Restructuring the Health System: Experiences of Advocates for Gender Equity in Bangladesh

Abstract

This paper illustrates how advocates for gender equity succeeded in influencing the restructuring of the health system in Bangladesh in the mid-1990s but failed to influence its implementation. Using published and unpublished documents and personal interviews, it traces the changing fortunes of health sector reforms and reform advocates from 1995 to 2002 and analyses the major challenges and strategies involved. It explores the gains advocates for gender equity made in the design of the reforms and also their limitations to counter some of the inherent risks. The paper highlights the gaps between the design and implementation of reforms. It argues that the key factor for success during the design of reforms was the participation of civil society, which enabled a large number of women, particularly poor women, to engage with the design of reforms. During the implementation phase, however, reforms became a government–donor driven programme disconnected from civil society and space for women's voices became limited. When a new government assumed power in 2001, opponents of reform succeeded in halting the restructuring of the health system, in part because civil society was no longer engaged with the reform process and registered little protest.

Résumé

Au Bangladesh, à la moitié des années 90, les avocats de l'égalité des sexes ont influencé la restructuration du système de santé, mais n'ont pu influencer son application. L'article décrit le sort inégal des réformes du secteur de santé et des défenseurs des réformes de 1995 à 2002, et analyse les principaux obstacles affrontés et les stratégies utilisées. Il recense les progrès des défenseurs de l'égalité dans la conception des réformes et leurs limitations pour contrer certains des risques inhérents. L'article note leur influence sur la conception mais non sur l'application des réformes. La participation de la société civile à la conception a été encouragée, ce qui a permis à beaucoup de femmes, en majorité pauvres, de s'y associer. En revanche, pendant la phase d'application, la réforme est devenue un programme dirigé par le gouvernement et les donateurs, coupé de la société civile. L'espace pour la voix des femmes a été réduit. Lors du changement de gouvernement en 2001, les opposants aux réformes ont réussi à arrêter la restructuration du système de santé, en partie parce que la société civile n'était plus engagée dans le processus de réforme et a opposé peu de résistance.

Resumen

Este artı́culo examina la incidencia de las activistas a favor de la equidad de género en la reestructuración del sistema de salud en Bangladesh a mediados de los años 90, y su falta de incidencia en la implementación de dichas reformas. El artı́culo analiza los principales desafı́os que enfrentaron los promotores de las reformas entre 1995 y 2002, y las estrategias que usaron. Examina sus logros y también los factores que limitaron sus intentos de contrarrestar algunos de los riesgos inherentes. Durante el diseño de las reformas, se alentó la participación de la sociedad civil, permitiendo a muchas mujeres, especialmente mujeres pobres, a participar en el diseño de las reformas. Durante la fase de la implementación, sin embargo, la reforma se convirtió en un programa dirigido por el gobierno y los organismos donantes, desconectado de la sociedad civil. Se limitó el espacio en que las mujeres pudieron tener voz. Cuando asumió el poder un nuevo gobierno en 2001, quienes se oponı́an a las reformas lograron parar la reestructuración del sistema de salud, en parte porque la sociedad civil ya no participaba en el proceso de la reforma y apenas protestó.

Key Words:

• gender issues
• health sector reforms
• advocacy and political process
• Bangladesh

Bangladesh between the 1994 International Conference on Population and Development (ICPD) and today presents an interesting case study of the making and unmaking of reforms in the health sector. Following the ICPD, Bangladesh adopted its first Health and Population Sector Strategy (HPSS) in 1996¹ and a five-year (1998–2003) national Health and Population Sector Programme (HPSP) in 1998, with an approximately US$3 billion budget.² The significant policy and organisational changes proposed in HPSS and HPSP, which had been agreed with donors in a 1995 meeting,³ were:

• financial sustainability;

• a shift from project-based planning to sector-wide planning, management and financing;

• a shift from vertical to integrated service delivery through unification of health and family planning services under a single management structure;

• a shift from bureaucratic/ technocratic to participatory planning through involvement of communities and stakeholders in policy and programme formulation, implementation and monitoring;

• a shift from centralised planning and management towards decentralisation;

• a shift from the public sector being the primary provider to partnership with private and NGO sectors;

• maintaining a basic services package in the public sector and addressing the health needs of vulnerable groups;

• a shift from separate women's projects in which gender equity was addressed, to a gender-mainstreaming approach.⁴

