” . It is real-time counting that provides data down to the district and local level. When a birth or death occurs, it should be recorded. It is different from estimates based on surveys and censuses, which although they often have a fair degree of accuracy only reflect events that took place in the past (e.g., survey-based estimates for child mortality are two to three years old, and for maternal mortality they are about eight to ten years old). Surveys also have the limitation that they provide national estimates without adequate disaggregation at the subnational, district, or local level. Given the large differences often found in results within a country, as well as the geographic, ethnic, and economic disparities within populations, true accountability to women, families, and communities requires disaggregated data for the areas in which they live and where they receive care. This information will help to ensure that the most vulnerable and economically deprived populations are given priority and that resources are channeled appropriately .

Countries have continued to devote time and attention to the development of functioning CRVS systems. The latest data suggest that 64 out of 75 priority countries either have CRVS assessment and national plans in place or are in the process of establishing them .

There are a number of reasons births and deaths are poorly registered. For example, there may be a lack of knowledge about the importance of collecting the information. In addition, families are busy when there is a birth and grieving when there is a death. Alternatively, there may be discriminatory laws or practices that prevent or discourage registration . Higher levels of government may not have identified registration as a priority, and often there is a lack of leadership to call attention to its importance. Champions are needed to strengthen CRVS. In order to scale up, barriers to reporting need to be identified and addressed, and specific guidelines need to be developed to enhance reporting. In many cases, cost alone has been a significant barrier to full CRVS, with estimates suggesting that in order to scale up and sustain CRVS in 73 of the countdown countries (excluding China and India) US$ 3.82 billion would need to be spent .

Some innovative approaches have been developed to overcome these barriers – for example, the use of mobile technology in Uganda, the deployment of community health workers to facilitate reporting in Bangladesh, and interinstitutional collaboration in Nicaragua . Although it may appear to be a technical issue, registering a birth is a matter of human rights. In 2014, the United Nations General Assembly adopted a resolution on birth registration in which it states that it is a universal human right “closely linked to the realization of many other rights such as the right to health and the right to education” . Although the accuracy of the data may be suboptimal, there seems to have been overall progress in birth registration throughout the world, with the registration of children under five increasing over a 10-year period from 58% to 65% in 2013, and the work continues .

Registration of deaths tends to lag behind registration of births. However, once a child’s death is registered then neonatal, infant, and under-five mortality can easily be reported because the definition does not depend on an accurate cause of death, rather only on the age at which it occurred. Therefore, it is important for the age at death to be accurately reported.

However, the identification of a maternal death depends on accurate identification of the fact that the woman was pregnant when she died or within 42 days of her death. This information is not consistently added to the death certificate. It may not be available to the individual registering the death, or the family may not want the information to be known. Abortion deaths are especially difficult to identify particularly where abortion is illegal. Families may not even be aware the woman was pregnant, or they may not want to report that she underwent an abortion. While some causes of death are automatically identified as maternal deaths (e.g., postpartum hemorrhage, eclampsia), the other causes are not (e.g., sepsis alone, unless it is identified as puerperal sepsis). It is important to train those who fill out death certificates to improve the quality of this information.

An important aspect of maternal death registration is whether or not a maternal death is a “notifiable event” – in other words, whether reporting is mandatory under the law. Countdown 2015 has provided information in this regard on 75 priority countries. Of these countries, 68 have available data on trends. For example, 47 of the 68 countries (69%) reported having a policy on maternal death notification as of 2013–2014, compared with only 34% with such a policy in 2008. Maternal death notification is an important way to ensure that all maternal deaths are identified. However, without a functioning CRVS system, legislation alone is often not sufficiently effective. Capacity building is needed to improve reporting and registration of maternal death, as well as the analysis of its causes . The WHO has developed an interactive training tool (ICD) directed at individuals and organizations responsible for completing death certificates . Some countries such as South Africa have developed their own training programs in an effort to improve the quality of death certificates (Personal communication: Debbie Bradshaw).

