The Republic of Indonesia, home to over 240 million people, is the world’s fourth most populous nation. Ethnically, culturally, and economically diverse, the Indonesian people are broadly dispersed across an archipelago of more than 13,000 islands. Rapid urbanization has given rise to one megacity (Jakarta) and to 10 other major metropolitan areas. And yet about half of Indonesians make their homes in rural areas of the country.
Indonesia, a signatory to the United Nations Millennium Declaration,1 has committed to achieving the Millennium Development Goals (MDGs). Moreover, the president of the Republic has been selected to play a key role in the design of the Post-2015 Development Agenda. However, recent estimates suggest that Indonesia will not achieve by the target date of 2015 MDG 4—reduction by two-thirds of the 1990 under-5 infant mortality rate (number of children under age 5 who die per 1,000 live births)—and MDG 5—reduction by three-quarters of the 1990 maternal mortality ratio (number of maternal deaths within 28 days of childbirth in a given year per 100,000 live births).2 Although much has been achieved, complex and indeed difficult challenges will have to be overcome before maternal and infant mortality are brought into the MDG-prescribed range.
This study was conceived during a visit to Indonesia in 2010 by Professor Bruce Alberts in his capacity as a science envoy to the Muslim world on behalf of U.S. President Barack Obama. In this role, Professor Alberts visited several communities in Indonesia, meeting with national leaders, government officials, and scientists and their students to discuss the role of science in a developing country. As a former president of the U.S. National Academy of Sciences (NAS), Professor Alberts was particularly interested in working with the Indonesian Academy of Sciences (AIPI), to enhance its role in providing technical advice to the government and the nation. The two academies and the U.S. Agency for International Development (USAID) agreed to collaborate in a study of the long-standing problem of maternal and neonatal mortality that might help Indonesia attain the targets set by the MDGs.
The task of the academies in this project was to evaluate the quality and consistency of the existing data on maternal and neonatal mortality; devise a strategy to achieve the Millennium Development Goals related to maternal mortality, fetal mortality (stillbirths), and neonatal mortality; and identify the highest-priority interventions and propose steps toward development of an effective implementation plan. NAS and AIPI were also to use the study process to train AIPI staff so that they could independently carry out similar science policy studies in the future.
Although the defined tasks relate specifically to the MDGs and the maternal and newborn mortality ratios, the study committee recognized that efforts to regulate pregnancy through fertility control and family planning also played an important role in mitigating the mortality risk attending childbirth. For example, pregnancies among very young and older women or excessively shorter or longer intervals between births are known to carry higher risk of complications for both mother and baby. However, the committee decided to maintain its focus on the problems affecting women already pregnant, according to the MDG definition, and to the issues surrounding childbirth that cause the great majority of maternal deaths. That
2 Following common usage, we use the abbreviation MMR for maternal mortality ratio, the number of maternal deaths per 100,000 live births, and NMR for neonatal mortality rate, the number of newborn deaths per 1,000 live births. The analogous maternal mortality rate is obtained by dividing maternal deaths by person-years of exposure. The maternal mortality rate can be converted to the MMR and expressed per 100,000 live births by dividing the rate by the general fertility rate.
said, a brief review of the history of fertility control in Indonesia appears in the appendix to this report, and the desirability of continued support for family planning and prevention of unwanted pregnancies is included in the recommendations.
The under-5 mortality rate has three components: newborns (neonates), infants (< 1 year), and children 1–5 years of age. The clinical events that undergird maternal, fetal (stillbirth), and neonatal deaths are often the same and as such are closely linked—that is, increased risk of maternal death implies a high risk of both stillbirth and neonatal death. It follows, then, that interventions aimed at reducing the incidence of maternal mortality will also reduce the prevalence of stillbirths and of neonatal deaths. Common determinants of maternal, fetal, and neonatal mortality include but are not limited to poor access to care, the poor quality of that care, the lack of education of many women, and living in poverty. Older infants up to 5 years of age are also affected by malnutrition and infectious diseases such as malaria. Nevertheless, in Indonesia the mortality rate for older infants has improved faster than that for mothers and neonates. Given that constellation, the joint committee convened by the U.S. and Indonesian academies of sciences for this study resolved to focus on the complex of problems afflicting mothers and newborns.
Mortality rates at childbirth are also affected by the levels and age patterns of fertility. In Indonesia, a vigorous national family planning program coupled with social change (particularly in the promotion of basic education for girls) saw fertility as measured by the total fertility rate (TFR) decline by more than 50 percent from the late 1960s to the early 2000s, with the TFR reaching a level of 2.6 by 2002, according to the Indonesia Demographic and Health Survey (IDHS). This overall decline has included major declines in fertility among women at higher-risk ages, particularly those in their teens and over 40, as well as marked increases in the average intervals between births.
