House of Commons - International Development

Select Committee on International Development Fifth Report

4 STRATEGIES FOR SUCCESS

92. This chapter will address strategies that have been shown to work in improving maternal health, and assess whether they can be replicated in other contexts and at scale. A number of promising approaches have been explored earlier in this report; for instance, the IMAGE microfinance, HIV and gender equity project in South Africa described in Chapter 2. A particular focus in this chapter will be what has worked in strengthening health systems—health workers, equipment and supplies (including essential drugs), clinics and infrastructure—and how donors can balance these supply-side approaches with appropriate demand-side interventions as discussed in Chapter 2. The chapter will also look specifically at successful approaches to improving maternal health in conflict-affected and fragile states. Finally, the chapter will address the crucial question of strengthening the measurement of progress in maternal health.

What works in preventing maternal deaths

THE EXAMPLE OF NEPAL

93. One country which was frequently mentioned when we asked witnesses for examples of successful donor support to maternal health was Nepal. In 1996, Nepal was reported as having one of the highest levels of maternal mortality in the world. This was due to several inter-related factors such as: low access to family planning; high prevalence of early marriage; the prohibition of abortion; low uptake and provision of skilled birth attendance; and limited availability of emergency obstetric care.[199]

94. A decade later, Nepal was reported to have experienced a significant decline in its maternal mortality ratio (MMR), from 539 in 1989-1995 to 281 per 100,000 live births for the period 1999-2005.[200] The DFID-supported Nepal Safe Motherhood Project (1997-2004) and Support to Safe Motherhood Programme (SSMP, running from 2004-2009 with £20 million DFID funding) were credited by witnesses as playing a key role in this reduction.[201] It is important to note, however, that there is debate over the true extent of the maternal mortality reduction, since there are wide and almost overlapping lower and upper limits for the MMR estimates of 581 for the period 1989-1995 and of 281 for 1999-2005 (392 to 686 versus 178 to 384 respectively).[202] The SSMP told us that there was sufficient circumstantial evidence to suggest that a real decline has taken place.[203]

95. The SSMP takes a multi-pronged approach that seeks to assist policy formulation, provide safe abortion services, improve emergency care and strengthen infrastructure.[204] DFID funds are given in the form of financial aid, technical assistance and direct support to UNICEF, the agency which helps to implement the programme.[205] The Towards 4+5 Research Programme Consortium highlighted a number of specific achievements under SSMP: support to emergency obstetric care; the training of more skilled birth attendants; the integration of maternal health work with the national health programme; and a financial incentive scheme to encourage women to have their deliveries in health facilities (as we described in Paragraph 41).[206] We applaud DFID for its contribution to the Nepal Safe Motherhood Project and Support to Safe Motherhood Programme, which have included a range of interventions relevant to maternal health in Nepal over a decade that has witnessed progress in reducing maternal mortality. We urge DFID to support independent comprehensive evaluations of this experience, with a view to sharing lessons in the region and globally.

96. Another DFID-supported project in Nepal highlighted to us was the 'women's group' approach, conducted as a research study, whereby groups were supported at village level to carry out advocacy for safer births, build awareness of maternal complications, promote hygiene and prevent delays in seeking care.[207] Evaluations of the approach suggest that neonatal survival benefits will be significant, with the possibility also of a direct effect on maternal survival (although this was not the main outcome being measured).[208]

97. Possibilities for replication of the 'women's group' approach appear high and trials to assess this are being conducted in Bangladesh, India, Pakistan and Malawi.[209] The Towards 4+5 Research Programme Consortium stated:

210

98. Dr Gill Greer, Director General of IPPF, was cautiously optimistic that the DFID-supported wider programme in Nepal had potential for replication.[211] However, YozuMannion criticised DFID for not already taking the opportunity to replicate successes from Nepal elsewhere.[212] Dr Stewart Tyson from DFID was wary about replicating "too much" and thought that many of the innovative approaches used "may or may not be transferable."[213] Baroness Vadera underlined the need to show caution in attempting to reproduce context-specific approaches.[214] We urge DFID to look closely at options for replicating successful approaches from Nepal where appropriate, and to identify factors relevant to scaling-up and transference. We appreciate that success is often context-dependent, but believe the DFID-funded approach to supporting women's groups, as in Nepal, is worthy of particular consideration wherever relevant.

