K8707 Maternal Health Malawi CS · A Major Factor in the High Maternal Mortality Ratio Is Delayed or Lack of Access to Good-Quality Health Services at Birth The maternal mortality

1

How to Improve Maternal and Neonatal Health in Malawi

CASE STUDYDecember 2016

PROJECT DATA

PARTNER ORGANIZATION:Ministry of Health of Malawi, the Norwegian Government, the German Federal Ministry of Economic Cooperation and Development, KfW

ORGANIZATION TYPE:Financial cooperation

DELIVERY CHALLENGES:Lack of understanding of importance of assisted delivery, and insuffi cient infrastructure at health care facilities.

DEVELOPMENT CHALLENGES: High maternal and neonatal mortality rates. How to increase numbers of pregnant women delivering at district health care facilities and staying 48 hours, and improve the quality of care they receive.

SECTORS:Maternal and neonatal health

COUNTRY:Malawi

REGION:Africa

PROJECT DURATION:August 2011–December 2017

PROJECT TOTAL COST: €17 million

CONTACTS

CASE AUTHORS:Jörg Freiberg, Linda Berk

PROJECT EXPERT:Fannie Kachale

TABLE OF CONTENTS

Introduction . . . . . . . . . . . . . . . . . . . . . . 3

Contextual Conditions . . . . . . . . . . . . . 3

Tracing the Implementation

Process . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Results Achieved . . . . . . . . . . . . . . . . . 14

Lessons Learned . . . . . . . . . . . . . . . . . 14

How the Case Study Informs the

Science of Delivery . . . . . . . . . . . . . . . 15

In Brief • Development Challenge: Despite recent improvements in mother and child

health in Malawi, maternal and neonatal mortality remains high. • Program Solution: Reward health care facilities for improving service, and

reimburse pregnant women for the costs associated with delivering in a facility and staying there for 48 hours postpartum.

• Program Results: Care improved, and the number of women delivering and staying in facilities postpartum increased.

Executive SummaryTh is case examines an ongoing results-based fi nancing initiative in Malawi that aims to reduce maternal and neonatal mortality. Th e initiative, being implemented in four rural districts that are home to some 2 million people, is an important one because, despite improvements in maternal and neonatal health, thousands of Malawi women still die in childbirth every year and many preventable deaths occur among newborns.

MALAWI

Global Delivery initiative

2

A Major Factor in the High Maternal Mortality Ratio Is Delayed or Lack of Access to Good-Quality Health Services at BirthThe maternal mortality rate continues to decrease from 984 deaths per 100,000 live births in 2004 through 807 in 2006 to 634 in 2015 (WHO/UNICEF/WB/UNFPA, 2015). There has also been a steady increase in the proportion of births attended by skilled health personnel from 55.6 percent in 2000 to 87.4 percent in 2013. The increase can be attributed to government policy of promoting institutional deliveries by changing the role of traditional birth attendants from delivering children to referring expectant mothers to health centers and hospitals. On the other hand, there is a significant difference with respect to proportion of births attended by skilled health personnel between rural and urban areas. All the cities of the country have proportions above 90 percent while all districts except Dedza have proportions below 90 percent.

However, there is a significant difference with respect to proportion of births attended by skilled health personnel between rural and urban areas. The difference can be attributed to good access to more improved health facilities and services in the cities than districts. As a result of years of underfunding, most Malawian health care facilities in rural areas face shortages of trained health care staff, low staff motivation, poor adherence to professional standards, stock-outs of essential medicines and consumables, lack of functioning equipment, and badly maintained buildings. The low-quality care these facilities provide is partly responsible for the high maternal and neonatal mortality ratios. In addition, lack of better transport systems in some rural areas makes it difficult for people to access health facilities (MoF 2014, 38, 39).

The Results-Based Financing Initiative Includes Both Supply- and Demand-Side IncentivesTo address these problems, the Results-Based Financing Initiative for Maternal and Neonatal Health (RBF4MNH) intervenes on both the supply and the demand sides. On the supply side, it pays health care facilities to improve the quality of care, providing financial rewards to both individual staff members (to motivate them personally) and facilities (to improve working and living conditions). On the demand side, it encourages women to give

birth in district facilities and to stay there for 48 hours postpartum by reimbursing them for some of the out-of-pocket expense they incur by giving birth at a facility.

The supply-side component rewards facilities for improving service. It measures performance based on a jointly agreed upon list of performance indicators and targets that are included in the quality and performance contracts (QPCs) negotiated with each facility. The process of designing these agreements involved consultations with the government and stakeholders at the zonal, district, and facility levels. The proposed maximum payments are 15–25 percent of the total salary envelope of each type of facility. The funds are divided between individual payments to individual staff members and rewards to be used by the facility or team as a whole. Decisions regarding the use of the facility portion of the reward are made by the facility itself.

On the demand side, all pregnant women between 15–49 who register for antenatal care services and are certified as residents of the facility’s catchment area are eligible for the program. They receive the payment (of about €6) if they give birth at a participating facility and remain there for 48 hours postpartum. Remaining under professional observation for 48 hours is important because it helps staff detect and address complications (especially infections and bleeding), most of which occur during the first 48 hours.

Initial Results Are Positive, Albeit Well Short of the TargetThe initiative reached only about a quarter of the 41,000 women it had hoped to reach in 2014. However, the number of women delivering at results-based financed facilities increased to 61,000 women between October 2014 and March 2015 (Options 2015). There have been various reasons for the difficulties in the first implementation phase, such as fear that the demand-side payments would encourage women to get pregnant for the purpose of receiving cash reimbursements; high staff turnover rate existing at local and national levels, affecting implementation as orientation of new staff took time; difficulties in arranging local bank accounts (Wilhelm 2016).

According to the impact evaluation, results-based financing has successfully changed the incentives driving some behaviors in the health sector. A clear shift in demand for maternity health services was observed. The increase in length of stay at RBF facilities was considerably larger. Health care workers were motivated by substantial changes


3

in their working environment, particularly in respect to infrastructure, equipment, and supplies. It has also affected the attitude and behavior of health staff, who are now more proactive in addressing maternal and neonatal mortality and put more emphasis on the quality of care. Teams gained skills and knowledge about how to solve delivery challenges themselves or with the support of higher-level authorities. They learned that their own actions, behaviors, and attitudes directly affect the quality of care. Teamwork at facilities improved, and staff increasingly see themselves as drivers of change (see Brenner and De Allegri 2016, 8).

IntroductionMalawi has seen remarkable declines in infant and under-five mortality and increases in the proportion of deliveries assisted by a skilled staff. Despite these improvements, however, maternal and neonatal mortality ratios remain among the highest in the world. In 2015, 634 women per 100,000 live births died in childbirth or as a result of a pregnancy-related illness (WHO/UNICEF/WB/UNFPA 2015). In 2013, 22 infants per 1,000 live births died (neonatal mortality rate) (UNICEF/WHO 2015).