One strategy being used to improve the capture of maternal deaths is the inclusion of pregnancy check boxes on the death certificate, although there is little information about how successful they are . Another strategy for increasing the identification of maternal deaths is linking birth and death certificates to determine whether a woman’s death was associated with a birth. However, this strategy is only effective when both the birth and the death are recorded, which may not be the case if the child was stillborn or if the maternal death was associated with an abortion or an ectopic pregnancy .

In May 2014, the WHO and the World Bank published a report that summarized why CRVS matters and how it can be improved and scaled up worldwide . Given the importance of CRVS in tracking maternal and child health (MCH) outcomes, indicators have been developed to enhance global and national accountability for these systems. Global indicators include the number of countries with a national CRVS plan based on a comprehensive assessment, as well as the number of countries with functioning CRVS committees and a legal framework for CRVS. Examples of outcomes that can and should be monitored and reported by countries include the percentage of births registered, the percentage of deaths in which the cause of death is medically certified and reported, and the percentage of maternal and newborn deaths registered and investigated by a medical practitioner. Ultimately, the goal should be to have the capacity to provide accurate annual reports on maternal and newborn deaths and their causes. The target set by the WHO and the World Bank is universal registration of births and deaths, including cause of death, by 2030 .

The new global strategy includes, for the first time, a target for the reduction of stillbirths. Measuring progress toward this target will require strengthening registration systems to capture stillbirths in both developed and developing countries . Stillbirths are currently not included in the WHO standard death certificate; they require a separate form.

Clearly, civil registration contributes greatly to effective monitoring of and accountability for maternal and perinatal deaths. However, as already noted, it often fails to correctly identify the cause of death, and it does not include the factors that may have led to or contributed to the death. This information is critical in order to ensure that interventions to reduce maternal and perinatal mortality are effective. A more in-depth review is needed in order to obtain this information.

Civil Registration and Vital Statistics

In its 2011 report, CoIA made three recommendations for obtaining “better information for better results,” the first of which was on the subject of vital events: “By 2015, all countries [should] have taken significant steps to establish a system for registration of births, deaths, and causes of death, and [should] have well-functioning health information systems that combine data from facilities, administrative sources, and surveys” . The UN defines civil registration as the “universal, continuous, permanent, and compulsory recording of the occurrence and characteristics of vital events pertaining to the population” . It is real-time counting that provides data down to the district and local level. When a birth or death occurs, it should be recorded. It is different from estimates based on surveys and censuses, which although they often have a fair degree of accuracy only reflect events that took place in the past (e.g., survey-based estimates for child mortality are two to three years old, and for maternal mortality they are about eight to ten years old). Surveys also have the limitation that they provide national estimates without adequate disaggregation at the subnational, district, or local level. Given the large differences often found in results within a country, as well as the geographic, ethnic, and economic disparities within populations, true accountability to women, families, and communities requires disaggregated data for the areas in which they live and where they receive care. This information will help to ensure that the most vulnerable and economically deprived populations are given priority and that resources are channeled appropriately .

Countries have continued to devote time and attention to the development of functioning CRVS systems. The latest data suggest that 64 out of 75 priority countries either have CRVS assessment and national plans in place or are in the process of establishing them .

There are a number of reasons births and deaths are poorly registered. For example, there may be a lack of knowledge about the importance of collecting the information. In addition, families are busy when there is a birth and grieving when there is a death. Alternatively, there may be discriminatory laws or practices that prevent or discourage registration . Higher levels of government may not have identified registration as a priority, and often there is a lack of leadership to call attention to its importance. Champions are needed to strengthen CRVS. In order to scale up, barriers to reporting need to be identified and addressed, and specific guidelines need to be developed to enhance reporting. In many cases, cost alone has been a significant barrier to full CRVS, with estimates suggesting that in order to scale up and sustain CRVS in 73 of the countdown countries (excluding China and India) US$ 3.82 billion would need to be spent .