Many elements of the reforms faced opposition from within the government. For example, the Planning Commission, which used to scrutinise and approve projects, was opposed to the idea of sector-wide programming because it transferred all the power to negotiate the programme with the donors from themselves to the Ministry of Health and Family Welfare (MOHFW). The Planning Commission wanted project-based planning to continue, in order to retain this control. Family Planning, particularly middle and senior managers, did not want unification with Health, as under the unified structure, health service people would be more privileged in getting management positions. Government officials, used to top-down centralised rule, were on the whole reluctant to engage in participatory planning, stakeholder consultation, decentralisation and partnership arrangements.

Despite these challenges, the MOHFW moved quickly to implement what were difficult structural changes. The MOHFW adopted sector-wide programming, management and financing, and reorganised service delivery in phases. In the first three years of HPSP implementation (1998–2000), the government unified health and family planning services at the upazila level (lowest tier of government services) and below, and started to establish “one-stop” community clinics to provide essential services package.⁵ On the other hand, stakeholder participation, NGO and private sector service partnerships with government services and social and gender equity aspects of the programme were not systematically promoted in this period.

The change of government brought about by national elections in 2001 provided the critics with an opportunity to find allies in the new administration, who succeeded in stalling the implementation of HPSP's key elements, particularly the planned unification of health and family planning services from the upazila level up to the centre, and the staffing of the one-stop community clinics. These critics also started questioning the merits of sector-wide programming. The government is now preparing the next health programme, but the future of the reforms has become uncertain.

The changing fortune of the reforms in Bangladesh calls for an analysis of the reformers and their strategies. Who pushed for reforms? Why and how? This paper will focus on the role of “gender equity advocates”, a term I use to refer to the coalition of women and men who worked to put a gender perspective during the process of developing the reforms. It will highlight their main concerns, identify the major challenges they faced and the strategies they used to influence the content of the reforms, and analyse their achievements and shortcomings.

The paper is based on a review of published and unpublished documents, primarily from government and international agencies, interviews in 2001 and 2002 with key informants, including eight government officials, ten representatives of international agencies and 11 representatives of civil society organisations, and the author's own personal experience as a member of the World Bank-led missions in 1996–1998 with responsibilities for integrating social and gender equity and stakeholder participation in the design of HPSS and HPSP.

Who advocated reforms: why and how?

In Bangladesh, the government and the international development agencies generally control policy and programme formulation. Citizens and civil society organisations have little access to these processes. As Bangladesh's dependence on international donor assistance started rising in the mid-1970s, the government began to initiate policy and organizational changes following donor prescriptions. USAID and the World Bank, the two most powerful donors, pressed Bangladesh to adopt a population control agenda. Since the 1970s, a World Bank-led donor consortium has been providing assistance to the MOHFW through four successive five-year population and health projects. These have been based on vertical family planning service delivery and the bifurcation of the health and family planning services of MOHFW, with most resources and programme interventions targeted to family planning.

Although many women-led NGOs were involved in supporting the country's family planning programme, many advocates of women's empowerment were critical of the family planning bias in the country's reproductive health services, in that they prioritised control of women's fertility over improvement of women's health.^{6 7} These advocates argued that policy and resource prioritisation and vertical programming contributed to success on one front and failure on another. In the first 25 years after independence, Bangladesh achieved remarkable success in lowering the fertility rate from 7 to 3.3 and increasing the contraceptive prevalence rate from 7% to 49%. However, it has failed to lower maternal mortality or increase the proportion of births attended by skilled attendants substantially from 5%–10%.⁸ Unfortunately, women's health advocates were limited in numbers and strength, and health issues have not been high on the agenda of the women's movement in Bangladesh. As a result, they have not been able to influence government policy priorities.

This situation started changing in the 1990s. The international women's health movement mounted a successful global mobilisation and forged an international consensus in support of a comprehensive agenda to promote women's reproductive health and empowerment at ICPD in 1994. At the same time, when the government started negotiations with the donors for the preparation and financing of the 5th five-year Health and Population project in 1995, the donors pressed for policy and structural reforms.

