House of Commons - International Development

Select Committee on International Development Written Evidence

Memorandum submitted by Columbia University, Mailman School of Public Health/Averting Maternal Death and Disability Program

SUMMARY

Of the eight Millennium Development Goals (MDGs) established in 2000, the least progress has been made toward MDG 5—to improve maternal health.[5] Maternal mortality remains at unacceptable levels in the developing world in large part because good-quality emergency obstetric care (EmOC) is still inaccessible to many women. In order to effect real change, donor agencies, health organisations and national governments must collaborate to strengthen health systems and improve the availability, quality, accessibility and utilisation of EmOC for women around the world.

1. Agency Overview

1.1 Columbia University's Mailman School of Public Health is a leading institution of higher learning in the United States and in the world. Its research, education, and service agenda addresses the critical and complex public health issues that affect millions of people locally and globally. The Mailman School is especially known for its pioneering work on women's reproductive health and human rights.

1.2 The Averting Maternal Death and Disability Program (AMDD), based at the Mailman School, was established in 1999 to reduce maternal mortality and morbidity in the developing world by improving the availability, quality and utilisation of emergency obstetric care (EmOC). AMDD believes that sustained reductions in maternal mortality can only be achieved through systemic country-wide improvements to heath systems, inclusive of EmOC.

2. Background

2.1 Major Causes of Maternal Mortality

According to the World Health Organisation (WHO), a maternal death is "the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management but not from accidental or incidental causes."[6] The maternal mortality ratio is defined as the number of maternal deaths that occur per 100,000 live births. Three-quarters of maternal deaths around the world can be attributed to five major causes, all of which can be treated if appropriate services are available. It is estimated that 25% percent of maternal deaths are attributable to haemorrhage, 15% to infection, 13% to abortion, 12% to eclampsia and 8% to obstructed labour.[7]

2.2 Stark inequities between developing countries and developed countries are evident with regards to maternal mortality. Over 500,000 maternal deaths occur each year around the world, and over 99% of these deaths occur in developing countries. A woman's chance of dying a maternal death in the developing world is 1 in 61, and in the developed world, only 1 in 2,800. Further inequities exist even within the developing world; a woman's lifetime risk of maternal death ranges from 1 in 20 in Africa to 1 in 160 in Latin America. Even within Africa, a woman's lifetime risk of such a death is 1 in 16 in sub-Saharan Africa, compared with 1 in 210 in Northern Africa. The major reason for these discrepancies is the lack of appropriate care available in developing countries.[8]

2.3 Even within countries there is inequitable access to the care that does exist, however inadequate. Differential access to care is often determined by degrees of social disadvantage, including those of educational level, wealth, urban versus rural residence, religion, ethnicity, and forced displacement status.

2.4 Approximately 15% of all pregnant women will suffer a direct obstetric complication, either during pregnancy, delivery or following delivery, all of which could lead to death or disability if left untreated. Some obstetric complications can be prevented: active management of the third stage of labour can reduce the incidence and/or severity of postpartum haemorrhage and access to family planning services and safe abortion can reduce deaths from abortion. However, most complications cannot be foreseen or prevented. These complications can only be treated. For this reason, access to EmOC and the presence of skilled attendants during delivery are indispensable in the fight to reduce maternal mortality.[9]

2.5 "Where is the M in MCH?" was a question posed in 1985 by Dr. Allan Rosenfield and Deborah Maine of Columbia University in response to the dearth of attention paid to maternal health within maternal and child health programmes. This article contributed to the creation of the Safe Motherhood Initiative, whose voice has grown over the past twenty years. During the 2000 Millennium Assembly of the United Nations (UN), 147 leaders from around the world signed the Millennium Declaration, which included the eight Millennium Development Goals (MDGs). The fifth MDG calls for the improvement of maternal health. More specifically, member countries intend to reduce their maternal mortality ratios by 75% between 1990 and 2015.[10] Data suggest, however, that progress has been slow, and maternal mortality in developing countries remains unacceptably high.[11] In order to begin to make real progress, health systems in developing countries must be strengthened, and EmOC must be made available and accessible to all women. Only with additional resources and increased political attention to the issue will these changes be possible.

