Please note: This archive was last updated in 2005.

RHO archives : Topics : Safe Motherhood

Key Issues

This section provides brief summaries of some major research areas related to safe motherhood, particularly in low-resource settings. More detailed discussions of specific key issues are included in the Annotated Bibliography.

Be sure to use the Glossary if you are unfamiliar with any of the terms on this page.

Causes of high maternal mortality

Research on mortality among women of reproductive age has confirmed the importance of maternal mortality in developing countries (Fauveau et al. 1989). Studies have investigated the causes and timing of maternal deaths (Pradhan et al. 2002; Dafallah et al. 2003; Li et al. 1996) as well as the barriers to timely and appropriate care that increase the likelihood of a woman dying after she develops an obstetric complication (Chiwuzie et al. 1995; Le Bacq and Rietsema 1997; Castro et al. 2000). A comparison of maternal deaths at Zambian and American referral hospitals found both sites had high maternal mortality from preventable causes (Kilpatrick et al. 2002). A retrospective, cross-sectional study in Uruguay found that women with short interpregnancy intervals (five months or less) are at increased risk of maternal death and pregnancy complications (Conde-Agudelo and Belizán 2000). A hospital-based study in West Africa confirmed hypertensive disorders and postpartum hemorrhage as key causes of maternal deaths (Thonneau et al. 2004).

Women whose health already is compromised by poor nutrition and disease are more likely to die during an obstetric emergency. Many studies have linked nutrition with two of the main causes of maternal deaths, hemorrhage and obstructed labor (Rush 2000). Severe anemia—often from iron and folate deficiency, malaria, hookworm, or HIV—is an important indirect cause of maternal mortality in many areas, yet current programs of iron supplementation for pregnant women are unlikely to be effective (Shulman 1999; Rush 2000). Malaria itself is a common indirect cause of maternal death, both during pregnancy and postpartum (Cot and Deloren 2003; Olsen et al. 2002). The full impact of HIV/AIDS on maternal deaths is difficult to determine, but HIV infection has a major impact on both direct and indirect causes of maternal deaths. Appropriate antiretroviral therapy during pregnancy and increased HIV prevention are important interventions that need to be included in safe motherhood programs (McIntyre 2003; Bicego et al. 2002).

In addition to medical and hospital factors, community-based or sociocultural factors such as attitudes and practices also influence maternal mortality (Stekelenburg et al. 2004; Kyomuhendo 2003; Griffiths and Stephenson 2001; Okolocha et al. 1998). Violence against pregnant women is increasingly recognized as a significant factor contributing to maternal deaths (Granja et al. 2002; McFarlane et al. 2002; Haque and Clarke 2002). As many as one in four women is physically or sexually abused during pregnancy, usually by an intimate partner (Heise et al. 2002). (For more information about the reproductive health effects of violence against women, please see RHO's Gender and Sexual Health section.)

Top of page

Preventing maternal mortality

Researchers have analyzed what kinds of interventions can best reduce maternal mortality (Bulatao and Ross 2003; Shiffman 2000; McCarthy 1997; Donnay 2000). Efforts are being made to close the gap between research and practice by advocating an evidence-based approach to decision-making in the care of pregnant women (Miller et al. 2003; Global Health Council 2002). Appropriate technological innovations—equipment, supplies, procedures, and techniques—can be used to reduce maternal mortality in low-resource settings (Tsu 2004a; Tsu et al. 2003; Tsu and Free 2002). Discussions and papers from a July 2003 workshop in Bellagio, Italy, on new and underutilized technologies to reduce maternal mortality indicate many existing technologies show promise in reducing maternal deaths (Tsu 2004b).

There now is international consensus that the following factors are important to reducing maternal mortality:

  • Reducing gender inequality.
  • Improving women's nutritional status.
  • Increasing girls' access to education.
  • Providing professional care at delivery.;
  • Building effective referral systems.;
  • Ensuring maternal care for all women.
  • Developing district-level planning and community participation.
  • Using process indicators to evaluate progress (UNICEF 1999).