A major factor in the high maternal mortality ratio is lack of or delayed access to good-quality health care service at birth. Specific problems include the following:

• insufficient infrastructure • lacking or nonfunctioning equipment • lack, or stock-outs, of essential medicines and

consumables • insufficient number and capacity of staff, especially in

rural areas • lack of staff motivation, commitment, and problem-

solving attitude • insufficient support and supervision by district health

management teams • limited outreach, as a result of the shortage of

midwives and fuel • lack of an adequate referral system and ambulance

services • short observation time after delivery (most women

who do deliver in facilities leave shortly after doing so).

A variety of factors make it difficult for women to reach facilities:

• distance and terrain • lack of resources and ability to pay for transport and

to remain at a facility after delivery

• lack of decision-making power within the household • traditional beliefs and lack of awareness of the

importance of medical care • poverty, youth, unmarried status.

This case study examines the ongoing Results-Based Financing Initiative for Maternal and Neonatal Health (RBF4MNH) in Malawi. As part of their long-standing commitment to support the health sector there, the governments of Germany (through the German Development Bank [KfW]] and Norway are financing the RBF4MNH (initiated by Federal chancellor Angela Merkel and Norway’s former prime minister Jens Stoltenberg), which aims to reduce maternal and neonatal mortality by improving the quality of care and increasing the share of institutional deliveries.

To increase the quality of medical care, the initiative rewards institutions that achieve certain outputs as well as pregnant women who deliver at district facilities. To motivate health care workers and institutions, it provides financial rewards to both individual staff members and facilities that meet preset quality targets. To incentivize women to deliver at facilities and remain there for 48 hours post-delivery, it provides conditional cash transfers that cover the costs of transport to and from the facility as well as the stay itself. The initiative also finances small-scale construction and the purchase of basic equipment.

The case study addresses three questions:

Question 1: Were implementers able to incentivize district hospitals, health centers, and district health management teams to improve the quality of care and increase access to good-quality maternal and neonatal health services?

Question 2: Were implementers able to incentivize women to deliver at health care facilities and remain there for 48 hours post-delivery?

Question 3: What delivery bottlenecks emerged, and how did implementers cope with them?

Contextual ConditionsMalawi is one of the world’s poorest countries, with average annual per capita income of about $350 (2012). The high fertility rate (5.7) has caused rapid population growth, which puts enormous strain on basic services, land, water, and other resources and undermines


4

economic growth. The high prevalence of HIV/AIDS (the ninth worst in the world) reduces output and is a major factor behind the country’s low life expectancy.

Despite Improvements, Maternal and Neonatal Mortality Ratios Remain High, and Access to and the Quality of Care Are PoorMalawi has achieved important improvements in maternal and child health. Infant and under-five mortality have declined dramatically since the 1990s, and the proportion of assisted deliveries increased from 57 percent in 2004 to 87 percent in 2014 (MoF 2014, 39). Malawi appears to have met the child mortality Millennium Development Goal (MDG) target of reducing the under-five mortality rate by two-thirds between 1990 and 2015 (UNDP 2015, viii).

These improvements notwithstanding, the maternal mortality ratio remains one of the highest in the world, at 634 maternal deaths per 100,000 live births in 2015 (WHO/UNICEF/WB/UNFPA 2015). The country is not expected to have met the MDG target of reducing the ratio by three-quarters between 1990 and 2015 (UNDP 2015, viii).

Although 91 percent of women gave birth at a health care facility in 2015 (NSO 2016, 21), they did not always have access to good-quality care. Systematic challenges such as a lack of qualified personnel; difficult working conditions, including heavy workloads and poor staff housing; the scarcity of functioning equipment; the irregular supply of essential drugs; and limited clinical and managerial competence contribute to the high maternal mortality ratio.

Access to services is another challenge, particularly in rural areas. About 20 percent of Malawi’s people live more than eight kilometers from a public health care facility. Seasonal difficulties travelling (e.g., rains that make roads unusable), lack of transport for pregnant women, and the costs associated with institutional delivery make giving birth in a facility unattractive or impossible for many women.

Since 2004 the Government Has Addressed Health Care Needs through a Sectorwide Approach (SWAP)The Health Sector Strategic Plan for 2011–16 (MoH 2011), which is aligned with the MDGs, guides implementation of health interventions in Malawi. It emphasizes increasing coverage of high-quality services defined in

the Essential Health Package (EHP)—the list of services, including maternal care, that should be provided free of charge—and improving the equity, efficiency, and quality of EHP services.

District councils and their committees, especially the district executive committee, manage the delivery of health services, working through district health offices. Traditional authorities and structures at the community level also play important roles. They are involved through area development committees and health care facility advisory committees.

The Government Launched the Presidential Initiative on Maternal Health and Safe Motherhood in 2012When Joyce Banda became Malawi’s first female president, in 2012, she made reducing the maternal mortality ratio and improving the health of children and mothers a top priority. The Office of the President and Cabinet launched the countrywide initiative in 2012, which seeks to achieve rapid and sustainable improvements in Malawi’s maternal health indicators by investing in infrastructure, expanding the professional health care workforce, and mobilizing traditional authorities as agents of change. It is working in tandem with the Ministry of Health to construct maternity waiting homes in each of Malawi’s 28 districts, train up to 2,000 community midwives, and transform community interactions with the public health sector through the creation of a chief ’s council.

Malawi’s Health Care System Is Highly Dependent on DonorsMalawi’s health care delivery system consists of government facilities (63 percent), faith-based facilities of the Christian Health Association of Malawi (CHAM, 26 percent), and some private for-profit providers.

External resources (donor aid) accounted for more than 60 percent of total health expenditure between 2004 and 2010. It fell dramatically to a level of 58 percent in 2011, as a result of concerns about governance (World Bank 2016). It declined again in 2013, after civil servants and high-ranking political figures were found with vast sums of money whose provenance they could not explain. The scandal, which led to calls for the president’s resignation, had devastating consequences for the health sector, affecting 37 percent of the funding provided by donors as pooled financing under the Common Approach


5

to Budgetary Support.1 However, funding by external resources has been increased between 2012 and 2014 up to a level of 74 percent mainly as discrete program financing. As a consequence of the decrease in budget support, district-level funding and funding for hospitals from the central government during the 2013/14 budget year was irregular and partial, resulting in a deterioration in health service.

Tracing the Implementation ProcessTo reduce maternal and neonatal mortality and strengthen the Malawian health system, in 2009 the Ministry of Health of Malawi requested a feasibility study undertaken by Options and intended to use both results-based financing (RBF) demand and supply measures. As part of their long-standing commitment to support the health sector in Malawi, the governments of Germany and Norway mandated the German Development Bank (KfW) to lead the initiative. In August 2011 the Ministry of Finance of Malawi and KfW signed a financing agreement for a contribution to the health SWAP for the initiative that made up to $10 million available, with equal contributions by the German and Norwegian governments. The RB4MNH has been implemented by the MoH (Ministry of Health) since April 2013 as a pilot program across selected basic health facilities in four districts.