In general, the reforms supported by the donors were geared towards improving the efficiency and financial sustainability of the health system. However, different donors promoted different issues. For example, the Swedish International Development Authority (Sida) prioritised social and gender equity and stakeholder participation, and urged the inclusion of all stakeholder voices, including women, in policy and programme formulation. The Netherlands and Canada also prioritised gender equity. Bilateral donors, particularly the Netherlands and Sweden, were strongly committed to a sector-wide approach, while the UK Department for International Development (DFID) prioritised service and management reorganisation, and the World Bank prioritised identification and costing of a basic services package in the public sector.

The government of Bangladesh was sensitive to the efficiency arguments, but equity was also an important consideration, to meet the needs of vulnerable groups who could not afford private services. In September 1995, at a joint government–donor consultation in Paris, it was the government of Bangladesh who put poverty issues on the agenda. Consensus was reached on the main elements of reforms, described above, which formed the basis of the HPSS and HPSP.³

Gender equity advocates: challenges and strategies

Following this government–donor consensus on the major elements of reforms, the donor consortium was invited to field project preparatory missions. At that point, Sida proposed to recruit two international experts to work as members of the World Bank-led missions to provide technical inputs on social and gender equity and stakeholder participation issues. These two experts worked as catalysts to build a larger coalition of advocates for gender equity, the core of which included advocates for women's health and empowerment, drawn from civil society in Bangladesh and from international development agencies, Sida, the Netherlands, the Canadian International Development Agency, UNICEF and the World Bank. This core group was able to attract the support of a larger number of key actors from diverse backgrounds: policymakers and programme implementers from within the government, academic researchers, activists in the women's movement and staff of international development agencies who also had a commitment to social equity, inclusion and participation.

The group realised that the reform package had significant potential for both benefits and risks. For example, sector-wide programming could free gender issues from the narrow confines of special projects and enable gender-mainstreaming across all policies and programmes. Unification of health and family planning services would facilitate delivery of integrated sexual and reproductive health services. Community and stakeholder participation would open doors for the inclusion of women's voices in male-dominated government decision-making processes. At the same time, in defining a basic services package, it was important to include comprehensive sexual and reproductive health services. Partnerships with private sector services and cost recovery schemes required special measures to prevent adverse impacts on poor women.

Though the government and the donors had already taken the hard decisions and bore the brunt of responsibility for restructuring the health system, these advocates had the task of bringing a gender perspective to the seemingly gender-blind issue of restructuring. While they were convinced about the need to engage with the reform process, they were also aware of their limited numbers, resources and access. They required evidence and analysis to make a convincing technical case, but their work was constrained by limited data and the fact that existing planning tools and indicators were not gender-sensitive. The collection of relevant evidence and creation of appropriate tools and indicators required time and resources.

Nor were the challenges only technical. Advocacy to create support for the technical work was also needed. Alliance-building with other reform advocates, both within the preparatory missions and outside, to get them on board on gender issues was important. Even more important was the creation of space for the participation of a critical mass of Bangladeshi women's health advocates, who had hitherto not been party to any government–donor negotiation. These multiple challenges required a strategic response.

The advocates essentially focused on four strategies. Their first strategy was to push for making community and stakeholder consultation an integral part of the preparatory process. Such consultation would also generate relevant evidence regarding gender-specific priorities, needs and constraints. Although both the government and the donors agreed to this proposal, additional resources—funds and people—were needed to make it happen. A full-time Bangladeshi consultant was recruited and special donor funds were allocated to set up nationwide consultations. MOHFW established 17 task forces with members drawn from government, donors, and civil society including representatives of women's organisations to define the various elements of HPSS and HPSP. The task forces worked for nearly two years (1996–1998). One of the MOHFW task forces was on community and stakeholder participation, which was chaired by a Joint Secretary of the MOHFW. The Bangladeshi consultant noted above acted as the member-secretary. The task force identified 34 stakeholder groups and organised consultations in five districts with health service users and vulnerable groups (poor women and men, the elderly and adolescents) and with fieldworkers of the MOHFW, using the participatory rural appraisal methodology. Stakeholder workshops were organised in 12 districts (two from each of the divisions) by the MOHFW, facilitated by NGOs.