Family planning can reduce maternal mortality and morbidity by reducing the total number of pregnancies, the proportion of high-risk births, and the proportion of pregnancies that end in unsafe abortions (USAID 1996; FHI 1995). One global study estimates that 21 percent of maternal deaths over a six-year period were due to unintended pregnancies (Daulaire et al. 2002). By enabling women to plan, space, and prevent pregnancies, family planning could prevent thousands of maternal deaths. Data also show that spacing births three to five years apart not only improves child survival, but also can save mothers lives (Setty-Venugopal and Upadhyay 2002).

The importance of adequate and accessible obstetric services at the community, health center, and hospital levels has become clear (Kwast 1995). While having a skilled attendant at delivery is associated with reduced maternal mortality, it has not been proven whether or not this association is causal or is confounded by other factors (Sloan et al. 2002). Where appropriately trained providers, equipment and drugs are available, many deaths due to postpartum hemorrhage can be prevented through the active management of the third stage of labor (Miller et al. 2004; McCormick et al. 2002; Geelhoed et al. 2002; PATH 2001). There is significant variation between and within countries in the practice of active management (Festin et al. 2003).

One study estimates that antenatal care and community-based interventions can prevent 26 percent of maternal deaths, and another 48 percent can be avoided by ensuring access to quality essential obstetric care (Jowett 2000). A study in Kenya found that offering free insecticide-treated bednets to pregnant women through antenatal care clinics is a cost-effective, equitable way to distribute this malaria prevention tool (Guyatt et al. 2002). Providing good care to women who have HIV/AIDS before, during, and after pregnancy can help reduce the spread of HIV, especially between mothers and infants (Rosser 2000).

New studies indicate that nutritional supplementation could reduce maternal morbidity and mortality (Villar et al. 2003; Tomkins 2001), especially among women with early or closely spaced pregnancies (King 2003). A large-scale study in Nepal found that supplementing reproductive-age women with vitamin A or beta-carotene reduced maternal mortality by 40 percent (West et al. 1999). Although iron supplementation during pregnancy can prevent a drop in iron storage associated with pregnancy, there is no single, universally acceptable approach to iron supplementation of pregnant women that provides a clear benefit to mothers and their children (Allen 2000; Beard 2000). Formative research in eight countries identified several common obstacles, as well as many facilitating factors, to iron supplementation programs. The findings have been used to improve compliance (Galloway et al. 2002).

Policy changes, appropriate allocation of resources, and community mobilization also play important roles in preventing maternal deaths (Koblinsky et al. 1999). Health sector reforms in China, which emphasize cost recovery, have had a direct negative impact on maternal health care (Bogg et al. 2002). A spreadsheet developed by the World Health Organization and the World Bank helps program planners and managers analyze and understand the costs associated with various maternal health services and interventions (Lissner and Weissman 1998). Professional obstetrical and midwifery associations have the responsibility and ability to reduce maternal mortality in their own countries and abroad (Chamberlain et al. 2003). Making better use of private-sector providers, especially nurses and midwives, in low- and middle-income countries could contribute to safe motherhood goals (Brugha and Pritze-Aliassime 2003).

Top of page

Human rights approach to safe motherhood

Safe motherhood is recognized as a basic human right, protected by a range of international human rights treaties and laws (WHO 1998; Cook et al. 2001). These treaties obligate governments to address the causes of poor maternal health through their political, health, and legal systems, and require that signatory nations report on compliance with treaty goals (Starrs 1997; Murphy and Ringheim 2001; Center for Reproductive Rights 2003). It has been proposed that compliance be measured by using the guidelines issued in 1997 by WHO, UNICEF, and UNFPA (Yamin and Maine 1999).