What works in strengthening health systems

Boosting human resources

99. 'Health systems' encompass infrastructure and equipment, human resources, financing and the various processes that enable staff to work with a system, such as communication.[215] Aspects of health systems strengthening that are particularly significant for maternal health relate to how the coverage and quality of skilled birth attendance at delivery and emergency obstetric care can be best improved. This often comes down to the issue of increasing the availability of skilled human resources.[216] In its written submission, the Royal College of Obstetricians and Gynaecologists identified the management of skilled human resources as "the most challenging aspect of MDG 5".[217]

100. Evidence we received welcomed the focus on health system strengthening within DFID's health strategy.[218] Immpact's written evidence stated that "DFID has been instrumental in putting the strengthening of health systems on the international agenda, and in making the case for maternal health as a health system issue."[219]

101. Several inter-related aspects of human resources issues struck us as needing urgent attention if progress is to be made on maternal health. The first concerns the sheer numbers of health professionals needed if more women are to deliver with a skilled attendant and to have ready access to emergency care should the need arise. There is an estimated global deficit of 2.4 million health workers, including 700,000 midwives alone.[220] Midwives are crucial to the achievement of MDG 5 as they provide both skilled birth attendance and some aspects of emergency obstetric care, if the right equipment and supplies are present. Dr Nynke van den Broek of the Liverpool School of Tropical Medicine told us that seven out of the eight key functions of obstetric care could be performed by midwives, including caesarean sections and blood transfusions.[221]

102. However, the quality of care that a midwife is able to provide will depend on their training and the wider health system in which they are working.[222] Using unqualified staff can worsen many obstetric complications.[223] In many regions, especially rural areas, training opportunities are sparse and of poor quality: Dr Monir Islam from WHO told us of a midwifery training school he had recently visited in Zimbabwe that had just one book and three teachers between 120 students.[224] Evidence from the Royal College of Obstetricians and Gynaecologists stated that "countries where progress towards MDG 5 has been achieved are those where midwifery as well as medicine is highly professionalised and well-recognised, including at government level."[225]

103. One written submission suggested that action-oriented research is needed into where staff shortages are most acute, what training is available and how midwives are paid and employed.[226] This would help target policies to address the human resource crisis. Similar research on clinical officers and medical assistants would also be helpful. The Peoples' Health Movement and Global Health Watch Secretariat suggested that DFID should consider working with the Global Health Workforce Alliance and the Alliance for Health Policy and Systems Research to help fill such a knowledge gap.[227] We were concerned to learn the extent of the global shortfall in health workers, particularly the lack of midwives. Boosting the numbers of midwives worldwide will be central to the achievement of MDG 5. Increasing the availability and quality of training opportunities for midwives is therefore of paramount importance. DFID should consider supporting action-oriented research into where human resource shortages and training needs are particularly acute and the options for addressing them in the short, medium and long term.

104. As Thoraya Obaid of UNFPA told us, health professionals should not only be well-trained but enjoy good working conditions, not least so that trained medical and midwifery staff do not migrate to more financially rewarding and satisfying jobs outside their home country (the so-called 'brain drain').[228] The UK Government has recently implemented a policy of avoiding the active recruitment of health care workers from developing countries for the NHS and this step was applauded in written evidence.[229]

105. Persuading staff not only to stay in their home country but to work in rural and remote areas is a key issue for donors and governments. In south and south-east Asia, the key human resources issue is not primarily one of absolute shortages but of poor distribution. In India, for example, a national assessment found that just 6% of the required obstetricians and 27% of the required nurses and midwives are currently deployed in rural settings.[230] Brigid McConville from the White Ribbon Alliance told us of a successful five-year advocacy campaign in Tanzania to persuade the Government to deploy more skilled birth attendants in rural areas. This has resulted in a doubling of health workers in some areas.[231] Ms McConville believed that DFID should support civil society to lobby governments for better salaries and incentives to work in rural areas.[232] The Royal College of Obstetricians and Gynaecologists said that DFID should fund projects that encourage health professionals to work in rural areas by offering financial, social and professional support.[233] We believe that DFID and other donors should find new ways to help governments encourage health professionals to provide quality services in remote and rural areas. This should include supporting civil society to lobby for better salaries and conditions for doctors and midwives working outside urban areas and to ensure the necessary infrastructure, supplies, transport and equipment are in place to enable these professionals to provide prompt and effective care.