Programmatically the RBF4MNH initiative is divided into three major components: a) a basic infrastructural upgrade of the 17 facilities including architectural modification to extend available space, replacement and provision of essential equipment, and maintenance of critical supply chains necessary in sustaining minimum standards; b) a supply-side RBF intervention consisting of quality-based performance agreements between the Reproductive Health Unit (RHU) of MoH on the one side and targeted facilities and District Health Management Teams (DHMT) on the other; and c) a demand-side CCT intervention consisting of monetary compensations to pregnant women for the recovery of expenses directly

1 The Common Approach to Budget Support (CABS) is a joint mechanism developed by a group of donors to provide budgetary support to Malawi. CABS funding is provided on an assessment of public finance management reforms, economic management, good governance, and a good track record of human rights in Malawi (see for more details Tavakoli and Hedger 2010), for an analysis of CABS impact on health expenditures, see Alavuotuniki 2015.

related to accessing and staying at target facilities during and at least 48 hours after childbirth. In early April 2013 performance agreements were signed between the MoH, the 17 health facilities, and the four respective DHMTs.

Concomitantly with the supply-side rewarding scheme, a demand-side scheme was introduced in July 2013. The demand-side intervention consists of Conditional Cash Transfers (CCTs) targeted toward pregnant women living in the areas of the 17 selected facilities. The CCTs are intended to support women a) to present to the intervention facilities for delivery in time; and b) to remain under skilled maternal care observation at these facilities for the initial 48-hour postpartum period.

The RBF supply-side incentives and demand-side CCTs are expected to increase the number of facility-based deliveries through their combined effect on improved quality, through changes in providers’ motivation and proactivity, and through the removal of financial barriers to access.

The second phase of the initiative included a second contribution by KfW (of €5 million). It was increased in January 2014 by another €5 million. Phase 2 aimed to achieve substantive consolidation and geographical expansion. Implementation of both phases took place between September 2012 and August 2015.

The Malawian Ministry of Health contracted with Options Consultancy Services Ltd. to design and manage the program. Options works closely with Broad Branch Associates, a company with expertise in innovative financing mechanisms in developing countries. The two partners are drawing on international and local expertise to build the capacity of the Ministry of Health and the Reproductive Health Directorate within it to implement results-based financing (RBF) approaches.

The RBF4MNH initiative introduces rewards based on measurable results, working on both the supply side (health care providers) and the demand side (pregnant women). The underlying principle is that demand should be increased only where existing supply is adequate. RBF4MNH also includes limited immediate investments in infrastructure upgrading and rehabilitation and essential equipment, to bring the maternity services in the four initial test districts (Mchinji, Dedza, and Ntcheu in the Central West zone and Balaka district in the South East zone) up to an appropriate minimum standard of care.

The key defining feature of a conditional cash transfer program is the direct link between the payment of incentives and a change in a specific behavior. In the


6

RBF4MNH initiative, cash transfers are not meant as an incentive but as reimbursement for out-of-pocket expenditure. All pregnant women 15–49 who register for antenatal care services and are certified as residents of the facility’s catchment area are eligible for the program. They receive the payment (of about €6) if they give birth at a participating facility and remain there for 48 hours postpartum. Remaining under professional observation for 48 hours helps staff detect and address complications of newborns and mothers (especially infections and bleeding), most of which occur during this period. Payments to mothers cover the costs associated with delivering at a health care facility, including transport to and from the facility, food, and essential childbirth items such as blankets, etc. The amount received depends on the distance to the facility.

The four participating districts are home to about 2 million people, the majority of them poor. Instead of working with a large number of facilities in each of the selected districts, the Reproductive Health Directorate and other partners at the Ministry of Health at the central and district levels indicated a preference for the World Health Organization–recommended Concentration Model of One Comprehensive Emergency Obstetric Newborn Care (CEmONC) per 500,000 people and four Basic Emergency Obstetric and Newborn Care (BEmONCs)

per CEmONC. Phase 1 was therefore limited to three or four facilities and one hospital per district.

The baseline assessment in the project districts focused on four areas of service provision: leadership and management, resource management, enabling of a safe and healthy environment, and service provision. The results were critical to the selection of the facilities in the four districts and were used continuously during the implementation phase.

Later assessment reports focused on improvements in maternal and neonatal care and utilities (water, electricity supplies, and sewerage systems). In order to initially bring all selected facilities in a position to be able to provide basic health services (which then would be assessed and rewarded in the supply-side RBF scheme), critical inputs had to be ensured and upfront improvements financed.

Lack of Maintenance and Repairs Is a Serious ProblemThe facility assessment revealed that one of the biggest delivery challenges is the lack of maintenance and repairs over a long period of time (pain point). Water, electricity, and waste disposal systems were malfunctioning, most facilities were cluttered with broken and unused equipment, and lighting and loose electrical wiring created serious fire safety issues.

Figure 1 Timeline of the Results-Based Financing Initiative for Maternal and Neonatal Health

04/2010 Feasibility study undertaken by Options06/2010 Appraisal of the planned Results-Based Financing Programme for Maternal and Newborn Health 04/2011 Signing of agreement on delegated cooperation between KfW and the Norwegian Ministry of Foreign Affairs08/2011 Signing of financing agreement between Ministry of Finance of Malawi and KfW 11/2011 – 08/2012 Design Phase prepared by Options09/2012 Submission of Inception Report for phase 1 by Options10/2012 – 12/2014 Phase 110/2012 – 03/2013 Preparation Phase04/2013 Performance Agreements signed04 – 10/2013 Facility Rehabilitation and supply-side component09/2013 registration and payment of cash transfer to eligible women begins10/2013 First verification exercise of supply-side RBF through peer review mechanisms; First incentive payments to facilities10/2013 Adaptation of the reward system (rewards paid for reaching 80 percent of targets instead of reaching 100 percent) 04/2014 Second incentive payments to facilities 09/2014 Submission of Inception Report for phase 2 by Options10/2014 – 12/2015 Phase 2 12/2014 Third incentive payments to facilities


7

As a result of these findings, the initiative decided to combine standard contracting procedures (contracting a local architect who subcontracted with builders and engineers and managed and oversaw the works) with an integrated rehabilitation strategy. The integrated rehabilitation strategy involved maintenance teams based at district hospitals led by a maintenance supervisor (refinement). Engaging district-level maintenance teams, which were supported by coaching, also promoted ownership and transferred skills. Use of maintenance supervisors helped monitor progress at construction sites, which proved to be of great benefit for the overall program.

Investment in infrastructure and equipment in phase 2 is following the approach taken in phase 1 while benefiting from the lessons learned. The initial structural and infrastructural inputs provided to RBF facilities independent of performance contributed heavily to the acceptability of the RBF intervention. However, infrastructure upgrades and equipment support were often delayed, which resulted in facilities being given additional financial start-up portions to obtain needed supplies directly. Lengthy procurement processes and the rainy season, which made some construction sites inaccessible, caused delays in the standard contracting procedures. Moreover, all financial transfers were complicated initially by the fact that facilities could not manage their own bank accounts.

The facility rewards have been used to improve facility infrastructure and resource availability. This RBF4MNH intervention led to greater autonomy for facility staff. Staff were now able to use the facility portion of rewards to buy items really needed by their facilities and facilities are now less reliant on DHMTs to resolve problems and obtain.

Timely Emergency Referral Is CriticalTimely emergency referral from BEmONCs to CEmONCs is often a matter of life or death for both mothers and their babies. Most lethal bleeding from complications during delivery and septic shock caused by prolonged labor and intrauterine infection can be prevented by timely referral to CEmONC facilities, where blood transfusions and Caesarean sections can be performed.