Approaching safe motherhood from a human-rights perspective emphasizes that women have the right to receive comprehensive reproductive health care, including family planning, education, nutrition, and basic health services (Yanda et al. 2003; Liljestrand and Gryboski 2001). Ensuring women's rights to life, liberty, and security of the person, health, maternity protection, and non-discrimination would facilitate safe motherhood (Cook and Dickens 2002). Human-rights principles can be incorporated into programs at the clinical, facility-management, and national-policy levels (Freedman 2001). A human-rights approach to addressing maternal mortality also means focusing strategically on power dynamics within the health care system (Freedman 2002).

Womens low status relative to men in society contributes to maternal mortality. A cross-national study of 79 developing countries found that womens status is a strong predictor of maternal mortality (Shen et al. 1999). A study of womens autonomy and use of maternal health care services in Uttar Pradesh, India, found that women with greater freedom of movement obtained more antenatal care and were more likely to use safe-delivery care (Bloom et al. 2001). In Afghanistan, high rates of maternal mortality may be an indicator of violations of womens human rights (Amowitz et al. 2002).

Governments can promote a human rights approach to safe motherhood by:

  • Working to eliminate discrimination against women, including violence against women and harmful practices affecting women's health.
  • Ensuring appropriate antenatal, delivery, and postpartum care to all women.
  • Repealing laws that prohibit health care procedures that only women need, including safe abortion procedures.
  • Ending discriminatory practices in public health care facilities, including requirements for spousal consent prior to accessing services.
  • Ensuring access to family planning services and information to all women and adolescents of reproductive age (for more information, see the Center for Reproductive Rights).

Top of page

Community mobilization

While interventions at many levels are needed to reduce maternal mortality and morbidity, experiences have shown that much can and needs to be accomplished at the community level. By working with the community, individuals and groups can be empowered to identify problems and derive solutions that work in the local context (Howard-Grabman et al. 1993). In Tanzania, communities have been empowered to develop community-based plans for emergency transportation of obstetric patients (Ahluwalia et al. 2003; Schmid et al. 2001). Another approach, establishing community loan funds to pay for emergency transportation, had limited success in Sierra Leone (Fofana et al. 1997).

Communication strategies that involve the community and make use of community-based tools and methodologies can be effective in promoting safer motherhood (Russell and Levitt-Dayal 2003; Moore 2000). In Nepal, a coalition of groups formed partnerships, and used innovative community mobilization and communication processes to convey reproductive health messages to people in remote areas (Levitt et al. 1997). Training community motivators or resource persons to increase use of health facilities for emergency obstetric care has had a positive impact, but the associated costs and need for continued supervision are substantial (Kandeh, et al. 1997; Nwakoby, et al. 1997). The "Home-Based Life Saving Skills" program is a community and competency-based program to educate, motivate, and mobilize pregnant women, families, and communities to reduce maternal and neonatal mortality (Sibley et al. 2004; Sibley et al. 2001). Having women keep their own home-based maternal records has been useful in identifying and referring women with increased maternal risks, improving reproductive health monitoring, and increasing the participation of the mother, her family and the community in their own health care (WHO 1994).

Top of page

Measuring maternal mortality and program impact

The difficulty of collecting reliable data on maternal mortality and the debate over which indicators to use are challenges for safe motherhood efforts (Graham and Airey 1987; Leete 1998; Stanton et al. 2000). Limiting the definition of maternal mortality to maternal deaths within 42 days of the end of a pregnancy has also been questioned (Høj et al. 2003). Necessary data on maternal mortality are lacking because of inadequacies in technology, leadership and resources (Graham 2002). Measuring and estimating maternal mortality in the context of HIV/AIDS is even more difficult due to the effect of HIV/AIDS on the risk of maternal mortality and on measuring maternal mortality (Graham and Hussein 2003). Measurement and evaluation issues are important because they influence program priorities and assessments of program performance (Milne et al. 2004). However, having valid numbers is not sufficient; it is also important to understand the underlying factors that contribute to the deaths (WHO 2004).

An analytical study of United Nations data found that attendance at birth by trained personnel correlated with lower maternal mortality rates (Robinson and Wharrad 2001), yet no causative relationship has been established. Skilled attendance at birth and rate of cesarean deliveries have also been used to assess trends in maternal mortality (AbouZahr and Wardlaw 2001). It is also possible to measure maternal mortality using a national census (Stanton et al. 2001).

