|1. Project Data:
ICR Review Date Posted:
|Bangladesh - Health Nutrition And Population Sector Program
Project Costs(US $M)
Loan/Credit (US $M)
|Health, Nutrition and Population
Cofinancing (US $M)
|CIDA, EC, KfW, Sida, DfID, UNFPA
Board Approval Date
|Health (40%), Central government administration (36%), Other social services (11%), General public administration sector (9%), General education sector (4%)|
|Nutrition and food security (25% - P)
Health system performance (25% - P)
Population and reproductive health (24% - P)
Tuberculosis (13% - S)
Child health (13% - S)|
||ICR Review Coordinator:
|Judyth L. Twigg
||Alain A. Barbu
||Christopher D. Gerrard
|2. Project Objectives and Components:|
a. Objectives:As stated in the Development Credit Agreement (DCA, p. 19), the project’s objectives were to: “(i) reduce infant, under-five, and maternal mortality and the proportion of malnourished children; (ii) eliminate the gender disparity in child malnutrition and mortality; (iii) ensure increased access to reproductive health services; (iv) lower total fertility with a view towards achieving replacement level fertility by 2010; (v) reduce the burden of tuberculosis, HIV/AIDS, malaria, and other priority diseases; (vi) initiate a system to control newer health threats and protect health risk by improving emergency services; and (vii) improve the prevention and control of non-communicable diseases.
The Project Appraisal Document (PAD, pp. 4-5) states that the project “assists the Government of Bangladesh in the implementation of its Strategic Investment Plan (SIP), 2003-2010,” and that “the main purpose of SIP (2003-2010) will be to increase availability and utilization of user-centered, effective, efficient, equitable, affordable, and accessible quality services, be it the Essential Services Package, improved hospital services, nutritional services, or other selected services.” The next sentence refers to these as “objectives,” and these are also listed as the objectives in the PAD’s data sheet and results framework (p. 33).
Since the objectives statements in the DCA and in the PAD are completely different, this Review will assess achievement of the objectives as stated in the DCA. The objectives as stated in the PAD will be used explicitly as part of the causal chain leading to the results envisioned in the DCA's objectives.
b. Were the project objectives/key associated outcome targets revised during implementation?
c. Components:The project contained three components, intended to constitute a specific set of activities within the Government’s overall investment plan for the sector:
1. Accelerating Achievement of Health, Nutrition, and Population (HNP)-Related Millennium Development Goals (MDGs) and Poverty Reduction Strategy Paper Goals (appraisal, US$ N/A; actual, US$ N/A). This component was intended to support the delivery of a package of essential services. The Essential Services Delivery (ESD) was to focus on reduction of maternal mortality; reduction of neonatal mortality; reduction in childhood morbidity and mortality; improvement in nutritional status, particularly of adolescent girls, pregnant and lactating women, and children; reducing fertility to replacement level; and reducing the burden of tuberculosis and malaria and preventing and controlling HIV/AIDS. Activities were to include a very broad array of public information campaigns, social mobilization and counseling, training of providers, provision of equipment, drugs, and laboratory supplies, provision of immunizations and micronutrients, and voucher programs to increase demand for services.
2. Meeting Emerging HNP Sector Challenges (appraisal, US$ N/A; actual, US$ N/A). This component was intended to support the development of policies and strategies for emerging challenges, with a focus on reduction of injuries and implementing improvements in emergency services; prevention and control of major non-communicable diseases (NCDs); urban health service development; and improvement of the HNP response to disasters. Activities were to include information campaigns, advocacy and counseling programs, legal assistance programs, various kinds of training, establishment of emergency care facilities in high-risk locations, improvements in screening and diagnosis for various conditions, improvements in intersectoral liaison and coordination, and development of a risk management plan.