These consultations enabled thousands of poor women and men to articulate their views on the future health programme. The views of services users as well as service providers were sought on different elements of the reforms, including unification of services, decentralisation, community and stakeholder participation, and the Essential Services Package (ESP). Reports from these consultations were discussed by the HPSS and HPSP preparatory missions and the various task forces,^{9 10 11 12 13 14 15} and key findings were incorporated in HPSP design. They identified priority needs of vulnerable groups, generating gender-sensitive information on violence against women and male responsibility in reproductive behaviour, contributed to partnership between the government, NGOs, civil society and communities and to establishing transparency and accountability in the health system. Participants strongly recommended community and stakeholder participation in the implementation and monitoring of the HPSP.

A second strategy was to adopt a mainstreaming approach to address gender issues. It was agreed between the government and the donors that the two international experts would work across all policy and programme elements and not confine themselves to providing feedback for a single section on gender in the HPSS and HPSP documents. Their Terms of Reference in the preparatory missions specified this mainstreaming approach, which facilitated their engagement with other reformers and getting broader support for an equity agenda. During the mission, the gender equity advocates would meet briefly as a group in the morning to discuss the critical issues of the day and then spread themselves across the different thematic groups, to raise relevant gender concerns within each group according to their individual expertise.

Because they could not work on all issues, however, their third strategy was to focus on selected key concerns. In addition to community and stakeholder participation, they decided to focus on reproductive health, human resources development, behaviour/change/communication, and financing, as they felt redefinition of these issues would create maximum impact. Reproductive health needed to be redefined in the light of ICPD and funded within the Essential Services Package. Community and stakeholder participation were key mechanisms to give women voices in decision-making. Gender imbalances in human resources (women dominating the lower ranks of services with only a few in managerial positions) needed to be redressed and gender sensitivity training of MOHFW officials was necessary to make services gender responsive. Emphasising male roles and responsibilities in reproductive health and the need for girls' and women's empowerment were important for changing mind-sets. Finally, they suggested regular monitoring of financing and resource allocation to ensure access for vulnerable groups.

Their fourth main strategy was the championship of issues other than gender during the HPSS and HPSP negotiations. The gender equity advocates were the main champions of social equity, quality of care, service users' rights and satisfaction, and partnerships with NGOs. The leadership on these issues helped to situate gender concerns in the context of a broader social agenda, so that gender advocates were not perceived as a narrowly-focused group interested solely in one agenda, which helped in alliance-building.

Successes in influencing HPSP design

The major areas that the advocates can claim as gains were, first, success in making social and gender equity a central concern. The main goal of HPSS and HPSP is improvement in the health of the most vulnerable groups—women, children and the poor—by providing a client-centered Essential Services Package delivered through the primary health care system.² Most of the HPSP budget (60–70%) is allocated to this. Gender equity became a main objective against which the programme's performance would be measured.

Second, HPSP used ICPD definitions in elaborating reproductive health in the Essential Services Package, which included safe motherhood and maternal nutrition, family planning, prevention and control of RTI/STI/AIDS and infertility, adolescent care and neonatal care. For reducing maternal mortality, interventions such as emergency obstetric care, referral and training of community midwives are included and budgeted for. Promotion of condom use is prioritised both as a method in family planning and in the prevention and control of RTI/STI/AIDS. Adolescents are recognised as an under-served group needing information on nutrition and safer sexual behaviour. Violence against women is also recognised as a public health issue requiring attention in public facilities. The importance of making services gender sensitive is also highlighted.²

Third, HPSP developed 41 indicators to monitor programme performance, a large number of which are geared to improving reproductive health and monitoring gender equity. The indicators include gender-disaggregated figures on life expectancy, malnutrition and infant mortality; the maternal mortality ratio; deliveries by skilled attendants; use of antenatal care; met need for emergency obstetric care; total fertility rate; contraceptive prevalence rate, including use of modern contraception; contraceptive discontinuation rate; STI prevalence in selected groups; extent of access for women, children and the poor to the Essential Services Package; awareness, especially by women, of selected interventions; hospitals certified as women-friendly; functional Unified Behaviour Change Communication (UBCC) strategy that includes gender considerations; user satisfaction; improved gender awareness across MOHFW; changes in male/female staffing ratios in MOHFW; nutritional status of women and children; and budget allocation for the Essential Services Package.¹⁶

Fourth, community and stakeholder participation was integrated in programme development, implementation, monitoring and review. For the first time, stakeholders are given a systematic role in MOHFW's policies and programmes. HPSP provides for establishment of national and local stakeholder committees to improve transparency and accountability of the health system and mandates stakeholder consultation as part of its annual programme review. It also provides for the preparation and promulgation of a patient's charter of rights in all union and upazila health centres.

