Report: Expansion and Replication of EMAS Program Approaches

Report: Expansion and Replication of EMAS Program Approaches

Contents I. INTRODUCTION... 3 III. PURPOSE... 6 IV. METHODS... 6 V. FINDINGS... 6 a. Replication within EMAS target district... 6 b. Replication outside EMAS target district... 11 VI. Reaching MOH Priority Districts... 15 VII. Replication Beyond EMAS Provinces... 17 VIII.Conclusion... 17

I. INTRODUCTION Project Overview In 2011, USAID launched the Expanding Maternal and Neonatal Survival (EMAS) Program to contribute to reductions in maternal and newborn mortality by improving the quality of care within health facilities and strengthening the referral system to ensure efficient and effective referrals from the health center to the hospital. EMAS is a 5-year program implemented across the six provinces in Indonesia with the largest burden of maternal and newborn mortality - North Sumatra, Banten, West Java, Central Java, East Java, and South Sulawesi. EMAS targets approximately 150 hospitals (both public and private) and 300 community health centers across 30 districts and cities in in these provinces (Figure 1). Project activities are implemented in close collaboration with Indonesian government agencies (national, provincial, and local), civil society organizations, public and private health facilities, and health professional organizations. The project is implemented through a partnership of five organizations including Jhpiego (lead partner), Lembaga Kesehatan Budi Kemuliaan (LKBK), Muhammadiyah, Save the Children and RTI International. EMAS is expected to contribute to an overall decline in national maternal and newborn mortality. The project focuses on two major objectives: 1) Improving the quality of emergency obstetric and neonatal care services in hospitals and community health centers; and 2) Increasing the efficiency and effectiveness of referral systems between community health centers and hospitals. Sustainability and Expansion Approach As part of its strategy, EMAS emphasizes sustainability as well as scale up of EMAS life-saving approaches in districts and provinces outside of the EMAS target districts to maximize project impact. The project focuses on helping districts and facilities operationalize existing GOI policies and programs and directly supports all three national MOH strategies and seven MOH programs outlined in the National Action Plan for Accelerating Reductions in MMR (Figure 2). EMAS also directly supports the strategies outlined in the Indonesian Newborn Action Plan and national health facility accreditation. Tools and approaches implemented as part of the project enable facilities to more readily meet established GOI standards and expectations regarding the provision of emergency ma ternal and newborn care within facilities. Support to District Health Offices utilize approaches that were specifically designed and implemented to operationalize and optimize existing, yet poorly functioning, referral systems. EMAS approaches strengthen weak points in system, using existing local budgets, implementers, and those accountable for quality services to improve the system from w ithin. At the provincial and district levels, EMAS approaches are fully integrated into District Health Office and health facility systems. Because EMAS focuses on supporting existing GOI policies and programs in the majority of its interventions, many EMAS activities are funded using local budgets.

Figure 1: EMAS Program Focus Area Map

Figure 2: EMAS Support to MOH Strategies and Programs

Nationally, EMAS has been successful in integrating its approaches into national guidelines, including the BUK 2015 Collaborative Improvement Guidelines, which incorporate all of EMAS s referral strengthening interventions. In addition, other national guidelines in the process of being developed, reflect EMAS approaches, including national death audit guidelines, the Indonesian Newborn Action Plan (INAP) mentoring guidelines, and national puskesmas accreditation guidelines. II. PURPOSE EMAS has received strong support from Provincial and District Health Offices, who benefitted from the project by creating highly-functioning and motivated facilities and referral systems which they can draw upon to expand quality services throughout their districts and provinces. Early in the project, local governments (DHO and PHO) demonstrated interest in replicating EMAS interventions and approaches. In 2015, the MOH Direktor Jeneral directed non-emas districts to replicate EMAS approaches in MOH priority districts using Dekon funds. As a result of on-going interest and support from EMAS, many districts and provinces have used existing line items within local budgets or incorporated new line items to fund the replication of EMAS approaches. The purpose of this document is to summarize the extent to which self-funded replication has occurred and to identify which approaches and interventions have been most frequently replicated. The summary includes replication within and outside of EMAS target districts and provinces. Information included in this report includes replication that occurred through June 2016. III. METHODS To quantify the details regarding replication, EMAS developed a standardized tool to retrospectively 1 collect information about which districts and provinces funded the replication of EMAS approaches and interventions, the type of approach/intervention replicated, and the number of facilities or districts where the approach or interventions were replicated. A consultant was hired by EMAS to w ork directly with EMAS district and provincial staff (through in person and telephone interviews) to collect information to complete the standardized tool. Information collected came from a variety of sources including district and province records as well as project records 2. The standardized tool also included data regarding the amount of funds used for replication. Where possible, EMAS staff sought to obtain this information directly from government budgets or activity budgets prepared to support the replication activity. However, in some cases this was not possible. IV. FINDINGS a. Replication within EMAS target district For the purposes of this report, within district replication includes implementation of EMAS approaches and interventions in facilities that were neither targeted nor funded by the EMAS project, but that lie within the 30 EMAS-supported districts. For example, in one target district, EMAS implements activities in 10 puskesmas. In the same target district, the DHO decides to expand the 1 Information collected before October 2015 was collected retrospectively. Data collected after October 2015 was collected using routine project reporting mechanisms and transferred into the standardized tool. 2 In the majority of cases, EMAS staff were directly involved providing short term TA when districts and provinces replicated approaches and interventions.