3. Advancing HNP Sector Modernization (appraisal, US$ N/A; actual US$ N/A). This component was intended to address three key HNP reforms:
a. Public health sector management and stewardship capacity, focusing on improving institutional and personal skills for better planning and monitoring; improved budget management through a medium-term budgetary framework process; reform management; improved aid management; development of proper contract documents and management of contracts with private and non-government organization (NGO) providers; information management; and development of alternative financing mechanisms. Also part of this component was a step-wise delegation of responsibility to promote decentralization and local-level planning.
b. Health sector diversification, through building Ministry of Health and Family Welfare (MOHFW) capability to become an active service purchaser in partnership with NGOs and private providers. The pattern of service provision was to be adjusted over time, with increasing use of contracts and commissions for NGOs to provide primary and secondary care in areas where they had a comparative advantage, and for private providers to provide secondary and tertiary services for poor people where they could do so cost-effectively and with high quality.
c. Stimulation of demand for HNP services, by improving the sector’s image and giving greater attention to effective communication, education, and information strategies for key health programs, and expanding demand-side financing, including piloting of several demand-side financing schemes.
d. Comments on Project Cost, Financing, Borrower Contribution, and DatesProject Cost: Neither estimated nor actual costs are provided by component. Total project cost at appraisal was estimated at US$ 4.3 billion.
Financing: The project was intended to finance a specific set of activities within the Government’s overall sector investment plan through a Sector-Wide Approach (SWAp), with the Bank and other partners (United States Agency for International Development, Asian Development Bank, Canadian International Development Agency, United Kingdom Department for International Development, the European Commission, the Government of Germany, the Government of Japan, the Netherlands Ministry of Foreign Affairs, the Swedish International Development Cooperation Agency, the United Nations Children’s Fund, the United Nations Fund for Population Activities, and the World Health Organization) contributing to implementation of the Government’s 2003-2010 Health, Nutrition, and Population Strategic Investment Plan. The Bank was to finance US$ 300 million of total anticipated plan costs of US$ 4.306 billion over the 2003-2010 time period. The Bank’s contribution was pooled with funds from seven of these partners into a multi-donor trust fund (MDTF), which the Bank administered. The total amount of this MDTF, including the Bank’s US$ 300 million contribution, was planned at US$ 687.9 million. Non-pooled partner contributions were planned at US$ 512 million. The actual disbursed amount was US$ 293.4 million from the Bank credit and US$ 387.8 million from the other seven contributors to the MDTF. The ICR does not provide actual total costs for implementation of the overall sector investment plan.
Borrower Contribution: As this was a SWAp, the Government contributed to the overall sector investment plan. The Government’s expected contribution over the 2004-2010 time period to the overall sector investment plan was US$ 2.726 billion (PAD, p. 62). The ICR does not provide the Government’s actual contribution to the plan. The project team subsequently clarified that the Government's actual contribution was US$ 3.1 billion.
On October 22, 2009, the project was restructured to modify several indicators, reallocate some funds from program activities that were progressing inadequately to increased pro-poor activities, and to extend the project’s closing date by one year from December 31, 2010 to December 31, 2011 to ensure full use of remaining project funds.
On July 30, 2010, the amounts to be contributed by the Canadian International Development Agency and the Government of Germany were increased, and the amount from the UK Department for International Development decreased.
On March 2, 2011, funds were reallocated between disbursement categories.
|3. Relevance of Objectives & Design:|
a. Relevance of Objectives:Relevance of Objectives is rated Substantial. At the time of appraisal, key issues faced by the sector included a wide gap in health conditions between rich and poor, inadequate reliance of the population on the public health sector (with most spending out-of-pocket, and on private and/or traditional sources of care), poor regulation of quality of care, serious governance issues in the sector, changing epidemiology with increased incidence of injuries, accidents, HIV/AIDS, Hepatitis B and C, cancer, and cardiovascular disease, and poor service to marginalized groups (especially poor women and children from tribal populations, people with disabilities, the elderly, adolescents, and HIV/AIDS patients). The Government’s Sector Investment Plan and the objectives of the project were substantially relevant to these issues. The project remains substantially relevant to Bank strategy, with the current 2011-2014 Country Assistance Strategy stressing inclusive growth and stronger governance through improved social service delivery. Sector-wide approaches with other partners to improve delivery of health, nutrition, and population services are explicitly endorsed in this Strategy (p. ii). The statement of objectives in the financing agreement was completely different from that in the PAD.