106. Recruitment, the brain drain and training and deployment of staff are all addressed under DFID's Emergency Human Resources Programme (EHRP) in Malawi (funded with £55 million over six years, 2005-2011). Malawi's human resources crisis is acute, partly because of the HIV/AIDS epidemic. The country's maternal mortality ratio (MMR) is estimated as 1,100 deaths per 100,000 births, making it one of only 11 countries worldwide with an MMR of over 1000.[234] The EHRP has several main elements:

improving incentives for recruitment and retention of Malawian staff (including salary top-ups);
using international volunteer staff in the interim whilst more Malawians are being trained;
expanding domestic training capacity for doctors and nurses/midwives by over 50%; and
strengthening Ministry of Health capacity to manage, monitor and evaluate human resources.

The programme specifically addresses career development and incentives for deploying staff to under-served areas.[235]

107. DFID says that since the programme began in 2005, provisional results are promising. Staff across the cadres of health professionals increased by 450 in the first 9 months, 60 volunteers have been recruited to fill gaps, training schools have increased intakes and there has been a significant decline in the number of nurses leaving the country to work abroad.[236]Whilst several witnesses cautioned that it was too early to show confirmed results,[237] the Peoples' Health Movement and Global Health Watch were impressed with the Programme and were keen to replicate it elsewhere:

—

238

Other evidence questioned why the EHRP is not being replicated by DFID elsewhere.[239] DFID deserves credit for its support to the Emergency Human Resources Programme in Malawi, for which initial results show expanded staff numbers and better uptake of training. We recommend that DFID move swiftly to support the replication, where appropriate, of efforts to address human resources problems as soon as conclusive results are available.

INCREASING THE AVAILABILITY OF EQUIPMENT AND SUPPLIES

108. The effective delivery of maternal health care relies on access to essential supplies and equipment, yet major shortages of even the most basic drugs, commodities and medical apparatus exist in many developing countries. The reality of giving birth in a facility with empty blood banks, no saline solution for drips and no routine medicines for complications was vividly conveyed by Dr Grace Kodindo, an obstetrician from Chad, in evidence to us and in her BBC Panorama film, Dead Mums Don't Cry.[240]

109. At the time of filming the Panorama programme, one very basic drug—magnesium sulphate, used to treat eclampsia—was not available anywhere in Chad. Given that the drug is extremely cheap—"cheaper than table salt" as Brigid McConville told us—yet also highly effective we could not understand why access is so limited in some developing countries.[241] Magnesium sulphate is not on the Essential Drugs Lists of many African countries and Dr Kodindo thought it vital that it be included.[242] We were concerned to hear about the lack of even very basic supplies and medicines in many developing countries. We recommend that donors, including DFID, work with the World Health Organization to advocate with national governments for national Essential Drugs Lists to contain drugs such as magnesium sulphate, which are crucial to maternal survival.

110. IPPF's evidence explained that the reasons for the shortages of basic drugs and supplies largely stemmed from insufficient funding, but also included: inadequate forecasting of supply needs; weak distribution systems within countries; and regulatory, tariff and tax barriers that hinder importation and provision.[243] Dr Kodindo explained that blood shortages were often due to a lack of appropriate storage and refrigeration—also a problem for some major obstetric drugs, including oxytocin—and also to cultural beliefs discouraging blood donation.[244] In rural areas, few hospitals will have blood banks.[245] Yet women with anaemia—a very common condition amongst poor rural communities—can deteriorate or die very quickly even after minor bleeding.[246]

111. Achieving the additional MDG 5 target of "universal access to reproductive health" will be dependent on addressing shortages of modern contraceptives and reproductive health commodities. Marie Stopes International said that many of their partner organisations are often forced to turn away a large proportion of clients at public clinics simply because they have run out of the basic contraceptive supplies.[247] Addressing such shortfalls will also make a direct contribution to reducing maternal deaths, since unwanted and unplanned pregnancies contribute to high levels of unsafe abortion, as described earlier.[248]

112. DFID is a member of the Reproductive Health Supplies Coalition which was set up to provide global leadership in increasing the availability of reproductive health products.[249]DFID admitted that family planning supplies are inadequately financed, both by national governments and global financing initiatives.[250] Of 16 countries receiving DFID support in Africa, for example, only 10 have specific budget lines for sexual and reproductive health supplies. Further, direct donor support for family planning supplies and services fell from $590 million in 1995 to $460 million in 2003.[251] In October 2007, DFID did, however, announce an additional £100 million for UNFPA, to be spent on family planning.[252] In addition to insufficient quantities of essential drugs, many countries have widespread shortages of other pre-requisites for maternal health and services, including adequate blood and family planning supplies. We believe that DFID should seek to build political commitment within countries to ensure that these crucial supplies are appropriately funded within national health plans and budgets. The Department should also campaign internationally for a reversal in declining budgets for family planning supplies and services.