Researchers continue to seek alternatives to maternal mortality to assess program impact over a short period of time (3 to 5 years) (Akalin et al. 1997; Graham et al. 1996; Ronsmans et al. 1997). The Maternal and Neonatal Programme Effort Index (MNPI) rates various components of national programs to reduce maternal and neonatal mortality (Ross et al. 2001). The index offers a tool for diagnosing and managing programs, and could allow comparisons between countries and regions over time. "Verbal autopsy" has also been used to identify and quantify maternal deaths in many places where health records and death certificates are unavailable, but its accuracy for determining medical causes of death is limited (Kodio et al. 2002; Sloan et al. 2001; Hoj et al. 1999). The "sisterhood" method of determining maternal mortality often provides data that are more comprehensive than facility-based records (Lech and Zwane 2002; Danel et al. 1996; Olsen et al. 2000; Font et al. 2000). Community-based studies of maternal deaths are costly and labor-intensive, but yield useful data for prioritizing program interventions and for evaluating existing monitoring systems (Mswia et al. 2003; Songane and Bergstrom 2001; Kilonzo et al. 2001). Data from household surveys has been used to link maternal death with poverty at the subnational level (Graham et al. 2004).

The use of process indicators and relevant proxies, such as maternal morbidity surveys (Filippi et al. 2000; Prual et al. 2000; Khosla et al. 2000), to target and assess programs designed to improve maternal health have also been suggested (Wardlaw and Maine 1999). Among the set of process indicators proposed are indicators of unmet need for family planning and obstetric services; utilization, coverage, and access; and quality of care (John Snow 1999; Leete 1998). Researchers in Zambia assessed the effectiveness of the existing district maternity-referral systems, using a mix of tools and relying primarily on routinely collected information (Murray et al. 2001). Self-reporting of reproductive outcomes and use of maternal health care services through health interviews with rural women in Zimbabwe offers another tool for assessing maternal health and the use of health care resources (Nilses et al. 2002).

In 1997, UNICEF, WHO, and UNFPA proposed using (1) the number of facilities providing emergency obstetric care, (2) the proportion of all women who deliver at emergency obstetric facilities, (3) the proportion of women with complications who deliver at emergency obstetric facilities, (4) cesarean deliveries as a proportion of all births, and (5) the case-fatality rate to measure and monitor the availability, use, and quality of emergency obstetric care (Maine et al. 1997). International consensus on the potential usefulness and drawbacks of these process indicators is emerging (Goodburn 2002; Gichangi et al. 2001; Ronsmans et al. 2002; Pathak et al. 2000; Ronsmans et al. 1999). The indicators have been used to perform needs assessments in several countries (Bailey and Paxton 2002; AMDD Working Group on Indicators, June2002; AMDD, September 2002; AMDD 2003). Experience with the first large-scale implementation and use of these "U.N. Indicators" in Malawi indicates that developing the necessary routine data-collection systems requires a great deal of focused attention and effort (Goodburn et al. 2001). Nonetheless, the monitoring system provides data that are immediately useful to program managers, service providers, and policy makers (Hussein et al. 2001). The UN Indicators also are being put to use in a collaborative effort by UNFPA and the governments of Cameroon, Côte d'Ivoire, Mauritania, Niger, and Senegal. The data collected are being used to develop key interventions to improve essential obstetric care in each country (UNFPA 2003). In a study in rural China, researchers propose complementary use of the UN Indicators along with reproductive health indicators developed through a community-based process to better reflect the local situation (Wang et al. 2003).