Improving the referral system is an area where further investment would bring substantial benefits. In the initiative’s geographical area, most of the fleet of ambulances is old, and many ambulances are in a state of

disrepair, making them unusable for long-distance travel. The initiative originally planned to provide one additional ambulance per district together with a maintenance plan included in the reward system. However, as other actors in this field plan to improve the referral system, the initiative may refrain from making further contributions in this area.

Quality and Performance Contracts Reward Facilities for Meeting Performance TargetsThe initiative’s supply-side component rewards facilities for improving service. It measures performance based on a jointly agreed list of performance indicators and targets that are included in the quality and performance contracts (QPCs). The selection of performance indicators and targets on which to pay incentives, including the nature and amount of the financial rewards, were was based on existing maternal and neonatal health indicators used by the MoH. As the implementation progressed, the Reproductive Health Directorate felt the indicators may be too ambitious, as one had to attain 100 percent of the target or the beneficiary would not receive any of the reward. District Health Management Team (DHMT) members stated that they had little involvement in the initial selection of indicators, but were allowed to give input into possible changes to the indicators and performance targets for the second phase of implementation.

The process of designing these agreements involved a series of consultations with the government and stakeholders at the zonal, district, and facility levels. Together they identified a range of indicators that could contribute to the objective of ensuring safe, good-quality delivery and neonatal care. The challenge was to narrow the list down to indicators that could be regularly measured and verified. The decision was made to start with a small number of indicators and to consider additional indicators in later phases. Figure 2 shows the targets and achievement of them in the first two cycles of the initiative.

Payments for performance are conditional on performance on the quality targets. In phase 1 the indicators for facilities comprised a set of weighted indicators plus a quality indicator that deflated the cash envelope to which the group is eligible based on performance on a quality measurement tool. DHMTs work toward a set of indicators that focus on the specific


8

requirements of facilities to improve their quality of care. They are incentivized to maintain and increase the number of deliveries as well as sufficient supplies of drugs and consumables across the district (to ensure that DHMTs continue to support nonparticipating facilities as well).2

All RBF schemes involve the signing of QPCs by the paying agent and the incentive recipient.3 The better a facility performs against each target, the higher the reward. QPCs are concluded for a term of one year.

Performance payments are disbursed in cycles (initially every six months, since October 2014 every quarter). Every cycle new targets are specified and new QPCs drafted using the previous cycle’s performance as the baseline. Tripartite agreements between the RBF4MNH Initiative, the district commissioner, and the recipient facility or DHMT should build ownership of the initiative at the district level and support the integration of the RBF approach into government structures from the outset.

2 The initiative monitors the risk that RBF could divert attention from other services.

3 Paying agents specify the responsibilities of each signatory, including what, to whom, and when the facility/DHMT must report; the targets they must reach and how they will be measured and verified; and penalties for administrative errors and fraud.

The eligible payment levels for facilities were determined by calculating a salary envelope for each type of recipient (BEmONC, CEmONC, DHMT). The proposed maximum payments are 15–25 percent of the total salary envelope of each type of facility. The funds are divided between payments to individual facility staff (the staff portion) and rewards to the facility or team as a whole (the investment portion).4 Each facility makes its own decisions regarding the use of the investment portion. Giving some of the reward to facilities allows them to address factors that affect job satisfaction, such as working conditions, in an effort to make staff agents of change.

District Health Offices Coordinate the InitiativeIn all four districts, district health offices coordinate activities and play an important role in the management of the initiative at the district level. The district commissioner

4 BEmONCs allocate 70 percent to the staff portion. CEmONCs allocate 60 percent to the maternity team (with 70 percent of that amount going to individual staff members) and 40 percent to the hospital as a whole. DHMTs allocate 60 percent to the staff portion.

Figure 2 Achievement of Quality and Performance Contract (QPC) Targets in Cycles 1 and 2

0

20

40

60

Perc

ent o

f qua

lity

targ

et a

chie

ved

80

100

120

Equipment

reported

HIV text a

nd treatm

ent

Magnesium Sulvate use

d

Infection preventio

n

check

list u

sed

Audits death

Up-to-d

ate card

s

Vitamin A given

Pregnancy-Specif

ic str

ess

questionnaire

done

Uteroto

nic use

d

HMIS reporte

d

RBF reporte

d

Partograph fil

led

Cycle 1 Cycle 2

Source: Options.


9

and the district council administer both the supply- and demand-side payments, which are coordinated by the RBF subcommittee. District RBF subcommittees are also responsible for linking the initiative with the activities and initiatives of the district council. They provide oversight to the DHMTs and health care facilities and inform the district executive committee on the status of the initiative.

As the design process progressed, it became apparent that additional human resources were needed for implementation. Through the DHMTs, the government appointed a district coordinator for each district and a central level coordinator. They support implementation of both supply- and demand-side activities and are critical to integrating the initiative into government policies and plans. They are also responsible for the initial trainings and capacity building for RBF measures.

Supportive Supervision and Continuous Training and Guidance Are CriticalThe baseline assessment identified serious gaps in knowledge and skills of standards for quality service provision, particularly regarding infection prevention and overall management. To address them, the initiative provided a series of trainings for DHMTs and health care workers early on. It also educated core staff of the Reproductive Health Directorate/Ministry of Health about the details of the supply- and demand-side approaches chosen.

RBF is a completely new approach in Malawi. As in other performance-based financing projects in Africa, the process of introducing and embedding it in the country’s health system and structures requires intensive support. Sensitization and continuous training and guidance are particularly important to build knowledge and create a solid understanding of the RBF concept and processes to enable strong management and ownership of the initiative. Efforts involve all key players at the district level, including the DHMTs, district councils, and traditional authorities.

The Reproductive Health Directorate, with technical support from Options and in cooperation with the DHMT, implemented a system of supportive supervision that covers recording, reporting, and monitoring data and interpreting performance and quality information. These support activities started before the RBF scheme was launched, in order to explain the value of RBF

incentives to facility staff, acquaint them with the new procedure, and motivate and support them in complying with the expectations set by the QPCs.

The second cycle revealed the need for intensive support, supervision, and mentoring in RBF and maternal, neonatal, and child health in order to improve and maintain performance, particularly given the high turnover of staff, decentralization, and political changes that occurred (pain point). Quarterly district review meetings were established for improving communication and addressing challenges. These meetings were useful to share information and suggestions, which allowed the implementation team to make necessary changes to the intervention. Supervision and feedback mechanisms were used at district and facility levels as another way to discuss gaps and strategies to improve performance. DHMTs stated they used a variety of methods to assist the facilities, which included coordination meetings, identifying problems in achieving the indicators, and, in one case, coming up with specific action plans after looking at patient exit questionnaires to identify causes of client complaints.

Despite challenges, adoption of the RBF concept began to occur not only at the district level, but also at the highest levels of the MoH and government. It took DHMTs several months to understand the concepts and to be trained in the additional tasks required for the RBF intervention, but DHMTs are now translating this knowledge into action to achieve the targets through better supervision and management.