Balancing the demand and supply-side of care

113. As The Lancet's series on Maternal Survival in 2006 made clear, the supply-side approaches described above need to be balanced with demand. The key constraints to maternal health vary significantly depending on the context and thus the challenge for donors lies in supporting the most appropriate configurations of care for the specific country or region. For example, where sufficient health care facilities are available but are under-used, demand-side approaches need to address issues such as transport and cultural barriers. However, where under-use is primarily due to supply-side constraints such as lack of drugs and supplies, insufficient health personnel and poor quality care, improvements in the supply-side need to be the priority.[253]

114. A well-performing health system will have an effective balance of demand and supply-side interventions. Countries which are struggling to meet MDG 5 often have an imbalance between the two and hence, in order to catalyse progress, governments and donors need to ensure that their assistance helps redress this imbalance, rather than making it worse.[254] As Richard Horton of The Lancet emphasised to us: "You need dual approaches. You are not going to solve this by a purely top-down building of clinics and facilities. You have got to mobilise the grass-root support in villages."[255]

115. In Chapter 5, we will address DFID's use of budget support and other modes of financing for maternal health. But the importance of donors retaining oversight of programmes when using budget support for maternal health is worth noting. This is necessary to ensure that balance between supply- and demand-side interventions is achieved. In order to achieve efficiently functioning health systems, there needs to be a balance of demand- and supply-side approaches. We believe that DFID needs to ensure that its support for demand- and supply-side approaches is flexible and reflects the needs of specific contexts, and that it is consistent with broader health systems strengthening in countries. Where budget support is being used, DFID and other donors should retain oversight of national programmes to ensure this balance is achieved. Monitoring systems need to be capable of tracking this balance.

Working in conflict-affected and fragile states

116. In fragile states, many of which are affected by conflict, maternal mortality ratios can be 2.5 times higher than in more stable countries at similar levels of development and income.[256] According to DFID, fragile states are those which have governments that cannot or will not deliver core functions to the majority of its people, including the poor.[257] Fragile states comprise 14% of the world's population but account for an estimated third of maternal deaths and almost half of all child deaths.[258]

117. There are both demand and supply aspects to poor maternal health in fragile and conflict settings.[259] Women are often exposed to greater risk of sexual violence—and therefore sexually transmitted infections and unwanted pregnancies.[260] It is estimated that in Liberia, for example, 60-75% of women of child-bearing age were forced into sex during the conflict between 1989 and 2003.[261] Conflict also constrains the supply-side of reproductive and maternal health services: health systems often collapse almost completely and geographical access is often more difficult because of insecurity and damaged infrastructure.[262] Health workers may move to safer areas.[263] Post-conflict countries, such as Sierra Leone, often find it can take many years to re-build infrastructure and develop sufficient human resources to provide maternal health care.

118. DFID supports maternal health care in such settings in a number of ways. It funds NGOs and other organisations working on maternal health in many conflict-affected countries, including Somalia, Sudan, Liberia, DRC, Sierra Leone and Burundi.[264] Supporting civil society organisations is important in fragile states because they often provide health services themselves where governments do not exist or are unable to provide services.[265] DFID also supports multilateral agencies working in conflict settings, including WHO, UNFPA and UNICEF.[266]

119. The UN recently introduced a "cluster approach" for emergency responses that aims to enhance co-ordination between different humanitarian actors and strengthen service delivery on the ground.[267] WHO is the lead agency for the cluster on health. Aasha Pai, Acting Regional Director for Africa and Latin America for Marie Stopes International broadly welcomed the cluster initiative as a serious attempt to improved co-ordination between different agencies.[268] Co-ordination is crucial given the presence of multiple partners in emergency situations.[269]

120. However, evidence from the RAISE initiative[270] expressed concern that the cluster approach fails to integrate sexual and reproductive health services as a core part of humanitarian responses.[271] Marie Stopes International (MSI) also acknowledged this, saying that family planning provision, in particular, was neglected within the cluster system and that donors needed to support humanitarian agencies to improve this.[272] Other witnesses believed that DFID could use its influence within the cluster system and strengthen its role at country level to help prioritise maternal, reproductive and sexual health within humanitarian responses.[273] We believe that maternal health should be an essential and integral part of all humanitarian responses. Women in conflict settings are more at risk of poor maternal health and have fewer—or no—services available to them. We recommend that DFID advocate within the UN cluster system—both amongst other donors and the lead agency, the World Health Organization—for maternal, sexual and reproductive health to be prioritised in humanitarian emergencies.