implementation area to an additional 12 puskesmas, and uses its own APDB funds to pay for the activities to be implemented. Findings from this analysis show that 20 of the 30 EMAS-supported districts have self-funded the replication of EMAS interventions and approaches to additional facilities within their district (Table 1) 3. In total, interventions and approaches to improve the quality of care were expanded to 274 additional facilities, while interventions and approaches focused on strengthening the referral system w ere expanded to 328 facilities 4. Table 1: Within District, Self-Funded Replication, by Province* Province # of Districts that Replicated Interventions and Approaches # of Facilities where Quality Improvement Interventions and Approaches (Component 1) were implemented # of Facilities where Referral System Strengthening Approaches (Component 2) were implemented # of RS # of PKM # of RS # of PKM North Sumatra 5 3 70 43 128 Banten 2 0 40 20 50 West Java 4 ** ** ** ** Central Java 3 0 58 0 58 East Java 3 0 38 0 0 South Sulawesi 3 4 61 3 26 Total 20 7** 267** 66** 262** * Total # of facilities may be duplicated across Component 1 and Component 2. ** As of the writing of this report, data by facility was not available. Within EMAS-supported districts, district replication was funded with APBD funds, utilizing both existing and special budget line items. Variation was seen across districts in terms of how budgets w ere allocated. For example, for districts in Banten Province, funds allocated to support replication w ere provided via a special fund designated for that purpose. In Karawang, West Java, funds were included in a separate task budget, listed separately from other MCH activities. However, in all districts, budgets allocated could be used to support either within-district replication or to provide support for other activities related to EMAS, including for facilities supported by EMAS. Among the districts that self-funded the replication of EMAS interventions to non-emas-supported facilities within the district, Figure 3 below shows the total number of districts replicating specific interventions, and whether the interventions were implemented in puskesmas or hospitals. Overall, districts already supported by EMAS have most frequently funded and expanded the implementation 3 Table 1 does not include data on the number of facilities within West Java where EMAS interventions and approaches have been replicated. Within district replication in West Java districts has utilized a different approach, where by details regarding which interventions and approaches have been replicated was not readily available at the writing of this report. 4 Note that the facilities captured under Component 1 and Component 2 in this table are not mutually exclusive (i.e. the same facility may be counted twice under each component).

of interventions (within district replication) to puskesmas within their districts. This finding is consistent with expectations, as EMAS already targets and provides direct support to the majority of hospitals providing maternity services within the 30 districts. Overall, among the most common interventions replicated outside of EMAS target facilities in the 30 EMAS districts was MOUs, followed by SijariEMAS, Clinical Performance Tools and clinical mentoring. Figure 3: # of districts replicating interventions to facilities within EMAS-supported districts, by intervention type and facility type 20 15 12 11 10 15 5 10 8 15 12 11 11 3 0 3 1 2 2 1 4 3 # of Districts Replicating within Hospitals # of Districts Replicating within Puskesmas A review of replication by facility and intervention type follows a similar pattern (Figure 4). SijariEMAS w as the most replicated intervention (in a total of 308 facilities), followed by MOUs (300 facilities), Clinical Performance Tools (246 facilities) and Clinical Mentoring (222 facilities). In addition, clinical dashboards, standardized registers and emergency trolleys were all replicated in large numbers of facilities. Figure 5 shows the extent to which each province has replicated interventions (both quality of care and referral system strengthening). Overall, EMAS districts in South Sulawesi have replicated the largest number of quality of care interventions within EMAS-supported districts, followed by Central Java and North Sumatra. Within district replication of referral system strengthening interventions occurred most frequently within facilities in North Sumatra and South Sulawesi.