Regarding performance feedback and recognition of effort, health care workers reported being motivated by the performance targets and feedback inherent in the PBF scheme. Health workers, especially following revisions after year one, very well accepted the indicator set. Health workers were motivated by its function as a clear and focused reminder of good clinical practice. They further described how the feedback and the interest in and appreciation of their work motivated them to improve their performance. Health care workers reported substantial improvements in supervision and performance feedback, both in quality and in quantity, which were perceived as helpful and supportive rather than controlling. In the first year, the indicator set and verification process were perceived as nontransparent and unfair by a number of health care workers. The established peer review verification process led to “foul play” in most health care workers’ perceptions. Changes in processes


1 0

(e.g., indicator set revisions, external verification teams, quarterly verification) during the second project phase were thus welcomed by all respondents, and led to greater acceptance and perceived usefulness of the verification results. However, health workers reported that the external performance verification process had not taken place since March 2015. While facilities continued to receive bonus payments based on the last conducted performance verification, health workers were not happy with this as they missed an opportunity for feedback, which would have permitted them to further improve their performance. As a result, many health care workers perceived negative change in the area of feedback and recognition of effort in the second year (Brenner and De Allegri 2016, 92,95,96).

Additional Training and Capacity Building Were Required after the 2014 ElectionsAfter the elections in May 2014, the newly elected district councillors were given powers to make decisions about RBF issues (before the election, district commissioners had the final say in their districts). Most of the newly elected councillors had not been exposed to the RBF4MNH initiative. Additional orientation and capacity-building activities were therefore needed.

RBF district coordinators and desk officers were replaced in 2014. These changes set back progress, as new staff required training and coaching and institutional memory was lost.

A strategic plan was developed to orient new members of parliament, chiefs, and ward councillors as well as new members of the RBF subcommittees in the four districts. As of December 2014, all councillors in the four districts had been oriented on RBF4MNH activities. These efforts continued in 2015 to ensure a good level of knowledge, understanding, and support at the district level.

Further implementation and management systems have been enhanced in phase 2 through the establishment of an RBF steering (or oversight) committee, as well as technical subcommittees. At the district and central levels, steering and management structures, including the existing RBF subcommittees, have been strengthened. The process of establishing and strengthening these committees is supported by the national office of the initiative, working closely with the Reproductive Health Directorate as principal implementing agency.

Phase 1 of the Demand-Side Component Fell Well Short of Its TargetThe CCT program implementation began in late 2013, but faced a number of difficulties and delays in some of the study areas. The number of women in RBF facilities who received cash payments increased every month between January and June 2014. However, phase 1 fell well short of achieving its target of reaching 41,000 women, providing financial support to only 9,150 women by the end of phase 1 (September 30, 2014).

The registration and the verification system faced some problems, partly due to the complexity of the system itself and partially due to the fact that the health surveillance agents (HSA) did not cooperate as expected. In respect to the verification process, health care providers at facilities faced challenges in so far as health surveillance assistants – responsible for verifying enrolled women’s residence in a facility catchment area – were not satisfied with the amount of rewards attributed to them from the supply-side RBF arm, which in turn led them to neglect their tasks in relation to the CCT program verification process, resulting in late arrival or disappearance of CCT cards (see Brenner and De Allegri 2016, 82, 85).

Most women enrolled in the CCT program were unfamiliar with or even confused regarding the purpose of CCTs and CCT eligibility criteria. Some women affirmed that not all eligible women are being enrolled into the program, especially those presenting to ANC clinics during their first pregnancy trimester. Women also perceived facility personnel as biased in enrollment in favoring enrollment among friends or relatives. Many stakeholders also critiqued the CCT eligibility criteria wherein only women residing in catchment areas of RBF facilities can be enrolled.

Also, the majority of women enrolled in the CCT program did not receive cash reimbursement after facility delivery. Difficulties were experienced in establishing bank accounts and preparing health care facilities to handle cash delayed payments to mothers (inflection point). Adoption of and compliance with the cash transfer program proved to be very challenging for health staff, who were not familiar with handling and accounting for cash. Corruption was also perceived to play a role wherein the health care providers did not pay out reimbursement amounts indicated on a woman’s CCT card in full.


1 1

In other instances, women were told that their CCT cards went missing during the verification process.

Intensive coaching, training, supervision, and audits were required to improve the performance of the demand-side component. Cash transfer operational procedures, including procedures for documentation, were strengthened. RBF desk officers, RBF district coordinators, district accountants, officers-in-charge of maternity, and nurses were trained to register women, verify their eligibility, and budget for conditional cash transfers by month and quarter. Facilities learned how to submit their quarterly budgets based on the projected number of deliveries. District authorities learned how to process the requests and forward them to the RBF4MNH office in Lilongwe in a timely manner. These initiatives resulted in a steadily rising of women receiving cash transfers: from 4,010 at June 2014 to 7,459 as the end of 2014 (Options 2014, 39) and 16,283 as of March 2015 (Options 2015). The steadily rising number of beneficiaries was encouraging. The cost of transport proved to an important factor in the decision to deliver in a health facility, particularly since costs have increased as a result of inflation and the devaluation of the kwacha.

The Proportion of Women Remaining in the Facility at Least 48 Hours after Delivery IncreasedThe major shifts in patterns of care seeking for delivery have been occurring within Malawi in the past decade. Malawi experienced an unprecedented increase in rates of skilled birth attendance (from 56 percent in the 2004 MDHS to 90 percent in the 2015 MDHS. However, only 39 percent of women received a checkup in the first two days after birth (MDHS 2015–16, 24). The average time spent at the facility after having delivered increased from 1.4 days to 2.3 days between April 2013 and May/June 2015 (Brenner and De Allegri 2016, 64). Considering that utilization of facility-based delivery and postnatal care increased at the national level, it is not surprising that there are no differences to be observed between RBF4MNH facilities and other facilities.

Facility Deliveries Increased Substantially between 2004 and 2010—But Problems RemainAt the national level, the proportion of live births that occurred in a health facility increased from 69 percent

(2004) to 92 percent (2015/16). According to the MDHS 2015–16, 39 percent of women reported having perceived a postnatal care of their health within 2 days of delivery (NSO, 24). The impact evaluation of RBF4MNH reported that RBF4MNH did not report any increase in the population-based rate of health service utilization for directly targeted services (i.e., facility-based delivery) or indirectly concerned services (i.e., antenatal and postnatal care services). This means that on a population-base level, the proportion of women using services did not increase substantially more in intervention compared to control catchment areas. The absence of an effect on the rate of facility-based delivery is largely due to the fact that utilization rates were already very high (approaching 90 percent). However, the actual number of facility-based deliveries increased substantially in intervention facilities over time and RBF4MNH produced also a significant effect on women´s length of stay at the facility after delivery. There was also a remarkable increase in the average time women arrived prior to the actual day of delivery in RBF4MNH facilities, from 3 days in 2013 to 4.9 days in 2015 (Brenner and De Allegri 2016, 64).

Staff Also Found Complying with Standard Operating Procedures DifficultThe majority of health workers are convinced that they generally examine their patients adequately. Health workers working at RBF facilities attributed these changes in routine patient care to the fact that a number of shortcomings have been minimized since the start of the RBF. They felt compelled to provide more systematic patient examinations since facilities were explicitly assessed based on the content and documentation of examinations. However, a detailed clinical assessment might not always be possible due to high workload. Clinical assessments are therefore kept short.