2.6 Emergency obstetric care is necessary to prevent maternal deaths

EmOC is a standard set of medical interventions which treat obstetric complications and, accordingly, prevent maternal deaths. The importance of EmOC has been endorsed by most international health organisations, including WHO, UNICEF, UNFPA and DFID. Basic EmOC comprises the following interventions, or "signal functions": 1) parenteral administration of antibiotics, for the prevention or treatment of sepsis, 2) administration of oxytocic drugs, which induce contraction of the uterus, to control bleeding, 3) administration of anticonvulsants (eg magnesium sulphate), for the treatment of pre-eclampsia and eclampsia, 4) manual removal of placenta, to control bleeding, 5) removal of retained products (eg, manual vacuum aspiration), to control bleeding and to prevent infection, and 6) assisted vaginal delivery (eg, vacuum), in the case of prolonged labour. Comprehensive EmOC includes all of the above interventions, in addition to surgery, namely caesarean sections, and blood transfusion. To provide the signal functions, facilities must have the requisite medications, supplies and functioning equipment and staff must be trained, competent and authorized to provide the intervention. In order to reduce maternal mortality in a real way, high quality EmOC must be available and accessible to all women and utilized by those women who suffer from obstetric complications.[12]

2.7 A set of six process indicators was established in the 1990s to evaluate health services' capacity to provide EmOC. These UN EmOC Process Indicators are: 1) The quantity of EmOC facilities in a region, generally described by the number of functioning basic and comprehensive EmOC facilities available per 500,000 population. The recommendation is for at least four basic facilities and one comprehensive facility per 500,000. 2) The geographic distribution of EmOC facilities, in order to ensure that the minimum standards of availability are met in all regions of a country. 3) The proportion of births which take place in a basic or comprehensive EmOC facility; the recommendation is that this proportion be at least 15%. 4) The met need for EmOC, or the proportion of women who require EmOC services who are treated in an EmOC facility. The goal is 100%. 5) The percentage of births which take place via caesarean section; the recommendation is between 5% and 15%. 6) The case fatality rate, which is the percentage of women admitted for obstetric complications who die as a result of those complications. It is recommended that the case fatality rate of women with obstetric complications be less than 1%.[13]

3. State of emergency obstetric care

3.1 The 2007 DFID Annual Report notes that progress toward MDG 5 is lagging. No change or negative progress has been made toward the reduction of maternal mortality in sub-Saharan African and in South Asia. Some or negligible progress has been made toward this goal in Northern Africa, Southeast Asia, West Asia, Latin American and the Caribbean.[14]

3.2 An important component of health systems which must be addressed is human resources. EmOC services cannot be provided without adequate numbers and levels of staff who are appropriately trained, posted and retained. One solution to the problem of insufficient staff, or inappropriate distribution of staff, is the expansion of mid-level providers' capacity and authority to provide EmOC services.

3.3 This strategy of training mid-level providers has been extremely successful in countries such as Mozambique, Tanzania and Malawi. In Mozambique, for example, the lack of physicians was a central factor leading to the training of dozens of surgical technicians to provide a variety of surgical interventions, including caesarean sections, throughout the country. The importance of the surgical technicians is particularly great at rural hospitals. While physician retention in these hospitals is extremely low, 88% of surgical technicians assigned to rural hospitals remained there seven years later. Data from 2002 show that surgical technicians performed more than half (57%) of major obstetric surgeries in the country, including nearly all such surgeries (92%) in rural hospitals. The case fatality ratio for surgeries performed by surgical technicians is only 0.4% and 0.1% for emergency and elective procedures respectively. Data showed that the only difference between outcomes of caesarean sections performed by surgical technicians compared to those performed by physicians was a slightly higher occurrence of superficial wound separations. The training of surgical technicians has also proven cost-effective; one recent study found that the cost of major obstetric procedures performed by a surgical technician is, on average, 27% of the cost of such a procedure when performed by a physician.[15][16][17][18][19]

3.4 The UN EmOC Process Indicators have been used in some 50 countries to assess the status of and to monitor progress in the provision of emergency obstetric services. Findings from needs assessments using these indicators are described below in paragraphs 3.5-3.7. It is clear from analysis of the assessments which have been conducted that availability, quality and utilisation of EmOC services at all levels need to be improved.