121. Aasha Pai of MSI thought that, as well as working to improve multilateral responses, DFID should do more at programme level to raise the profile of maternal health in conflict settings:

"DFID has been a leading donor in terms of humanitarian issues. DFID has also been a leading donor in terms of gender and reproductive health [...] At the programming level DFID could do more to bring [the two] together also through specific mechanisms, like through DFID's Conflict, Humanitarian and Security Department, CHASE."

Ms Pai said that it was important to include maternal health care—for instance, emergency obstetric services—as part of the "basic health package" and in the health system right from the beginning in emergency and conflict settings.[274] She also emphasised the need to address the particular needs of individual maternal health programmes; for instance, in Afghanistan the cultural requirement to have only women doctors treating women.[275] Programme level work will often involve building up health systems from a state of collapse and this will require long-term political and financial commitment.[276] Save the Children UK recommended a twin-track approach that combined health systems building and the continuation of emergency relief until alternative structures are in place.[277]

122. Baroness Vadera told us that she believes that DFID has a comparative advantage in working on "difficult" issues such as safe abortion.[278] DFID's ability to work on these issues marks them out from NGOs and donors who cannot or will not give support to family planning or abortion services. Capitalising on this comparative advantage in conflict-affected and fragile settings would help ensure the necessary support for services that are essential to women at risk of forced sex, such as abortion and family planning services. We believe that DFID should go beyond immediate emergency relief and build on its ability to work on sensitive issues such as abortion, for which there is greater demand in conflict-affected and fragile settings and which urgently needs support. Efforts should be made to ensure that maternal care is a core part of both DFID's and national health programmes from the outset. A long-term dual approach that seeks to strengthen or re-build systems whilst continuing some aspects of emergency care is likely to work best.

123. Giorgio Cometto, Health Adviser for Save the Children UK, highlighted several examples of good practice of maternal health provision in fragile contexts, some of which had DFID involvement. One was in Afghanistan, where a relatively quick improvement in health outcomes had been achieved by sub-contracting the provision of health services to non-state providers.[279] This had boosted take-up of health services enormously, and there had been an increase in skilled birth and antenatal care attendance, and a decrease in child mortality from 257 per 1000 to just over than 190 per 1000 in five years (although thus far there had been no such dramatic reduction in maternal mortality).[280] Another example of good practice was in South Sudan, where Giorgio Cometto thought DFID had bridged the gap well between the winding-down of emergency funding and the establishment of health systems with a 'basic services fund' that addressed health care needs immediately.[281] DFID was also credited for its support to the development of a sexual and reproductive health policy in Sierra Leone.[282]

124. Baroness Vadera told us that DFID had benefited from its previous experience in Nepal, where DFID's maternal health programme had continued throughout a decade of conflict. She was keen to do more of this systematic programme work—work that went beyond emergency relief—in conflict and post-conflict situations.[283] We believe that DFID should learn from what has worked in terms of supporting maternal health programmes in fragile, conflict and post-conflict settings and share this knowledge appropriately elsewhere. This should include successful examples from DFID's own programmes, such as recent experiences in Nepal, Sudan and Afghanistan.

The need for improved health information systems to monitor progress

125. Seven years on from the launch of MDG 5, implementing accurate methods to measure maternal mortality remains a major challenge.[284] The fact that only 30% of countries have routine death and birth registration indicates the challenges in routinely collecting accurate data about maternal health and maternal deaths.[285]Maternal deaths are often not recorded, let alone investigated.[286] This means that official maternal mortality data are unreliable in many countries: a recent study conducted by Médecins Sans Frontières in the Democratic Republic of Congo estimated maternal death rates to be 10 times higher at 5,200 deaths per 100,000 live births than the national reported average of 520 deaths per 100,000 live births.[287] As we said in Paragraph 8, because of the absence of data from countries with some of the worst maternal mortality ratios, it is hard accurately to assess either progress in reducing deaths or the most effective strategies to improve maternal health.