Some innovative approaches have been developed to overcome these barriers – for example, the use of mobile technology in Uganda, the deployment of community health workers to facilitate reporting in Bangladesh, and interinstitutional collaboration in Nicaragua . Although it may appear to be a technical issue, registering a birth is a matter of human rights. In 2014, the United Nations General Assembly adopted a resolution on birth registration in which it states that it is a universal human right “closely linked to the realization of many other rights such as the right to health and the right to education” . Although the accuracy of the data may be suboptimal, there seems to have been overall progress in birth registration throughout the world, with the registration of children under five increasing over a 10-year period from 58% to 65% in 2013, and the work continues .

Registration of deaths tends to lag behind registration of births. However, once a child’s death is registered then neonatal, infant, and under-five mortality can easily be reported because the definition does not depend on an accurate cause of death, rather only on the age at which it occurred. Therefore, it is important for the age at death to be accurately reported.

However, the identification of a maternal death depends on accurate identification of the fact that the woman was pregnant when she died or within 42 days of her death. This information is not consistently added to the death certificate. It may not be available to the individual registering the death, or the family may not want the information to be known. Abortion deaths are especially difficult to identify particularly where abortion is illegal. Families may not even be aware the woman was pregnant, or they may not want to report that she underwent an abortion. While some causes of death are automatically identified as maternal deaths (e.g., postpartum hemorrhage, eclampsia), the other causes are not (e.g., sepsis alone, unless it is identified as puerperal sepsis). It is important to train those who fill out death certificates to improve the quality of this information.

An important aspect of maternal death registration is whether or not a maternal death is a “notifiable event” – in other words, whether reporting is mandatory under the law. Countdown 2015 has provided information in this regard on 75 priority countries. Of these countries, 68 have available data on trends. For example, 47 of the 68 countries (69%) reported having a policy on maternal death notification as of 2013–2014, compared with only 34% with such a policy in 2008. Maternal death notification is an important way to ensure that all maternal deaths are identified. However, without a functioning CRVS system, legislation alone is often not sufficiently effective. Capacity building is needed to improve reporting and registration of maternal death, as well as the analysis of its causes . The WHO has developed an interactive training tool (ICD) directed at individuals and organizations responsible for completing death certificates . Some countries such as South Africa have developed their own training programs in an effort to improve the quality of death certificates (Personal communication: Debbie Bradshaw).

One strategy being used to improve the capture of maternal deaths is the inclusion of pregnancy check boxes on the death certificate, although there is little information about how successful they are . Another strategy for increasing the identification of maternal deaths is linking birth and death certificates to determine whether a woman’s death was associated with a birth. However, this strategy is only effective when both the birth and the death are recorded, which may not be the case if the child was stillborn or if the maternal death was associated with an abortion or an ectopic pregnancy .

In May 2014, the WHO and the World Bank published a report that summarized why CRVS matters and how it can be improved and scaled up worldwide . Given the importance of CRVS in tracking maternal and child health (MCH) outcomes, indicators have been developed to enhance global and national accountability for these systems. Global indicators include the number of countries with a national CRVS plan based on a comprehensive assessment, as well as the number of countries with functioning CRVS committees and a legal framework for CRVS. Examples of outcomes that can and should be monitored and reported by countries include the percentage of births registered, the percentage of deaths in which the cause of death is medically certified and reported, and the percentage of maternal and newborn deaths registered and investigated by a medical practitioner. Ultimately, the goal should be to have the capacity to provide accurate annual reports on maternal and newborn deaths and their causes. The target set by the WHO and the World Bank is universal registration of births and deaths, including cause of death, by 2030 .

The new global strategy includes, for the first time, a target for the reduction of stillbirths. Measuring progress toward this target will require strengthening registration systems to capture stillbirths in both developed and developing countries . Stillbirths are currently not included in the WHO standard death certificate; they require a separate form.

Clearly, civil registration contributes greatly to effective monitoring of and accountability for maternal and perinatal deaths. However, as already noted, it often fails to correctly identify the cause of death, and it does not include the factors that may have led to or contributed to the death. This information is critical in order to ensure that interventions to reduce maternal and perinatal mortality are effective. A more in-depth review is needed in order to obtain this information.