3.5 National assessments of health centres and hospitals in Africa, Asia and Latin America have shown that many countries have inadequate numbers of EmOC facilities for their population size; most facilities providing delivery care only provide some of the signal functions that would qualify them as EmOC facilities. This means that even women who live near health facilities may not find the medical services needed to treat life-threatening complications.[20] Assessments of health facilities serving forcibly displaced populations in Africa show similarly poor capacity to deliver EmOC.

3.6 Most assessments showed that few health facilities delivered the full range of EmOC signal functions they were expected to provide. Administration of oxytocics, antibiotics and anticonvulsants were available more frequently than procedures such as manual removal of the placenta and removal of retained products. Assisted vaginal delivery was reported to be the basic EmOC signal function least likely to be performed. In terms of comprehensive EmOC, the capacity to perform caesarean deliveries varied widely, ranging from 100% of hospitals in Nicaragua and the Sofala province in Mozambique to 33% of hospitals in one region of India.[21]

3.7 Geographic distribution of EmOC facilities remains a problem, especially in rural areas. While many countries have the minimum recommended number of comprehensive EmOC facilities per 500,000 population, these facilities tend to be very poorly distributed. Many sub-national regions are lacking comprehensive EmOC. Of a sample of nine developing countries, for example, only Benin had the minimum number of comprehensive EmOC facilities in every sub-national region.[22][23]

3.8 Met need for EmOC is quite low. National needs assessments in nine countries in sub-Saharan Africa showed that met need was on average 28% (ranging from 12% in Mali to 48% in Benin), far below the UN recommended level of 100%. Such low met need indicates that too many women in those countries are not receiving treatment for their obstetric complications.[24]

3.9 Health care is not only a service provided to individuals, but also a basic human right. National governments are obligated under international human rights law to progressively realize the highest attainable standard of health, which includes ensuring that all women have access to good-quality pregnancy-related care.

4. Case Study: Chad

4.1 A woman born in Chad has a one in 11 chance of dying as a result of complications of pregnancy or childbirth, compared with one in 3,800 in the UK. A 2002 study evaluating the six EmOC process indicators in Chad found that few targets were being met. According to the process indicator recommendations, Chad should have a minimum of 63 facilities with the capacity to provide basic EmOC in addition to a minimum of 16 facilities with the capacity to provide comprehensive EmOC. Study authors, however, found that Chad had only 40% of the EmOC facilities recommended for its population size. Distribution is also a problem; no region had the minimum number of recommended basic facilities, and the northern region had no EmOC facilities at all. The met need for EmOC was only 12%, diverging sharply from the recommended 100%, and the CFR was 3.9%, four times the recommended maximum of 1%.[25][26]

4.2 Nearly one-quarter (23%) of maternal deaths in Chad in 2002 were attributable to pre-eclampsia and eclampsia, hypertensive disorders experienced during pregnancy. Both of these conditions can be easily treated with magnesium sulphate, an inexpensive anticonvulsant medication common in the developed world. Unfortunately, the same medication is nearly impossible for health providers to obtain in many developing countries.[27][28][29]

4.3 Dead Mums Don't Cry

In 2005, a programme aired on BBC as part of the Panorama Series. This programme, Dead Mums Don't Cry, chronicled the unflagging work of Dr. Grace Kodindo, an obstetrician fighting to save women's lives daily in Chad's major referral hospital, the Hôpital Général de Référence, located in the capital city of N'Djamena. At the time of filming, there was no magnesium sulphate in the entire country of Chad. As Dr Kodindo notes of her native country, "Everything is lacking here."[30]

4.4 The response to the account of Dr Kodindo's struggle and the situation in Chad has been remarkable. A UK-based non-profit organisation created in response to the programme, "Hope for Grace Kodindo," has sent medications, baby clothes, and other vital supplies to the Hôpital Général de Référence. One accomplishment has been the provision of magnesium sulphate: as a result of these donations, over 1,100 women's lives have been saved.[31]