With little recorded change in fertility rates since 2000 and a consistent gap in the unmet need for contraception as measured by the 2007 and 2012 IDHS, there is scope for further declines in fertility that would likely have a positive impact on maternal health. These declines should be encouraged through a reinvigorated family planning program that would be an integral part of the safe childbirth services offered to Indonesian women.
The current Indonesian programs directed at reducing maternal and neonatal mortality have proven insufficient to meet the MDG targets, according to the commonly accepted data sources. Many of the actions needed must be implemented locally, in or near where people live. Because maternal and neonatal mortality and some of their major determinants vary considerably among districts and municipalities, designing locally appropriate solutions is all important for program success. One striking example is the disparity in mortality rates between the more densely populated areas of the country and those that are more sparsely populated and thus face far more substantial transport and communications challenges. Life-saving programs are generally more cost-effective where population density, personal wealth, and quality of facilities are highest, but cost-effectiveness and absolute impact must be balanced against equity in a country as geographically diverse as Indonesia. Moreover, in 2001 Indonesia adopted a decentralization policy that shifted political and budgetary power to the districts and municipalities, and so most decisions are now made at that level, which further increases the need for locally feasible and acceptable strategies.
In 1989 the Indonesian government launched a Midwifery Education Rapid Training Program to increase access to basic midwifery services in the villages. By 1998 this initiative had led to the establishment of midwifery academies with a three-year curriculum. The National Education System that followed in 2003 transferred all sectoral education programs, including health education, to the Ministry of Education. This resulted in a 15-fold increase in the number of midwifery academies. Concurrently, the number of midwives increased from 52,000 in 2006 to over 200,000 in 2012. This number of trainees exceeds the capacity of any academy to offer adequate hands-on childbirth training experience in the face of a relatively stable national birth cohort. Also, many of the midwives constituting the current workforce are products of the earlier one-year course that offered little in the way of hands-on experience with childbirth emergencies. Moreover, many of the village-bound midwives are relocating to the cities in search of patients, thereby diluting their presence in underserved rural districts. With recent statistics indicating the country’s inability to effectively reduce its maternal mortality rate, the large midwife contin-
gent does not appear to be in a position to advance maternity care to the level needed to save maternal and newborn lives in Indonesia.
Most births in Indonesia still take place in the home of the mother or in a birthing room in the home of a village-based midwife. Regrettably, however, it is not usually possible to predict with high reliability when a pregnant woman will experience a life-threatening emergency during childbirth. Moreover, the system of referral and emergency transfer of the mother from home to hospital in many areas of Indonesia, especially those with a low population density, has not proven effective in saving lives.
The Joint Committee on Reducing Maternal and Neonatal Mortality in Indonesia of the U.S. National Academy of Sciences and the Indonesian Academy of Sciences has developed eight recommendations designed to guide future efforts to reduce maternal and neonatal mortality in Indonesia (these recommendations are described more fully in Chapter 8). The recommendations, tempered by the unique challenges of Indonesia, are based on established medical practice paradigms that have been shown to dramatically reduce mortality at childbirth in developed countries.
Recommendation 1—Facilities. Indonesia should seek to ensure that all births occur in a certified (i.e., independently accredited) facility with either basic or comprehensive emergency obstetric and newborn care (BEmONC or CEmONC) capabilities. As defined by the World Health Organization (WHO), CEmONC capabilities consist of trained doctors, nurses, and midwives who possess the abilities to perform a cesarean section, provide a blood transfusion, undertake vacuum delivery of a baby, administer magnesium sulfate and antibiotics, and render the other services necessary to remedy those conditions that most commonly cause maternal or neonatal mortality in Indonesia. BEmONC facilities, by contrast, lack the capability to perform a cesarean section and to administer a blood transfusion—the actions necessary to prevent the most common direct causes of maternal death in Indonesia. As an immediate first step toward realizing universal access to comprehensive emergency obstetric and newborn care when necessary, clear, rapid referral links should be established between certified BEmONC facilities and their CEmONC counterparts throughout Indonesia. Every effort should be made to ensure that every facility designated as a CEmONC facility has the necessary staff and equipment and that it is fully prepared to provide the required emergency obstetric and neonatal care 24 hours a day, seven days a week.
Meanwhile, some notable efforts are already under way. One is a program to upgrade the status of local health clinics (puskesmas) to BEmONC status. Another is a program of maternal health insurance (Jampersal). Among other things, it provides for universal free delivery services at public clinics and hospitals. These kinds of insurance schemes have been shown internationally to be instrumental in making it easier for women to give birth in health institutions rather than at home. In this regard, the latest (2012) Indonesia Demographic and Health Survey has revealed a marked increase in births occurring in health institutions. Although it may be too early to draw a definitive link between this and the current efforts, these findings could auger well for the future.