Fifth, HPSP recognises gender as a cross-cutting issue and addresses relevant gender concerns in many different parts, and mandates production of gender-disaggregated evidence and tools for gender analysis.

Limitation of the HPSP

Institutional capacity to carry out the reforms was weak, however. During the design phase the reformers worked on elaborating principles, activities and budget, but did not weigh carefully the institutional capacity of the government to implement the reforms. HPSP design acknowledged the institutional weaknesses of government but assumed these would be corrected through technical assistance, and that detailed strategies and institutional arrangements would be elaborated during the implementation phase. But this posed the risk that the planned activities would not be implemented on time or in the ways envisioned. Gender, community and stakeholder participation and partnership parts of the programme faced particular difficulties as the government had little institutional capacity for implementing these. During the design phase, it was primarily the donors and civil society who advocated these issues; the government did not oppose them but its attention was mainly focused on reorganisation of services and financing.

Furthermore, bipartisan political support for the reforms was tenuous, even though such support is necessary if reforms are to survive regime change. Unfortunately, it was difficult to mobilise such support given the extreme partisan and confrontational politics of Bangladesh. Several other reforms, for example, decentralisation, have also fallen victim to partisan politics.¹⁷ The preparation of HPSS and HPSP started during a period of political turmoil in which political parties were focused on election campaigns. Reform negotiations were primarily carried out by the donors and Bangladeshi government officials. Political parties and their leaders were hardly involved. Agreement on the major elements of the reforms was reached when the Bangladesh Nationalist Party (BNP) was in power (1995). The reforms survived regime change from the BNP to the Awami League (AL) in 1996; indeed, the AL government (1996–2001) enthusiastically embraced them. HPSP was finally adopted only after it got the approval of the then Prime Minister, Sheikh Hasina. This facilitated the implementation of restructuring, most notably unification of health and family planning services and sector-wide programming. But the AL's strong support inhibited the creation of bipartisan support and HPSP's future became uncertain when the AL lost power in 2001.

Third, gender equity advocates who were autonomous remained few in number, and it was the government and the donors who established the space for them to participate, not civil society organisations themselves. The continuation of their participation during HPSP implementation was therefore dependent on the will of government and donors. The HPSP did not address the strengthening of autonomous civil society, instead it gave government the responsibility for establishing spaces for civil society participation. This posed a risk to stakeholder participation, as there was no strategy to sustain civil society involvement in the case of government resistance or inertia.

Gender equity advocates were aware of this risk and wanted to develop a process outside the aegis of government to strengthen civil society advocacy on health, particularly sexual and reproductive health, but they were unable to generate any support from either the government or donors for this. The government was suspicious in case strong civil society advocacy turned critical. Donors underestimated the significance of and need for civil society advocacy. They were thrilled with the existence of strong government commitment and assumed it would be sufficient for implementing HPSP, and even those who strongly supported gender equity and civil society participation pooled and committed all their money to the government. They did not maintain any separate funding channel to support autonomous civil society advocates. The donor consortium also scaled down their office staff and did not retain any full-time professionals to work with MOHFW on gender equity and participation. When the government of Bangladesh later neglected these issues, civil society advocates were not strong enough to get them to alter their stance.

Design vs. implementation of the HPSP

There was a big gap between the design and implementation of many aspects of the HPSP. The momentum of stakeholder consultation during the preparatory phase (1996–1998) stopped when the implementation phase began, which reduced the role of gender equity advocates. In 1999, the MOHFW established a National Steering Committee (NSC) for stakeholder consultation but the Committee hardly convened a meeting and no national level stakeholder dialogue was organised during the first two annual programme reviews of HPSP. NSC did, however, facilitate the formation of 16 union and 9 upazila primary stakeholder committees on a pilot basis. The members of these committees were nominated by the community during workshops organised by MOHFW through a transparent process to ensure the participation of vulnerable users of services. Though these committees never received formal government recognition, they carried out their work with enthusiasm and performed well as local health watch groups.¹⁸

MOHFW also proposed to establish community groups consisting of MOHFW field workers, members of local government and representatives of primary stakeholders as management committees for the community clinics. But here, again, there was a gap between commitment and implementation. In most cases, these groups were formed by MOHFW without community participation, and they remained non-functional.¹⁹

Gender equity and maternal health strategies received government approval only in 2001 and are yet to be translated into operational plans. Strategies for partnerships between government and NGO/private sector services and pro-poor targeting were never developed. The institutional arrangements for operationalising gender mainstreaming and participation were unsatisfactory. The two focal points for gender within MOHFW, Gender Issues Office (GIO) and Gender, NGO, and Stakeholder Participation Unit (GNSPU) did not co-ordinate with each other and the latter was abolished in 2002.