Qualitative findings of the impact study, however, suggest that health workers feel insufficiently familiar with the clinical guidelines relevant to their clinical performance. Frameworks to support health worker competence and adherence to clinical standards are often limited or insufficient. Rotating staff from other departments are sometimes difficult to orient, do not consistently adhere to maternal care guidelines, and are perceived as resistant to familiarizing themselves with prevailing standards of midwifery given that they will


1 2

leave maternal care services due to the rotation system. Incomplete patient assessment performed by colleagues also poses challenges to ensuring a sufficient level of quality and patient safety. When patient documentation is incomplete or missing, sometimes health workers document findings in medical charts of partograph forms without actually having checked or examined the patient. Given that existing medical records and clinical documentation are perceived as incorrect, incomplete, and unreliable, risk assessments and identification of likely complicated cases remained of poor quality.

Important Changes to the Initiative Were Made in Phase 2As part of the preparation of phase 2, RBF4MNH teams, the Reproductive Health Directorate, health care workers, and DHMTs reviewed and analyzed data from phase 1 to determine how best to improve the QPCs. Based on this analysis, they made major changes to the agreements (adaptation).

Sharing Results Across Districts Promoted Cross-LearningSince the second reward cycle (October 2013–March 2014), results have been shared at district reward and review meetings and comparisons made between districts. This process promoted cross-learning across districts and inspired improvements by weaker teams, based on the sharing of examples of good practice (feedback loops).

The Reward System Was Adapted after the First Cycle, after Too Few Facilities Met Their TargetsBefore October 2013 the first verification exercise of supply-side RBF was achieved through peer review mechanisms. They were followed by a first reward review meeting and reward distribution in October 2013. The average reward in the first cycle was just 18 percent of the facility total salary envelope, and performance on reaching a number of targets was extremely low. In this cycle targets were either reached by 100 percent or not reached (an all or nothing system). In October 2013 the system was changed so that rewards were paid for reaching 80 percent of targets (adaptation). The change was made to motivate staff and avoid entering into a

vicious cycle of low rewards–low purchase capacity–low performance–low rewards, which might have threatened the aim of the initiative to improve quality of care.

To fine-tune the program, Options highlighted some lessons learned from other African countries that introduced performance-based financing:

• Mobilizing people requires informing them of how much they could have earned but did not (opportunity costs).

• People need to be convinced that funds will truly flow. • It takes time to understand, react, and put plans into

action.

After the first cycle, participating facilities and DHMTs understood what was at stake. Although participants were disappointed over the low level of rewards, they were motivated to perform better and earn more in the next cycle. Indeed, rewards increased in cycles 2 and 3 (figure 3).

The increases in rewards can be attributed to the following factors:

• improved understanding of the initiative at all levels • efforts to improve the quality of care and achieve the

targets set • The desire to earn more, having seen the potential

reward envelopes for cycle 1 • assurance that funds actually flow.

The third change was an increase in the overall reward envelope. The reward envelope was increased for both facilities (by 20 percent) and DHMTs (by 40 percent),

Source: Options.

Figure 3 Results-Based Financing Rewards Earned in Cycles 1, 2, and 3, by Type of Institution

0

1020304050

607080

Rew

ards

ear

ned

by fa

cilit

ies

(per

cent

age

of to

tal r

ewar

d en

velo

pe)

90100

Cycle 01 Cycle 02 Cycle 03

BEmONC CEmONC DHMT


1 3

to take account of inflation and rising costs in Malawi and to include a wider group of medical personnel in reward payment at the DHMT level (district maintenance team and HMIS officers).

In phase 2, verification and payment of rewards was changed from twice a year to four times a year (adaptation). Although the change increased the financial and administrative cost of the verification process, it improved motivation, enhanced learning, and facilitated course correction, allowing DHMTs to respond more quickly to identified weaknesses.

A review of the indicators and targets underlined the need for regular in-depth review and revision (refinement). These changes enabled facilities and DHMTs to earn rewards that are larger and better reflect the improvements made in delivering care. Changes included the following:

• Setting of a ceiling of the institutional delivery rate (IDR): Given the very high IDR of some participating BEmONC health facilities, a ceiling of 85 percent has been set, above which the RBF4MNH will no longer incentivize further increases. This has been done in order not to encourage overcrowding in the facilities and in order to not penalize those facility teams that are unable to increase the IDR. This means that the BEmONC health facilities have been divided into two groups: (1) those with an IDR of 85 percent, and above, and (2) those with an IDR of below 85 percent, which will continue to be rewarded an increase in the IDR.

• Flexible performance-based indicators: During the first two reward cycles for phase 1, health facility teams often achieved important improvements in performance, but still missed their 80 percent targets. Consequently, these facilities also missed out on the portion of the reward allocated to this indicator and this had a demotivating effect, particularly where great effort has been made and the target is only just missed. Therefore, during phase 2 a revised list of indicators and weights has been introduced to the districts and RHD/MoH, which will reward facility teams for actual progress made. The system of “one-or-nothing” target setting was replaced by one where rewards are earned according to the level of actual achievement. For example, if a facility team achieves 75 percent of the indicator, they would receive 75 percent of the portion of the reward allocated to this indicator. Under the old system, where the targets were set at 80 percent or more, the facility would have forfeited the whole portion of the reward allocated to this indicator despite achieving 75 percent.

• Change of the verification method: Qualitative indicators have been integrated into the indicator list for calculating the reward payment instead of calculating, first, the reward on the basis of quantitative indicators and deflating them by underachievement of qualitative indicators (phase 1).

• Change in the weighting of indicators: The weight of each indicator group reflects its importance for achieving the key objectives of RBF4MNH. The group of indicators measuring the quality of care is weighted with 58 percent, indicators measuring the quality of management and reporting are weighted with 16 percent, indicators measuring the quality of basic services (energy, water supply, quality of equipment) are weighted with 17 percent, the indi-cator measuring feedback mechanism of patients is weighted with 4 percent and the family planning indicator with 5 percent.

Phase Reached 61,000 WomenIn addition to the 18 facilities supported in the first phase, 10 new facilities have been selected for the second phase of RBF4MNH. The aim was to have 61,000 women deliver in RBF facilities between October 2014 and December 2015. To overcome the difficulties and delays that resulted in not achieving the targets planned for the first phase, the following procedures were established:

• To avoid the delays in the start of the infrastructure works the contracts with the design team of phase 1 were extended for phase 2. Also the District Maintenance Team was involved in the construction and rehabilitation processes.

• The cash payment was raised to reflect the sharp increase in prices since the amounts were set in 2012 (adaptation).

• Annual audits of the four district accounts and corrective action plans were established in partnership with the district authorities (feedback loops and refinement).

• The facilities were supported to better manage the cash transfer component. The finance officer of the RBF4MNH team visited districts monthly; detailed reviews of the funds flow mechanisms have been con-ducted; improved templates for monthly reconcilia-tion of the accounts were introduced; and cash books were introduced at the facility level that show all pay-ment amounts received by the facility and paid out to pregnant women.