Top of page

Antenatal care and risk assessment

The role of antenatal care is being increasingly questioned, particularly in resource-poor settings (Prual et al. 2000; Vanneste et al. 2000; Amooti-Kaguna 2000). Some experts have questioned the actual impact on mortality of many elements of antenatal care (McDonagh 1996; Villar and Bergsj 1997; Jahn et al. 1998), although few would argue with the benefits of regular care—including counseling on nutrition and screening for certain disorders, such as syphilis (Gloyd et al. 2001)—on women's and infant's health. A new approach to antenatal care suggests making use of existing family and community support systems, and developing partnerships beyond the health services to help women prepare for safe pregnancies and deliveries (Gerein et al. 2003).

Researchers are also studying the factors associated with use of antenatal care (WHO and UNICEF 2003; Magadi et al. 2000). A study in rural Nepal found that the quality of services at health posts and the presence of village outreach workers, not the quantity of facilities, were the strongest determinants of use of antenatal care services (Acharya and Cleland 2001). Similarly, a study in Cape Town, South Africa, found that womens attendance at antenatal care is influenced by their perceptions of the quality of care, and the perceived benefits and risks of antenatal care. This study showed the need to better address womens informational needs and to improve their interactions with providers (Abrahams et al. 2001). In Mozambique, women’s fear of witchcraft encourages them to hide pregnancies and delay prenatal care to protect themselves and their children (Chapman 2003). In rural Nigeria, women do not seek prenatal care because of the lack of financial resources, God’s will, and husband’s denial (Adamu and Salihu 2002).

Screening women for risk factors has been recommended as a strategy to reduce maternal mortality, but recent studies have cast doubt on its ability to predict which women will develop complications (Majoko et al. 2002; Yuster 1995). A prospective study in Guinea-Bissau found risk-factor screening would do little to reduce maternal mortality (Høj et al. 2002). Other studies have tried to identify which risk indicators—for example obstetric history, age, parity, and height—are most reliable (Fortney 1995; Ujah et al. 1999). A study in Nigeria found that certain risk factors were associated with anemia among pregnant women (Oboro et al. 2002). Although risk screening can identify groups of women who are more likely to have an obstetric complication, these risk factors do not predict which individuals actually develop complications. The majority of "high-risk" women deliver without problem, while most women who do develop problems are from the "low-risk" group.

A WHO systematic review of randomized controlled trials of routine antenatal care in several countries found that a lower number of antenatal visits has little effect on the risk to mothers and babies, and would reduce costs (Carroli et al. 2001; Munjanja et al. 1996). Women and providers generally accept the new antenatal care model, but they may need to be reassured about the safety of fewer, more widely spaced visits for those with no complications (Langer et al. 2002). A WHO study on antenatal care in Argentina, Cuba, Saudi Arabia, and Thailand found considerable variability of antenatal care services among selected clinics in the study countries (Piaggio et al. 1998). A study in India found that antenatal care utilization is an important determinant for safe delivery care (Bloom et al. 1999). Antenatal care can contribute to reducing the complications of hypertensive disorders of pregnancy (for example, eclampsia) among developing-country women. Antenatal care is also used to deliver intermittent preventive treatment of malaria, but a study in Malawi found poor coverage despite high antenatal attendance (Holtz et al. 2003).

Top of page

Skilled attendance during pregnancy and delivery

About half of all births in developing countries are attended by a skilled health care provider (see Figure 2).

Many studies have shown the value of medically trained midwives in providing adequate and accessible obstetric care to women in developing countries (Peters 1995; de Bernis et al. 2000). Studies have tested whether midwives at the village level can improve the referral system and reduce maternal morbidity and mortality (Fauveau et al. 1991; Senah et al. 1997). WHO has developed a series of training modules to teach midwives about the concepts of safe motherhood and to upgrade their clinical skills (O'Heir 1997). The American College of Nurse-Midwives Life Saving Skills Training Program also has been recognized as an important step in enhancing midwives' role in safe motherhood programs (Buffington et al. 1998). However, studies also show the need for skilled midwives to work in collaboration with emergency obstetric care to improve maternal health (Ronsmans et al. 2001).