126. One approach to addressing the difficulties in measuring maternal mortality is to help countries to develop reliable information systems, which are often weak due to financial constraints and skills shortages.[288] Accurate data are also crucial to ascertaining which national policies work most effectively.[289] Boosting national capacity to gather and use statistics can also strengthen advocacy for maternal health within countries: Richard Horton from The Lancet said that in Guatemala a mortality survey amongst women of child-bearing age had been a major factor in driving improvements in maternal health.[290]

127. DFID told us that it was "the biggest funder of statistics".[291] But Baroness Vadera recognised the difficulties in getting maternal health data: she said that lack of data was "possibly one of the reasons we have not been as effective as we could have been on maternal health."[292] DFID currently funds several initiatives working to improve maternal health data. The Health Metrics Network is a global partnership working from the premise that health information is not an end in itself, but a route towards better health.[293] DFID is providing £500,000 to the Network between 2007 and 2010. Dr Stewart Tyson highlighted the work being done by the Network to build the foundation for a comprehensive health information system, for instance by addressing registration of births and deaths.[294]

128. DFID is also funding the Immpact project, an international research initiative co-ordinated by the University of Aberdeen, with £7.5 million awarded for the period 2002-08. Other funders include the Gates Foundation, USAID and the European Commission. Through, for example, its 'Sampling at Service Sites' methodology, Immpact has pioneered a simple, low-cost way of estimating the levels of maternal mortality within countries. This method collects data from women respondents where they gather in large numbers (for example, markets and clinics) and typically costs much less than traditional house-to-house type surveys.[295] The studies conducted by Immpact have found that mortality levels are often considerably higher than officially recorded.[296] Alec Cumming of Immpact told us that the aim now was to offer this and other evaluation tools developed by the project as global "public goods" and that training courses in developing countries were about to begin.[297] DFID funding for the project will expire in August 2008.[298]

129. We received evidence which proposed that in-country data collection should be nationally 'owned' and locally relevant and that DFID should help build up in-country capacity to collect and use data on maternal health.[299] The Royal College of Obstetricians and Gynaecologists said a good starting point would be to seek the compulsory registration of all births and deaths in all countries.[300] Supporting improved health information systems in developing countries is of crucial importance to identifying and sustaining successful policies for maternal health. We believe that DFID should continue to support initiatives addressing weak information systems, such as the Health Metrics Network and Immpact. DFID should ensure that its programmes include a focus on strengthening national capacity to collect, analyse and use maternal health data.

130. More could be done to emphasise the importance of more accurate data within global financing and advocacy networks such as the International Health Partnership. As we said in Paragraph 22, maternal health can be viewed as a barometer of a nation's development and accordingly the use of maternal health data as a 'yardstick' for financing decisions by donors may act as an additional incentive for countries to improve maternal health itself. Baroness Vadera told us that DFID is encouraging the IHP to use maternal mortality as one of their key indicators for success.[301] The opportunities to highlight and address the urgent need for improved data that arise through various international initiatives, such as the International Health Partnership, should be seized and championed by DFID. The use of maternal indicators as a basis for financing decisions, for example, is likely to be a powerful stimulus to countries to improve maternal health itself.

131. Evidence from the Royal College of Obstetricians and Gynaecologists (RCOG) suggested that auditing the quality of maternal care and the reasons for deaths was also important. This is also true for other adverse outcomes, such as stillbirths. The RCOG pointed out that audit systems had been successfully developed in Sri Lanka and South Africa, where the system has helped to identify the nature of maternal health problems and where interventions and budgets have been adjusted accordingly.[302] Helping countries to monitor maternal deaths and the quality of care through routine audit systems will help to focus policies. We believe that DFID should help share lessons from developing countries that have successfully implemented audit systems of maternal deaths.

199 Ev 159-160 Back

200 Ev 207 Back

201 Ev 160 and Q 245 [Dr Gill Greer] Back

202 Ministry of Health and Population (MOHP) [Nepal], New ERA, and Macro International Inc. 2007. Nepal Demographic and Health Survey 2006. (Kathmandu, Nepal: Ministry of Health and Population, New ERA, and Macro International Inc.). Back

203 Ev 207-208 Back

204 Ev 208-210 Back

205 Ev 95 Back

206 Ev 152 Back

207 Ev 155 and Anthony Costello et al, 'An alternative strategy to reduce maternal mortality', The Lancet maternal Survival Series (September 2006), p.10 Back

208 Ev 154 Back

209 Ev 154 Back

210 Ev 155 Back