b. Relevance of Design:Relevance of Design is rated Modest.The ICR (p. 7) notes that, while the choice of a SWAp that financed a specific set of activities was appropriate, a broader SWAp (financing a time slice of the entire investment program, rather than financing a specific set of activities) may have laid a better foundation for more far-reaching sector dialogue. Indeed, the choice of such a hybrid approach between a project and a sector-wide program led to serious implementation difficulties and delays resulting from the need to apply the SIM's fiduciary requirements to a very large investment program and the Government's weak capacity in that regard. The use of the MDTF mechanism to pool funds was an “efficient and practical” way to align funding from a large number of development partners (ICR, p. 16).The PAD (pp. 33-51) contains a detailed results framework linking the project’s activities to expected outcomes. However, during implementation significant shortcomings in the design of these activities became apparent. The project’s Program Implementation Plan contained 38 Operational Plans that were poorly integrated. The activities under the third component designed to shift the Ministry of Health and Family Welfare from being a provider to purchaser of health services weighed heavily on the scarce managerial resources of the Ministry, and were made even more challenging by political transitions that took place in 2006 and 2009. Human resource constraints (borne largely of a structural divide between the Revenue and Development parts of the government budget) were not taken adequately into account during project design, resulting in failure to create necessary posts, frequent unfilled vacancies, and ultimately newly built facilities failing to provide planned services due to shortages of needed manpower.
|4. Achievement of Objectives (Efficacy) :|
Attribution: The pooled funds to which the Bank contributed accounted for only part of the overall sector program, and therefore the observed outcomes relied on a number of factors both within and outside the pooled fund. These include funding from other sources (Government revenue, plus funding from other development partners outside the pool), as well as non-health sector factors that contribute to achievement of many of the project’s objectives. In terms of health sector funding, the ICR (p. 18) points out that the majority of the MOHFW’s revenue funding was allocated to salaries, with little flexibility in the short term, and therefore the contribution of the pooled funding appears to have been substantial in terms of the proportion of key inputs it financed (new initiatives, drugs, etc.). However, some key development budget interventions of the MOHFW were outside the purview of the project, including a project on revitalization of community clinics (ICR, p. 41). According to the main text of the ICR, it is reasonable “to assume that the HNPSP has contributed significantly” to observed outcomes in the sector (ICR, p. 18), within the context of the Bank’s and the pooled funding’s overall support to the Government’s program; but comments on the ICR by the Ministry of Planning’s Implementation, Monitoring & Evaluation Division (IMED, ICR, p. 41) state that “in view of its limited boundary, it will be unrealistic to assign the national health outcomes to effects of the HNPSP.”
Other elements of the ICR raise caveats about the quality of data available to assess progress. According to comments by one development partner, there is reason to be concerned that reported data come largely from the Government’s own systems, and “the lack of third party verification or triangulation of data and results continues to be a concern deepened by the recognition of overall ‘weak internal controls’ and limited progress in monitoring and evaluation in the sector” (ICR, pp. 25, 50). Also, although precise data are not provided, one development partner cites “considerable private sector health provision in Bangladesh, some of this by public providers themselves,” raising questions about the extent to which observed results can be attributed to the public sector through the SWAp (ICR, pp. 25, 50). The project team subsequently clarified that the majority of the data on project outcomes are from Demographic and Health Surveys (DHS) and the Multiple Indicator Cluster Survey, and that the Government takes its information from those surveys. Additional data for this Review has been added from the 2011 DHS.
One development partner also points out that the revision of indicators and targets in 2009 translated into “the lowering of expectations concerning the program results. If one had not done this, the assessment would have been not so good” (ICR, p. 49).
Outputs Relevant to All or Most Objectives:
The MOHFW finalized a National Health Policy that was aligned with the goals established by the Second National Strategy for Accelerated Poverty Reduction. A regulatory framework for pharmaceuticals was established. Overall, according to the ICR, “there has been limited progress in implementation” of health sector modernization, “due in part to the setting of over ambitious expectations that did not sufficiently take into account political economy considerations” (p. 29).
The Ministry of Local Government, Rural Development and Cooperatives (MOLGRDC) contracted with NGOs to implement health services in urban areas. Under these arrangements, MOHFW provides required commodities such as contraceptives and TB drugs. MOLGRDC is currently in the final stages of development of a strategy, in coordination with MOHFW, for urban health services.
Overall, the ICR provides little detail on the project’s specific outputs, making it difficult to construct a plausible causal chain linking the project’s activities to observed outcomes. The project team subsequently clarified that the project financed a slice of the Government's overall sector program, making it impossible to disaggregate project-specific outputs from the entire range of health sector activity during the time period of the project.