The role of traditional midwives (TBAs)—medically unskilled birth attendants—in improving maternal health has also been studied in many communities (Ray and Salihu 2004; Minden and Levitt 1996; Bolam et al. 1999). Researchers have examined the impact of training on TBAs' obstetric knowledge and practices, as well as on maternal mortality and morbidity (Goldman and Glei 2003; Goodburn et al. 2000; Hoff 1997; Kamal 1998). The impact of training TBAs on maternal mortality appears limited and the greatest benefit may be improved referral and linkages with the formal health system (Bergström and Goodburn 2001). Results from a meta-analysis suggests TBA training may increase antenatal attendance rates (Sibley, Sipe and Koblinsky 2004). In Zambia, traditional birth assistants serve as culturally knowledgeable, social support women during labor and delivery, but have little accurate knowledge of appropriate management of labor and delivery (Maimbolwa et al. 2003). A study in rural Bangladesh suggests that improving the timely referral of complicated deliveries to medical facilities or to trained TBAs could reduce maternal morbidity and mortality (Paul and Rumsey 2002). Programs are testing alternative instructional methods (Matthew et al. 1995) and curricula (Sibley and Armbruster 1997) for training TBAs. With appropriate training and supervision, TBAs may be helpful in preventing transmission of HIV (Bulterys et al. 2002). Continuous one-to-one support of women during labor has been shown to improve satisfaction and outcome (Hodnett et al. 2004).

While assistance at delivery is associated with lower maternal mortality, research has shown that the presence of a skilled attendant is not necessarily sufficient to reduce the likelihood of maternal death or disability (Graham et al. 2001). Not all health care providers have the necessary midwifery skills to be considered skilled birth attendants. Comprehensive definitions and lists of skills have now been developed to define a skilled attendant (Safe Motherhood IAG 2003). It is also now recognized that the context within which a provider works has a large effect on the outcome of the delivery (Maclean 2003). The broader concept of “skilled attendance” has been developed to describe the presence of a skilled provider within a supportive environment. It is defined as: “…the process by which a woman is provided with adequate care during labour, delivery, the postpartum and immediate newborn periods. In order for this process to take place, the attendant must have the necessary skills and must be supported by an enabling environment which includes adequate supplies, equipment and infrastructure, as well as an efficient and effective system of communication and referral/transport” (Safe Motherhood IAG 2003).

Skilled attendance depends on a partnership of skilled attendants, an enabling environment, and access to emergency obstetric care services. As a result of this broader emphasis on skills and context, safe motherhood programs are shifting away from focusing on training skilled birth attendants to ensuring skilled attendance at delivery (de Bernis et al. 2003; Bell et al. 2003; Voet 2003; WHO 2002).

Top of page

Essential obstetric care

Most life-threatening obstetric complications cannot be predicted or prevented, but they can be treated (Maine and Rosenfield 2001). Research by the Prevention of Maternal Mortality project drew attention to the importance of essential obstetric care, particularly emergency care, in reducing maternal mortality (Prevention of Maternal Mortality Network 1996; Maine 1997; Post 1997; Maine and Rosenfield 1999). The "three-delays" model outlines the factors that delay (1) the decision to seek emergency obstetric care, (2) a woman's arrival at a health care facility, and (3) her treatment at that facility (Thaddeus and Maine 1994). The facility must have the necessary personnel, equipment, and drugs to provide emergency obstetric care, but the demand for these resources, their costs, and cost-effectiveness can be estimated in advance (Desai 2003; El Tahir et al. 1995). Despite the perception that hospital services are costly, a study in Bangladesh shows that basic services, including emergency obstetric care, can be made available with low overall increases in cost and large benefits for maternal health (McCord and Chowdhury 2003).

Studies have found that access to essential obstetric services is limited in developing countries (Nirupan and Yuster 1995; Barnes-Josiah et al. 1998) and that building a functional chain of referral to obstetric services can reduce maternal mortality (Maine et al. 1996; Figa-Talamanca 1996). One such system is provided by maternity waiting homes that are linked to facilities offering essential obstetric care. Women who are at increased obstetric risk, but who do not have access to essential obstetric services, are referred to maternity waiting homes a few weeks prior to delivery. Once a woman is in labor or in need of care, she is transferred to the hospital (WHO 1996).