4.5 While such efforts are inspiring, and important, they are not sustainable. Other crucial steps must be taken to improve women's health. The national essential drug list of all countries must be updated to include such vital drugs as magnesium sulphate, and staff must be trained in their use. Drug and equipment procurement systems must be strengthened. National protocols on EmOC must be updated, and medical staff must be trained and supported in the use of new medications.[32]

5. Required actions

5.1 The single most important task of organisations seeking to decrease death and disability due to pregnancy and childbirth is to strengthen health systems at all levels to deliver EmOC—especially at sub-national levels. A crucial component of this task will be to focus on the systematic implementation of EmOC services, with equitable distribution to all women. Only when this is achieved will the reduction of maternal mortality become a reality. Columbia University's Averting Maternal Death and Disability Program (AMDD), for example, was able to help upgrade hundreds of existing facilities in 18 developing countries to increase met need for EmOC by at least 100%, and to significantly reduce case fatality rates, often by 50% or more, over the course of only a few years, through effective partnerships and a systematic focus on improving facility, district and national systems.[33][34][35][36][37]

5.2 Organisations must ensure that refugee and internally displaced women are not excluded from the prioritisation of EmOC and other reproductive health services for women in developing countries. DFID is an important donor to UN agencies and to humanitarian relief organisations, and as such should encourage these organisations to include comprehensive reproductive health services, including EmOC, in their programmes.

5.3 Needs assessments have shown that in many places facilities exist but lack the trained staff, essential drugs, equipment and supplies to make them function as EmOC facilities. In order to effectively strengthen health systems, donor agencies must encourage the systematic upgrading of existing health centres and hospitals to provide the full range of life-saving signal functions.[38]

5.4 Donors should encourage national governments, implementing organisations and policy-making bodies to expand the role of mid-level providers to provide good quality EmOC, with requisite training and support, and within strengthened health systems.

5.5 DFID's maternal health strategy appropriately focuses on the most critical elements of maternal mortality reduction, notably EmOC. Moreover, DFID is a global leader in this field, together with its partners in the International Health Partnership. It is crucial that DFID maintain this evidence-based focus on what is needed to reduce maternal mortality. It is also crucial that DFID maintain and expand its global leadership in maternal health.

5.6 Funding

The charge to strengthen health systems and all its components necessitates additional funding; donors such as DFID will be crucial to the success of the process of strengthening health systems. There are several ways in which donor agencies can be most efficient. First, they can make commitments to long-term investments. Second, they can provide funding which complements national health programmes and ensures donor harmonisation. Finally, stakeholders at every level should be included in plans for funding.[39] DFID's participation in the International Health Partnership will help to accomplish these goals, and DFID should continue to pursue such collaborations.

5.7 In the 2005-2006 fiscal year, DFID spent over £2.5 billion on its Bilateral Programme. Over £200 million, or 17%, of this was spent on the health sector, representing an increase in funding from the previous four fiscal years.[40] However, the health sector requires additional support. In addition, of the health spending only £16 million was spent on maternal and newborn health (excluding Poverty Reduction Budget Support).[41] Given DFID's strong strategy and leadership position, further spending is called for. In addition, allowing for explicit funding for maternal health, including EmOC, family planning and safe abortion services, can help to ensure that these effective interventions are prioritised. Without additional funding and attention from DFID and other donors, little, if any, progress will be made toward MDG 5.

5.8 Donors have the obligation to challenge governments of countries suffering from high maternal mortality to redouble their efforts to improve maternal health and reduce maternal deaths. International organisations must work in concert with national and sub-national actors in order to effect real change in maternal mortality.

5.9 Health is a basic human right, and as such, donors must support national governments in their progressive realisation of this right.

5.10 Donors must also promote increased accountability at all levels of the health care system, from health workers to donor agencies, in order to reach the common goal of maternal mortality reduction. As noted in the Task Force on Child Health and Maternal Health's progress report, "Accountability should lie at the heart of the MDG initiative."[42]

September 2007