Outputs related to the availability and utilization of effective, user-centered services: The percentage of total government expenditures allocated to the MOHFW increased from 6.5% in 2004 to 6.7% in 2007/2008, not meeting the target of 10%. There is still chronic underspending of the MOHFW development budget, even though the program took “several steps” (not specified) to expedite fund release (ICR, p. 32). The unspent portion of the development budget fluctuated during the project period: 27.6% in 2003/2004; 17.2% in 2004/2005; 13.6% in 2005/2006; 25.2% in 2006/2007; 17.1% in 2007/2008; and 26% in 2008/2009. According to an evaluation performed by the Ministry of Planning’s Implementation, Monitoring & Evaluation Division (IMED), 80% of 490 patients surveyed reported that services provided at government health facilities had improved over time. The same IMED evaluation conducted focus groups with patients, members of local government bodies, private physicians, teachers, community youth leaders, and farmers. These focus groups, according to the ICR (p. 39), indicated that the project’s interventions had resulted in “marked improvement” in the quality of health services provided through Upazila Health Complexes, the availability of doctors, nurses, and equipment, and the supply of essential drugs to patients (especially to women, children, and the poor). However, these focus groups also reported shortcomings in equipment available to health workers and supervision at health facilities.
Outputs related to the availability and utilization of efficient services: According to the ICR (p. 28), major policy decisions on decentralization and hospital autonomy were not reached due to the changing political situation. Pilots on management and autonomy were carried out in six district hospitals and 14 Upazila Health Centers. Pilot local-level planning was carried out in six districts and their Upazilas, and 2009 budgets reflected those pilots. The percentage of MOHFW expenditure on medical and surgical requisites at the district level and below increased from 9% in 2004 to 67% in 2006/07 (there was no target). The percentage of MOHFW expenditure at the Upazila level and below declined from 51% in 2004 to 47% in 2008/09, not meeting the target of greater than 50%.
Outputs related to the availability and utilization of equitable, affordable, and accessible services: The percentage of total MOHFW expenditure allocated to the 25% poorest districts was 15% in 2006/2007. No baseline or more recent data are available. The target of 40% appears not to have been achieved. The focus groups conducted as part of the IMED evaluation reported shortcomings in access to health care for the poor, specifically, that “there should be more attempts to take health services to the door-steps of poor people” (ICR, p. 39).
(i) Reduce infant, under-five, and maternal mortality and the proportion of malnourished children, and eliminate the gender disparity in child nutrition and mortality: Modest achievement. Targets set under this objective were derived from the Millennium Development Goals, adjusted to the project's timeline. Many targets were not met, although there was progress toward targets on almost all indicators, with reason to expect that this progress will continue. It is also important to note that observed outcomes are likely attributable to many factors beyond the scope of the project.
The percentage of children under one year of age who were fully immunized increased from 73% in 2003 to 80% in 2011, not meeting the target of 85%. The 2011 DHS reports that 83% of children age 12-23 months had been fully immunized by their first birthday. The percentage of newborns protected at birth against tetanus increased from 86% in 2003 to 91.3% in 2010, not meeting the target of 95%. The 2011 DHS states that 89.9% of newborns were protected at birth against tetanus.
The percentage of children age 1-5 receiving Vitamin A supplements in the previous six months increased from 81.8% in 2004 to 92.2% in 2011, exceeding the target of 90%. There was a less than 10% difference in progress between the poorest and richest wealth quintiles between 2004 and 2010 in vitamin A supplementation (for children 9-59 months) and full immunization coverage (for children under 2).
The percentage of children under 5 with symptoms of acute respiratory infection who sought care from a trained provider increased from 74.6% in 2004 to 82.7% in 2010. No target was specified. The 2011 DHS reports that the percentage of children under five with symptoms of acute respiratory infection in the preceding two weeks who had sought care from a health facility or health provider decreased from 37% in 2007 to 35% in 2011.
A demand-side financing pilot (a maternal voucher scheme for pregnant women) was implemented in 33 Upazilas. In 2010, 64% of pregnant women who were targeted by a voucher scheme delivered their babies using a skilled birth attendant, exceeding the target of 60%.