Even while actions are being taken toward achieving this goal, community health centers and midwives will continue to serve their patients. Thus the transition will be gradual and will occur only as CEmONC hospitals and BEmONC clinics, supported by effective health insurance programs, become available and gain the confidence of families. Meanwhile, ongoing mass efforts to train midwives for home delivery will create a larger number of midwives who are unable to treat the most common causes of maternal mortality and may eventually be displaced. Actions to upgrade hospitals and facilities should not be delayed because it is unlikely that further mortality reduction will be made before more women are able to give birth in accredited facilities.
The interventions needed to reduce maternal, fetal, and neonatal mortality are well known. But, with a few important exceptions, most complications leading to maternal and fetal deaths during childbirth cannot be predicted. Thus medical interventions for these complications must be readily available for all women and newborns. Because most maternal, fetal, and newborn deaths occur during labor, delivery, and the immediate postpartum period, the interventions for any complications that arise need to be readily available during labor and after delivery. Few of these interventions are available in the home, and, unfortunately, in most low-income countries many of these interventions are not available in clinics. For these reasons, this committee advocates a systems approach based on facilities with CEmONC capabilities in which all the required signal functions are available for all births.
The following recommendations deal with the major components of this system:
Recommendation 2—Strategies and Plans. Indonesia’s districts vary widely in population density, personal wealth, and numbers of doctors and other health workers, as well as in unalterable features such as isolation, island geography, and transport difficulties. Within the resources of Indonesia, and probably those of any country, it is inevitable that technology and services-centered indicators such as maternal mortality ratio and neonatal mortality rates will vary. Technical strategies, implementation plans, and a road map to achieving high coverage with quality maternal and neonatal health (MNH) services should be developed across a range of Indonesia’s diverse environments—all reflecting local conditions and feasible approaches over 5-, 10-, and 20-year time horizons.
Those responsible for creating an Indonesian plan to reduce maternal, fetal, and neonatal mortality should seek to reduce the disparities in pregnancy-related services and outcomes between east and west, rich and poor, and rural and urban Indonesia. A one-size-fits-all plan will not sufficiently address the significant regional and socioeconomic differences in the country. Indeed, several different strategies may be required. In allocating resources, choices will inevitably have to be made. However, these choices should be made openly, with local participation, and based on balancing the objectives of service equity and total lives saved.
The Indonesian health system was designed to be community-based. However, this approach has not reduced mortality at childbirth. At present, Indonesia has an overabundance of incompletely trained midwives who offer delivery services at home. They now face increased competition for their services and difficulty in accessing employment opportunities within the system. Because most complications cannot be reliably predicted before or during labor, they all too often go unrecognized in the home until it is too late. It follows, then, that the most promising future strategy for reducing the maternal mortality ratio (MMR) and neonatal mortality rate (NMR) in Indonesia is to ensure that most if not all deliveries are in BEmONC or CEmONC facilities. In principle, these facilities can be found in every district, and efforts must be directed at strengthening their capabilities and providing access, including transport, to all women.
Recommendation 3—System of Care. Organization of this system of care and standardization of the training and licensing of providers should be centralized under one Indonesian government agency that is represented within the central government and at the provincial, district or municipality, and community levels. Such a system should address population-based planning and the implementation of all services involved in childbirth.
Creation of a systems approach to obstetric and neonatal care would be greatly facilitated if there were greater coordination among organizations charged with training and licensing care providers, operating health care facilities, setting standards for quality of care, approving birth plans, and collecting data for monitoring performance.
Recommendation 4—Training. The system for training skilled birth attendants should be revised to include the training of physicians and nurses who specialize in emergency obstetric, neonatal, and
anesthesia services. The training of midwives should be strengthened to emphasize recognition of obstetric emergencies and the design and implementation of birth plans that include a planned birth at a BEmONC or CEmONC facility and early referral for complications.
WHO and other organizations recommend the use of skilled birth attendants for delivery. However, the training and skills of those labeled “skilled attendants” in Indonesia vary widely, and many cannot perform many interventions that are often necessary to save a life. Even the ability of the most skilled attendant to save a life is limited if no blood or antibiotics are available, or if the facilities needed for a cesarean section are not accessible. As an interim step, all midwives should be given training in neonatal resuscitation, and simple resuscitation equipment should be a standard item in the midwifery kit. In the end, however, the continued emphasis on training large numbers of birth attendants who cannot provide emergency obstetric and neonatal services and who attend deliveries at home or in non-BEmONC and non-CEmONC facilities will not result in substantial further reductions in maternal mortality in Indonesia.