Having access to emergency obstetric care does not ensure that patients receive quality care (Miller et al. 2003). Many projects have increased the use of essential obstetrical services and reduced case fatality rates by improving the quality of care at referral facilities (Gill and Ahmed 2004; Djan et al. 1997; Ifenne et al. 1997; Leigh et al. 1997; Mbaruku and Bergstrom 1995). This includes low-cost improvements in the design and functioning of operating rooms (Abreu and Potter 2001). Training to improve essential obstetric care skills of providers is key to providing quality obstetric care, and new training methods and resource manuals have been developed to address this need (Penny and Murray 2000; WHO 2000). Changing policies to enable existing medical staff to address obstetric emergencies may also reduce mortality in rural areas (Mavalankar 2002). Maternal mortality/morbidity review meetings, where each case of maternal death and/or severe obstetric morbidity and contributing factors are discussed among hospital staff, can be useful tools for identifying problems and improving essential obstetric care services at the district level (Gohou et al. 2004; Stekelenburg and Roosmalen 2002). Criterion-based clinical audits can also be useful tools for measuring and improving the quality of obstetric care in developing countries (Filippi et al. 2003; Weeks et al. 2003; Wagaarachchi et al. 2001).

Experience with the family-centered maternity care approach in the Ukraine, Moldova, and Russia showed that maternity care systems in these countries can be improved and lead to changes in policies and practices that offer better maternity care options for women and their families (John Snow, Inc. 1998). A study in rural India found that patients have learned to determine when hospital care is needed, resulting in the effective and efficient delivery of emergency obstetric services. The cost of these services, while low on a per-capita basis, is a substantial barrier for many poor families (McCord et al. 2001). Cost of services is also a barrier to use of hospital-based delivery care in West Africa (Borghi et al. 2003). Several initiatives to increase the availability and utilization of emergency obstetric care services are being implemented through partnerships between associations of obstetricians and gynecologists in more and less developed countries (Jamisse et al. 2004; Benagiano and Thomas 2003b); Mekbib et al. 2003; Benagiano and Thomas 2003a; Lalonde et al. 2003; Curet et al. 2003).

Top of page

Postabortion care

Deaths due to abortion-related complications are preventable if they are identified early and treated appropriately. However, many women suffering from abortion complications delay in seeking care, resulting in many maternal deaths (Goyaux et al. 2001). Although 13 percent of maternal deaths worldwide are estimated to be due to complications from unsafe abortions, this figure undoubtedly underestimates the total in places where abortion is illegal or very restricted (Thonneau et al. 2002). Postabortion care (PAC) has been endorsed as an effective strategy to reduce maternal mortality caused by unsafe abortions. A PAC approach involves three integrated components:

  • Emergency treatment of medical complications of spontaneous or induced abortions.
  • Postabortion family planning counseling, services, and referral to prevent future unintended pregnancies and repeat abortions.
  • links to other reproductive health services, such as diagnosis and treatment of reproductive tract infections (Greenslade et al. 1994).

A summary of operations research on PAC indicates the most effective intervention strategies are: improving clinical care; providing family planning counseling and services; expanding access to PAC; planning comprehensive PAC services; and involving male partners (Huntington 2001). Experiences in Ghana and Kenya showed that training nonphysician providers (such as midwives) in the provision of PAC is a feasible and acceptable strategy for decentralizing PAC services to the community level, and has resulted in increased access to postabortion family planning and other reproductive health services (Yumkella 2000; Baird et al. 2000; Billings 1998). In Ethiopia, involvement of the private sector can substantially improve access to quality postabortion care (Yeneneh et al. 2004).