The percentage of births attended by skilled personnel increased from 13.4% in 2004 to 26.5% in 2011, not meeting the target of 50% (or a 2009 revised project target of 28%). The 2011 DHS reports that this percentage increased from 16% in 2004, to 21% in 2007, to 32% in 2011. The percentage of antenatal care provided by medically trained providers increased from 48.7% in 2004 to 56.0% in 2010, not meeting the target of 75%. The 2011 DHS reports that the percentage of live births in the preceding three years for which women received at least one antenatal care visit from a medically trained provider increased from 50.5% in 2004, to 53.4% in 2007, to 54.6% in 2011.
The National Nutrition Program area-based community nutrition program was expanded to 63 new Upazilas.
The maternal mortality rate declined from 322/100,000 in 2001 to 194/100,000 in 2010, exceeding the target of 240/10,000.
The neonatal mortality rate declined from 41/1000 in 2004 to 32/1000 in 2010, nearly meeting the target of 30/1000. The 2011 DHS also reports neonatal mortality at 32/1000.
The under-five mortality rate declined from 88/1000 in 2004 to 56/1000 in 2010, nearly meeting the target of 52/1000. The 2011 DHS reports under-five mortality at 53/1000, essentially reaching the target.
The infant mortality rate declined from 65/1000 in 2004 to 45/1000 in 2010, not meeting the target of 37/1000. The 2011 DHS reports infant mortality at 43/1000, not meeting the target.
The percentage of children under five (age 6-59 months) who were underweight declined from 47.5% in 2004 to 38.6% in 2010, not meeting the target of 34%. The 2011 DHS reports that 36% of all children under five were underweight, more closely approaching the target. The percentage of children under five (age 24-59 months) whose growth was stunted declined from 43.0% in 2004 to 35.5% in 2010, not meeting the target of 30%. The 2011 DHS reports that 41% of all children under five were stunted, again not meeting the target.
The ICR does not provide data related to gender disparities. The project team subsequently provided the following additional tabular data, demonstrating a narrowing of the gap between boys and girls from 2004-2007 along most indicators, and for Vitamin A supplementation, girls surpassing boys in 2006 and 2010. The 2011 DHS indicates that the infant mortality rate for girls in was 37/1000, and that for boys 43/1000, further narrowing the gap. The under-five mortality rate as reported by the 2011 DHS, however, was 50/1000 for girls and 57/1000 for boys, representing a widening of the gap between 2007 and 2011. The 2011 DHS reports that 59% of boys and 60% of girls age 6-59 months were given Vitamin A supplementation in the preceding six months, essentially indicating gender parity. The 2011 DHS gives figures of 42% for females and 41% for males for stunting under age five, and 39% for females and 34% for males under age five for underweight. This represents a maintenance of gender parity for stunting, but an increase in the gap between boys and girls for underweight.
(ii) Increase access to reproductive health services: Substantial achievement.
The ICR reports no outputs related to this objective. The project team subsequently added that the government implemented programs to address and strengthen local-level planning, including family planning.
The percentage of births attended by skilled personnel among the lowest two wealth quintiles increased from 3.3% in 2004 to 11.8% in 2010, exceeding the target of 10%.
The percentage of antenatal care provided by a medically trained provider for the lowest two wealth quintiles increased from 24.9% in 2004 to 40.3% in 2010, meeting the target of 40%.
The contraceptive prevalence rate (modern methods) increased from 47.3% in 2004 to 54.1% in 2010, not meeting the target of 60%. The 2011 DHS indicates that use of modern methods increased from 47.3% in 2004, to 47.5% in 2007, to 52.1% in 2011, approaching but again not meeting the target. The percentage of eligible couples or women on long-lasting birth control methods remained essentially stable, at 7.2% in 2004 and 7.4 in 2010, not meeting the target of 9.3%. The 2011 DHS reports that 11.2% of women used injectables and 0.7% used intrauterine devices, surpassing the target.
(iii) Lower fertility with a view toward achieving replacement-level fertility by 2010: Substantial achievement. It is important to note that other factors, such as increasing age of marriage, were likely contributors to observed outcomes.
The total fertility rate (TFR) declined from 3.0 children per woman in 2004 to 2.5 in 2010, not meeting the target of 2.2. Replacement-level fertility was not achieved. However, the 2011 DHS gives the TFR as 2.3, essentially meeting the target.