1 4

Results AchievedThe RBF4MNF initiative has achieved good results:

• The institutional delivery rate increased from 92 percent in 2012 to 96 percent in December 2014.

• The length of stay after delivery increased considerably larger in RBF4MNH facilities than in other facilities (see Brenner and De Allegri 2016, 29).

• Among women without complications, 87 percent stayed at the facility for 48 hours post-delivery (up from a baseline of 0).

• Ninety-five percent of neonatal deaths in participating facilities and 100 percent of maternal deaths were properly audited (up from a baseline of 0).

• Eighty-two percent of pregnant women who delivered in RBF facilities with unknown HIV status were tested and managed accordingly (up from a baseline of 23 percent).

• Stock-outs of drugs and supplies decreased: The average number of stock-out days for HIV testing kits during each six-month reward cycle fell from 14 at baseline (2012/13) to about 5 in December 2014.

• Thirty-six percent of women who delivered in an RBF facility between July and December 2014 received cash for transport, up from 23 percent in January–June 2014.

Lessons LearnedRBF supply- and demand-side incentives are expected to increase the number of facility-based deliveries through their combined effect on improved quality, changes in providers’ motivation and proactivity, and removal of financial barriers to access. So, how well did the initiative perform?

Question 1: Were implementers able to incentivize district hospitals, health centers, and district health management teams to improve the quality of care and increase access to good-quality maternal and neonatal health services?

The initiative established several mechanisms to improve the quality of services at health care facilities and the overseeing DHMTs, the direct link to pregnant women. The average reward envelope achieved increased steadily, as a result of improved performance, better understanding of the approach, greater availability of critical inputs, and upfront improvements. These positive indicators are worth tracking and learning from over the next several

years, as the results of the next phases of the program become apparent.

The initiative changed the incentives driving behaviors in the health sector. In addition to the measured outputs and outcomes of quality of care and institutional delivery rate, it affected the attitude and behavior of health staff, who are now more proactive in addressing maternal and neonatal mortality and put more emphasis on the quality of care. Teams gained skills and knowledge about how to solve delivery challenges themselves or with the support of higher-level authorities. They learned that their own actions, behaviors, and attitudes directly affect the quality of care. Teamwork at facilities improved, and staff increasingly grew to see themselves as drivers of change.

Improved quality is essential to the program, but finding verifiable indicators for rewarding it is difficult. This initiative highlights the importance of monitoring, reviewing, revising, and recording the RBF approach to make the original program objectives and results framework more realistic and to increase its development impact. As districts learned about RBF, introduction of the revised approach to new facilities proceeded more rapidly and efficiently.

This initiative also shows that a one-size-fits-all design would not have been effective; the approach must be flexible to adjust to local circumstances and to the progress the program makes. The programs’ processes are not generally applicable to other contexts. Solutions must be individually crafted and, if necessary, supported by intensive technical training and advice.

Question 2: Were implementers able to incentivize women to deliver at health care facilities and remain there for 48 hours post-delivery?

The number of women who delivered safely in RBF facilities rose. Transport cost proved to be an important factor in the decision to deliver at a health facility; it has become even more important with the increase in costs as a result of inflation and the devaluation of the national currency. The aim of keeping women in the facility for 48 hours post-delivery was achieved. However, there is not yet evidence that the cash payments motivated women to deliver at facilities rather than at home. An ongoing independent evaluation is expected to answer this question.

Many women who deliver at RBF facilities still do not receive the cash payment, despite being eligible. Challenges with the flow of funds, verification of eligibility, and knowledge gaps have prevented payments from being made to all eligible women.


1 5

The demand-side approach did not target women most in need of financial support. The initial reason for introducing a cash transfer was to address financial barriers to institutional deliveries by compensating women for the out-of-pocket costs of institutional delivery. In order to introduce a more sustainable and targeted approach, the initiative plans to adjust eligibility criteria so that only women who need financial support receive it. Poverty targeting will begin in 2016. Registration and certification processes will also be reviewed and simplified.

Going forward, it would be useful to understand the impact of the demand- and supply-side components of the initiative on the power of women to decide to deliver at a facility; their traditional beliefs and awareness of the importance of institutional delivery; and the stigmatization that poor, young, and unmarried women face at health care facilities.

Question 3: What delivery bottlenecks emerged, and how did implementers cope with them?

Capacity constraints were the single greatest risk factor for the success and sustainability of the RBF4MNH initiative. Staff turnover rates were high, as a result of the reorganization of the civil service after the presidential election of May 2014, causing loss of institutional memory and experience. Employment restructuring had an adverse effect on performance, leading to delays as well as increased financial and human resource inputs to coach and train new employees, visit facilities, and conduct capacity-building workshops.

A key delivery lesson from implementation is that embedding the approach in government systems is crucial to enable continuous support; country ownership is an important precondition for success. All stakeholders should be involved in program design, and programs should be aligned with national health strategies and policies. Although the involvement of all levels of government in all steps of design is time consuming, it initiates a change of mind-set and action for change. The RBF4MNH initiative had to engage in a program of high-level advocacy in order to promote RBF within the Ministry of Health and to influence future developments in health policy. Doing so required strong advocacy, communication, and cooperation with government ministries and departments and external partners.

The initiative worked closely with the Reproductive Health Directorate of the Ministry of Health. It also garnered district- and community-level support, which brought

early gains and enhanced local ownership. However, high turnover negatively affected the ownership of RBF by district authorities, which required critical inputs such as advocacy visits and training workshops. Overall, ownership of district and traditional authorities is now considered to be very high.

There is a growing recognition in international development cooperation that RBF can be more than a health financing strategy; in the long run, it can also become a health system innovation. This mere possibility can be seen as an opportunity, such as an entry point to reform the health sector, particularly for improving the quality of health care services.

How the Case Study Informs the Science of DeliveryThe emerging science of delivery entails five elements that should be considered when analyzing a case study.

1. Focus on measurable welfare gains of citizens The program’s monitoring data show an increase in

the number of women who delivered at RBF facilities. The actual number of facility-based deliveries increased substantially in intervention facilities over time and RBF4MNH produced also a significant effect on women´s length of stay at the facility after delivery.

2. Adopt multisector, interdisciplinary multistakeholder approaches

The RBF4MNH initiative is embedded within the Reproductive Health Directorate of the Malawian Ministry of Health at the national level, Ministry of Health structures at the district level, and local government and community structures. It embarked on high-level advocacy within the scope of the SWAP and to influence future developments in health policy. With its communications and advocacy strategy, the initiative aims to increase awareness and understanding of RBF in Malawi and among stakeholders at different levels (central, district, facility, and community); to increase awareness of the initiative as a government-owned program within both Malawi and the global RBF community; to communicate in a language that is easily understood by the general public, communities, and beneficiaries in order to raise awareness; and to increase positive media coverage of RBF4MNH. Although involvement of all levels of government proved to be time consuming, it initiated a change of mind-set and more action for change.


1 6

3. Use evidence to inform experimentation and learn from, adapt, and measure results

The detailed reward verification system provides evidence of achievements by health care facilities and DHMTs. Demand-side verification is carried out by interviews with randomly selected women. Without these data, the program would not be able to validate the impact of activities or confidently allocate funds.