A prospective study in Zimbabwe found that providing contraceptive services to women hospitalized for incomplete abortions reduced subsequent unplanned pregnancies (Johnson et al. 2002). In Nepal, a model PAC service and training program for the management of incomplete abortions showed that manual vacuum aspiration is a safe, effective procedure that can increase access to family planning and other reproductive health services (Ghosh et al. 1999). A comparison of models of postabortion care in Mexico City showed that women preferred comprehensive PAC services to the standard model (Billings et al. 2003). The Egyptian Postabortion Care Initiative’s move from small-scale research to a national program provides many lessons for use in other settings (Huntington and Nawar 2003). Overall, successful development and implementation of PAC services requires careful planning and coordination among the different parties and health care providers involved (Ghosh et al. 1999).

Top of page

Use of misoprostol for obstetric and gynecologic indications

With support from the Packard Foundation, PATH and EngenderHealth performed a qualitative study on ob/gyn uses of misoprostol. The summary below reflects some of the outcomes of this work.

Misoprostol is a prostaglandin E1 analog indicated for the prevention and treatment of gastric and duodenal ulcers. Closely related to other prostaglandins, misoprostol is also used as an adjunct to medical abortions performed with methotrexate or mifepristone. Misoprostol-only regimens have shown promise for obstetric and gynecological indications including postabortion care, treatment of miscarriage, cervical priming, induction of labor , and prevention and management of postpartum hemorrhage (Goldberg et al. 2001; Gulmezoglu et al. 2001). Among the key advantages of misoprostol for these indications are its effectiveness, low cost, stability (tablets have a shelf life of several years at room temperature), accessibility, ease of administration (oral, vaginal, rectal), and potential to lead to safer reproductive health outcomes than currently used therapies (Blanchard et al. 2000). Despite its appeal, a study of misoprostol use in Brazil, the United States, and Jamaica indicates there is considerable variation in the regimens used, which often differ from those recommended in medical literature (Clark et al. 2002).

Research continues into the use of misoprostol during the third stage of labor (after delivery of the infant and prior to delivery of the placenta) to prevent postpartum hemorrhage. A WHO multi-center, randomized trial of misoprostol in the third stage of labor found that oxytocin is preferable to oral misoprostol in the active management of the third stage of labor in hospital settings where active management is the norm (Gulmezoglu et al. 2001). However, it may be useful to help prevent deaths from postpartum hemorrhage in the absence of other interventions (Joy et al. 2003; El-Refaey and Rodeck 2003; Darney 2001). Studies in Zimbabwe and Nigeria have shown that oral misoprostol is as effective as intramuscular oxytocin in preventing postpartum blood loss (Oboro and Tabowei 2003; Kundodyiwa et al. 2001). Rectal misoprostol shows promise as a useful first-line drug for the treatment of primary postpartum hemorrhage (Mousa and Alfirevic 2003).

Interest in misoprostol and its potential to reduce the complications of unsafe abortion and other gynecological and obstetric conditions has been demonstrated by researchers in Asia, Africa, Europe, Latin America, the Middle East, and North America (PATH/EngenderHealth 2001). Available studies, however, have shown great variations in dosing guidelines, populations studied, evaluation criteria, and outcomes. When used during the first trimester, misoprostol-only regimens administered vaginally can result in complete or partial abortions in 5 to 94 percent of cases (Blanchard et al. 1999). Because the uterus becomes increasingly sensitive to misoprostol as gestation advances, slightly higher effectiveness rates have been observed with lower doses administered in the second trimester. When used during spontaneous abortion or cervical priming, misoprostol may enable clinicians to avoid mechanical procedures. Misoprostol's side effects are common but generally mild (for example, gastrointestinal effects, shivering, fever), but on occasion can be severe or life-threatening (for example, uterine rupture). Misoprostol's effects on the fetus exposed in utero are unclear, and may be associated with limb defects and facial paralysis (PATH/EngenderHealth 2001). Reports of women who use misoprostol to self-induce abortion have been common in Latin America, particularly Brazil (Barbosa et al. 1993).

Additional studies, determination of appropriate guidelines, and increased information dissemination will help ensure the successful use of misoprostol during pregnancy, delivery, and postpartum care (PATH/EngenderHealth 2001).

Top of page