(iv) Reduce the burden of TB, HIV/AIDS, malaria, and other priority diseases: Substantial achievement.
The TB case detection rate increased from 46% in 2004 to 74% in 2010, essentially meeting the target of 75%.
The percentage of districts with disease surveillance reports increased from 52% in 2004 to 95% in 2011, meeting the target.
100% of districts were using Directly Observed Treatment Short-Course (DOTS) by project closure.
The TB cure rate increased from 85% in 2004 to 92% in 2010. The ICR gives targets of 85% (Data Sheet) and 95% (p. 54).
Although no outputs on malaria were reported, the ICR's economic analysis includes data that the malaria death rate per 1000 declined from .0053 in 2003 to .0034 in 2010.
The project team later explained that an HIV epidemic feared possible at the time of preparation did not materialize, and therefore no HIV-related outcomes were monitored or reported.
(v) Initiate a system to control newer health threats and protect health risk by improving emergency services: Modest achievement.
Standard operating procedures were put in place for mass casualty management and disaster mitigation for field-level health personnel. However, according to the ICR (p. 28), “the quality of any response may be hampered by the often fairly low level of institutional capacity and service quality in the health sector.”
Four one-stop crisis centers were established in four divisional Medical College Hospitals to address violence against women. The manual that guides management of violence was updated in 2007, and training in this area was initiated for health service personnel. According to the ICR (p. 28), there are “limitations regarding appropriately trained personnel which constrain the availability of some services.”
(vi) Improve the prevention and control of non-communicable diseases: Modest achievement.
Two strategies were developed and endorsed: a Strategic Plan for Surveillance and Prevention of NCDs 2011-2016, and a National Cancer Control Strategy and Plan of Action 2009-2015. An NCD risk behavior survey was conducted.
Tobacco usage among men and women over age 15 increased from 20.9% in 2004 to 23% in 2009 for smoking tobacco, and from 19.7% in 2004 to 27.2% in 2009 for smokeless tobacco, not meeting the target of 15% for both smoking and smokeless tobacco.
According to the ICR (p. 28), progress in this area was “limited.”
Efficiency is rated Modest.
The PAD (pp. 97-108) calculated a Net Present Value of US$ 610 million (5 year) and US$ 2.975 billion (10 year) and an Internal Rate of Return of 21% and 51% for 5 and 10 year horizons, based on a set of assumptions that were largely achieved during implementation (though data are not available for many of the assumptions) (ICR, pp. 36-37). The ICR did not repeat the calculations at closing, and therefore little data are available on the efficiency of investment of Bank resources, resulting in a Modest rating.
The project’s primary health care interventions, including immunization, antenatal care, increasing access to skilled birth attendants, etc., were among those generally considered to be cost-effective. The project’s poverty focus successfully directed funding toward those most in need (ICR, p. 18). Although overall per capita financing of the public sector for health is low (US$ 4-5/person/year), a 2010 Bangladesh Public Expenditure and Institutional Review found that high value is obtained from those expenditures compared to other countries in the region (ICR, p. 19).
The 38 “silo-like” Operational Plans were identified as an important problem by the 2008 mid-term review, producing inefficiencies such as constructed facilities sitting idle because of poorly coordinated procurement of equipment/drugs and allocation of human resources (ICR, p. 9). A 2008 Bank survey of procured equipment found “considerable wastage of project resources due to uncoordinated central procurement” (ICR, p. 46). Throughout the project, expenditures were higher in the last quarter than in the first two quarters, likely caused by delays in release of funds to cost centers (ICR, p. 33). Furthermore, the bifurcation of services between the Directorate of Health and the Directorate of Family Planning, from the center down to the lowest service delivery points at the field level, led to duplication and avoidable waste of financial and human resources (ICR, p. 44). The mid-term review commented on the negative effects of this bifurcation and recommended improved coordination, but the Ministry of Health and Family Welfare continued to support the division. The project team subsequently stressed that these structures are an unavoidable part of the government's operational arrangements, and that the team worked consistently with the Bank and the government to mitigate these challenges in the health and other sectors. These efforts included development of a single work plan and harmonization of the development and revenue budgets.
a. If available, enter the Economic Rate of Return (ERR)/Financial Rate of Return at appraisal and the re-estimated value at evaluation:
* Refers to percent of total project cost for which ERR/FRR was calculated