The program’s monitoring and evaluation system provides timely information for decision making and helps stakeholders to understand the context of RBF implementation, identifying successes and failures and accountability. It also helps monitor project progress and mitigate risks. The reward system, reward indicators and targets, and quality and performance contracts were revised based on experience during implementation and frequent interaction with participating facilities. The RBF4MNH team learned that continuous supervision and coaching improved results on both the supply and the demand sides.

4. Consider the role of change management, leadership, and learning from practitioners

The RBF4MNH initiative supported organizational change processes. It succeeded in turning healthcare facilities and DHMTs into change agents. Program achievements were regularly discussed with the Reproductive Health Directorate and shared at district review meetings. Comparisons of districts enabled cross-learning and spurred improvements by weaker teams based on the sharing of examples of good practice.

During the first years of implementation, the complexity of the approach required substantial time and resources. Since then health care workers have increasingly understood and adopted RBF concepts. Health care staff and DHMTs are now more proactive in addressing maternal and neonatal mortality, having learned that their own actions, behaviors, and attitudes directly affect the quality of care. They have taken on more leadership of the program and the overall provision of good-quality services.

Within KfW and German Development Cooperation (GIZ), this program is used to gather and analyze positive (and negative) experiences and to brand RBF within the toolkit of development cooperation instruments. However, sharing experience and learning is not always easy to coordinate given the large number and variety of stakeholders involved.

5. Be adaptive, flexible, and iterative when implementing solutions

The RBF4MNH initiative exhibited a high degree of flexibility by addressing delivery bottlenecks with several adaptations. The program’s management explicitly refers to regular review, replanning, and reporting. Moreover, the following changes has been introduced between cycles 1 and 2:

• Sharing results across districts promoted cross-learning

• Adaptation of the reward system • Increase in the overall reward envelope • Introduction of flexible performance-based

indicators and changes in the weighting of indicators

• Change of the verification system • Increase of training and awareness activities.

RBF approaches must be tailored to local circumstances and remain flexible, responding to the progress the program makes. The case study shows the importance of planning experimental components in order to identify the most effective solutions while taking local conditions into account.

ReferencesAlavuotunki, K. 2015. General Budget Support, Health

Expenditures, and Neonatal Mortality Rate: A Synthetic Control Approach. Helsinki: UNU-WIDER. See https://www.wider.unu.edu/sites/default/files / wp2015-108.pdf.

Brenner, S., and M. De Allegri. 2016. Final Results of the RBF4MNH Impact Evaluation. Washington, DC: USAID. See http://sphfm.medcol.mw/wp-content / uploads/2016/07 / Final-Results-Report-1.pdf.

MoF (Ministry of Finance, Economic Development Planning and Development). 2014. 2014 Millennium Development Goal Report for Malawi. See http://www.undp.org/content/dam/malawi/docs/general / Malawi_MDG_Report_2014.pdf.

MoH (Ministry of Health, Malawi). 2011. Malawi. Health Sector Strategic Plan 2011–2016: Moving towards Equity and Quality. See http://www.nationalplanningcycles .org / sites / default/files/country_docs/Malawi/2 _ malawi_hssp_2011_-2016_final_document_1.pdf.


1 7

MoH (Ministry of Health, Malawi), UNICEF, UNFPA (United Nations Population Fund), WHO (World Health Organization), and AMDD (Averting Maternal Death and Disability). 2011. Malawi 2010 Emergency Obstetric & Newborn Care Needs Assessment. Final Report. See http://www.healthynewbornnetwork .org/hnn-content/uploads / Malawi-EmONC -Report -June-2015_FINAL.pdf.

NSO (National Statistical Offi ce). 2016. Malawi: Demographic and Health Survey, 2015–16: Key Indicators Report. See https://dhsprogram.com / pubs/pdf/PR73/PR73.pdf.

Options. 2012. Inception Report: Results Based Financing for Maternal and Neonatal Health (RBF4MNH Initiative), 2012–2014. Lilongwe and London: Option Offi ce.

Options. 2014. Inception Report: Phase II; Results Based Financing for Maternal and Neonatal Health (RBF4MNH Initiative), 2014–2015. Lilongwe and London: Option Offi ce.

Options. 2015. Improving Maternal and Newborn Health Using Results-Based Financing, Malawi. See http://www.options.co.uk/work/improving-maternal-and -newborn-health-using-results-based-fi nancing-malawi.

Tavakoli, H., and E. Hedger. 2010. Aid Eff ectiveness in Malawi: Options Appraisals and Budget Support. London: ODI (Overseas Development Institute). See https://www.odi.org/sites/odi.org.uk/fi les/odi -assets/publications-opinion-fi les/5800.pdf.

UNDP (United Nations Development Program). 2015. Malawi Millennium Development Goals Endline Report. See http://www.mw.undp.org/content/dam /malawi/docs/general/UNDP_MW_EDP_MDG _book_fi nal.pdf.

UNICEF/WHO (UNICEF and World Health Organization). 2015. A Decade of Tracking Progress for Maternal, Newborn and Child Survival: Th e 2015 Report. Geneva: WHO. See http://www .countdown2015mnch.org/documents/2015Report / Malawi_2015.pdf.

WHO. 2015. Malawi: Neonatal and Child Health Profi le. See http://www.who.int/maternal_child_adolescent / epidemiology/profi les/neonatal_child/mwi.pdf.

WHO/UNICEF/WB/UNFPA. 2015. Maternal Mortality in 1990–2015. See http://www.who.int / gho/maternal_health/countries/mwi.pdf.

Wilhelm, D.J., S. Brenner, A. S. Muula, and M. De Allegri. 2016. “A Qualitative Study Assessing the Acceptability and Adoption of Implementing a Results Based Financing Intervention to Improve Maternal and Neonatal Health in Malawi.” BMC Health Services Research, 16: 398. See http://bmchealthservres .biomedcentral .com / articles/10.1186/s12913-016-1652-7.

World Bank. 2016. External Resources for Health (% of total expenditure on health), see http://data .worldbank .org/indicator/SH.XPD.EXTR .ZS?locations=MW.

KfW has been helping the German federal government achieve its goals with respect to development policy and international development cooperation for more than 50 years. KfW’s role in the fi eld of German development cooperation is that of an experienced bank and an institution specializing in development policy. On behalf of the German federal government, primarily the Federal Ministry for Economic Cooperation and Development (BMZ), KfW promotes and supports programs and projects that mainly involve state actors in developing and emerging economies—from their conception and execution through to monitoring their success.

To get closer to projects and programs in partner countries, KfW has regional offi ces in almost 70 countries in addition to offi ces in Frankfurt, Berlin, and Brussels.

© 2016 German Development Cooperation: KfW. All rights reserved. Th e fi ndings, interpretations, and conclusions expressed in this work do not necessarily refl ect the views of the German Development Cooperation: KfW. Th e German Development Cooperation: KfW does not guarantee the accuracy of the data included in this work.

K8707

Documents

K8707 Maternal Health Malawi CS · A Major Factor in the High Maternal Mortality Ratio Is Delayed or Lack of Access to Good-Quality Health Services at Birth The maternal mortality