An assessment of HPSP's performance indicators, carried out in 2002 for the joint government–donor annual programme review, presents an interesting picture. There was a steady improvement in female life expectancy from 58 to 60, a significant decline in female infant mortality from 70 to 56 and under-five child mortality from 111 to 84. Severe malnutrition declined from 20% to 12%. The most significant decline was in the maternal mortality ratio, which had remained stagnant from 1990 to 1995 (470 to 450), from 410 per 100,000 live births to 320. The percentage of women receiving antenatal care nearly doubled from 26% to 47%. However, fertility rate registered only a slight decline from 3.3 to 2.9 and contraceptive prevalence rate remained stagnant from 49% to 50%.

Protagonists of HPSP have used these statistics to claim success, pointing out that in many other reform settings, services and outcomes have actually declined. But opponents have used the same statistics to discredit HPSP, arguing that structural changes such as unification of health and family planning services have demoralised family planning officials, who lost authority under restructuring. House visits by family planning outreach workers have indeed declined. The critics argued that, as a result, progress in fertility reduction and improvement in the contraceptive prevalence rate have halted.^{20 21} Many experts, however, have noted that the total fertility rate plateaued in the early 1990s and that a further sharp decline in fertility cannot be sustained only through family planning.²²

All programme reviews of HPSP carried out jointly by the government and the donors bi-annually have acknowledged that the goals and targets of HPSP are commendable but would not be reached during the first five-year programme period. Indeed, some of the targets established in the performance indicators were considered too ambitious. But all reviews have strongly recommended that the government stay the course on restructuring. None of the successive reviews has suggested any shift in the directions of HPSP.

However, as noted earlier, the government that assumed power in 2001 stopped the restructuring processes. Halfway through implementation, HPSP was halted. The contestation between the supporters and critics of HPSP, which centred around disagreements as to whether data collected on the performance indicators¹⁶ is indicative of success or failure, has continued. It is likely that the next programme will try to find some middle ground, particularly on the issues of sector programming and unification of health and family planning services. The government has already indicated that it will continue with sector-wide programming, but a decision on proceeding with unification of services is pending the results of an evaluation commissioned by the Planning Commission.²³

Conclusion

What lessons can be drawn from the experiences of the gender equity advocates with the health sector reform process in Bangladesh? The main lesson is the need to strengthen the capacity of autonomous civil society advocates. In Bangladesh, advocates successfully engaged in the process of reform design at the invitation of donors and the government to participate. But their place in the government–donor dialogue was tenuous and ad hoc, facilitated by gender experts from international agencies. During the brief implementation phase, when the government and international agencies did not prioritise consultation with advocates, the latter could not mobilise sufficient constituency pressure to open up the process. Herein lies a major weakness of the advocates' position. In future, advocates need to develop the capacity for sustained public advocacy and forge alliances with other groups, particularly the women's movement, and to expand the base of their constituency support.

International agencies and reformers within the government can also draw some lessons from this experience. Both need to revisit the strategy of promoting reforms primarily through government channels. While government ownership and commitment is essential for implementing such reforms, civil society has an equally important role to play in proposing, debating, implementing and evaluating many aspects of the reform agenda, not least in helping to create greater public understanding of the reforms. Unfortunately, during implementation in Bangladesh, HPSP gradually became a government–donor driven programme disconnected from engagement with civil society. When it came under attack from the government and critics, civil society had largely been silenced and defence fell largely on the shoulders of donors. In future, both the government and the international agencies need to create an enabling environment for strengthening the role of autonomous civil society advocates, who can promote as well as monitor the restructuring of the health system in Bangladesh.

Acknowledgements

This article is the product of an International Women's Health Coalition-Columbia University collaborative action–research project on Health Sector Reforms in Bangladesh. A research grant from the International Women's Health Coalition enabled the author to undertake this analysis and writing.