Recommendation 5—Financing. Sufficient and effective financing mechanisms for obstetric and newborn services should be established under the supervision of the government of Indonesia to ensure universal access to quality MNH care and strengthen the organization of that care. These mechanisms should include sufficient overall funding for the needed services; the appropriate incentives for women to deliver at a certified facility capable of providing optimal care and for providers to reduce unnecessary or inappropriate care; and effective monitoring and accountability mechanisms to plan and track financing, including some mechanisms for peoples’ participation.
The government of Indonesia has made significant efforts in recent years to increase its spending on health and to focus more of that spending on the poor and on priority health needs. However, without greater efforts to reform reporting and accountability, the government will continue to find it difficult to identify and remedy the bottlenecks hindering the translation of increased spending into better birth outcomes.
Recommendation 6—Data. Making the appropriate decisions about programs likely to reduce maternal, fetal, and neonatal mortality rates in Indonesia and allocating resources appropriately will require the collection of high-quality data on the relevant rates and causes of morality. These data must be routinely collected nationwide in a standardized manner and used routinely and frequently at the local level for improving the quality of programs and at the national level for planning programs and allocating resources.
Presently, Indonesia relies on data from periodic household surveys to obtain official measures of maternal-, neonatal-, and child health–related deaths. However, these surveys may lack the necessary sample sizes or the most reliable methods to determine mortality rates and must therefore rely on estimations and model-based adjustments that yield a wide range of values. Thus these surveys do not provide the precision and reliability needed to guide decisions optimally at the national and district levels. Nor can they provide the timely guidance needed to achieve rapid progress in programs and policies.
Ultimately, a comprehensive vital registration system, including registration of maternal and neonatal deaths, is the only way to fully provide the kinds of statistically valid data required for planning and monitoring, particularly at the regional and local levels. Initially, strong support should be offered to efforts such as the Indonesia Mortality Registration Strengthening Project under the Ministry of Health. This support might take the form of follow-up and training activities directed at local governments to build understanding and capacity. In the end, a broader commitment to and cooperation in building the system will be required from a range of key stakeholders, including the Ministry of Health, Ministry of Home Affairs, Ministry of Finance, Ministry of State Apparatus Utilization and Bureaucratic Reform, and Statistics Indonesia.
Recommendation 7—Community Involvement. Health care at the community level has many strengths, including the women's volunteer committees, posyandus (integrated health posts), and the
district health departments. These organizations could make a greater contribution to lowering maternal and infant mortality by supporting the creation and execution of certified birth plans. Such plans would include facilitating transportation and utilizing antenatal care services, immunization, and contraception optimization for all women and children and, most important, planning delivery in a certified BEmONC or CEmONC facility. These volunteer committees should be integrated into an overall system designed for the provision of services for women and newborns.
Maternal health insurance that provides for free or very low-cost birthing service at health facilities (particularly those with BEmONC or CEmONC capacity) should be encouraged and strengthened to support these community-based initiatives.
Recommendation 8—Education and Empowerment. Programs that encourage the education and empowerment of girls and young women in general and relating to childbirth issues in particular should be strengthened. This high-priority initiative must include clearly measurable and frequently assessed indicators of progress, with additional resources provided whenever needed. In addition, a reinvigorated family planning program should constitute an integral part of safe childbirth services offered to Indonesian women to reduce the number of high-risk pregnancies associated with unwanted births, the number of underage and overage mothers, and very short and very long birth intervals.
In Indonesia, over 60 percent of the women who die in childbirth have not had the benefit of a primary education. The desires, knowledge, and decision-making power of the mother are critical factors in family decisions regarding safe childbirth. Educating girls and young women saves lives by enabling a mother to make better decisions, by empowering her in her dealings with other family decision makers, and by giving her the knowledge she needs to effectively nourish and care for her infant.
The committee believes that the recommendations made offer Indonesia the greatest likelihood of practical success. The world is divided between countries with an MMR under 20 per 100,000 live births and those with an MMR over 200. Some developing countries are in the former group, such as Sri Lanka and Malaysia. What these countries have in common is that they have adopted a program that corresponds to recommendation 1: enabling all women to give birth in suitable facilities (CEmONC or BEmONC). Most of the latter group of countries, not unlike Indonesia, promote home births attended by midwives or traditional birth attendants. This approach has thus far failed to maintain the expected rate of improvement in Indonesia. Accordingly, this committee believes that Indonesia is likely to be best served by investing in the future in facility-based services, with the improvements suggested in the recommendations.