Figure 4: # of facilities where EMAS interventions were replicated, by intervention Audit SijariEMAS MOU Clinical Performance Tools Clinical Mentoring Clinical Dashboards Standard Register Emergency Trolley Clinical Rotations PKM 275 247 241 221 175 175 164 34 0 Hospital 33 53 5 1 4 2 3 4 5 Total 308 300 246 222 179 177 167 38 5 Audits Figure 5: # of facilities replicating interventions, by province and intervention type 300 Replication of Quality of Care Interventions 250 200 45 63 58 150 100 50 0 34 76 58 40 34 32 5 34 58 51 25 40 13 58 59 35 32 38 38 38 32 38 South Sulawesi Central North Sumatra Banten East Java Clinical Mentoring Clinical Rotations Clinical Dashboards Hospital Death Audit Clinical Performance Tools Emergency Trolley Standard Register

Replication of Referral System Strengthening Interventions 350 300 250 177 200 150 100 50 0 89 153 51 58 46 34 North Sumatra South Sulawesi Central Java Banten East Java Sijari EMAS PK (MoU) Despite efforts to gather information regarding the amount EMAS-supported districts spent on replicating interventions within their own district, it was not possible to disaggregate the funds spent on w ithin district replication of EMAS-supported interventions. Through work with the district Pokjas and other advocacy activities, EMAS worked with the DHO to ensure sufficient funding for MNH priorities. In addition, w here possible, EMAS sought to use existing district budget line items to fund activity implementation within target facilities and districts. For example, funds to host clinical rotations in hospitals were utilized from existing line items for training in district budgets. To provide context for this report, EMAS identified total district budget allocations (w here possible) for maternal and child health activities from 2013 to 2016 in EMAS-supported districts (Figure 6) 5. Four of the six provinces saw overall increases in MCH budgets from 2013 to 2016. Budgets in East Java showed an overall decline and those in South Sulawesi showed uneven trends during the time period reviewed. In East Java, the 2013 and 2014 budget in one district (Malang) was especially high as significant funds were allocated for facility renovations. It should be noted that trends in several districts and provinces show uneven growth patterns. These trends can be partially explained by the fact that some interventions required large, up-front costs (eg. renovations, purchase of an ambulance). This pattern is especially clear in West Java, where significant amounts of APBD funds were spent on the purchase of hardware for SijariEMAS in 2015 in that year. 5 *Budget figures for South Sulawesi and Banten include the sum of district budgets in EMAS-supported districts used to fund EMAS interventions only. Figures from the other four provinces are representative of the sum of the district MCH budgets in EMAS-supported districts.

Hundreds Figure 6: District Budget Trends (2013 2016) for Maternal and Child Health Activities, EMAS-Supported Districts 400,000,000 350,000,000 300,000,000 250,000,000 200,000,000 150,000,000 100,000,000 50,000,000 0 2013 2014 2015 2016 North Sumatera Banten* West Java Central Java East Java South Sulawesi* Linear (North Sumatera) Linear (Banten*) Linear (West Java) Linear (Central Java) Linear (East Java) Linear (South Sulawesi*) b. Replication outside EMAS target district For the purposes of this report, replication districts includes implementation of EMAS approaches and interventions in facilities or districts that were neither targeted by nor funded by the EMAS project. Findings contained in this section of the report are limited to replication districts that lie within EMAS - supported provinces. Section IV below will discuss the extent to which replication has occurred outside of the six EMAS-supported provinces. In total, 35 districts and cities have replicated EMAS approaches and interventions (Figure 7). West Java has the highest number of replication districts and cities (11), followed by North Sumatra (8), East Java (7), Central Java (6) and Banten (3 each). Plans to replicate EMAS interventions in Bone, South Sulawesi w ere underway, but not yet finalized at the time data was collected for this report 6. 6 Plans are underway in Bone, confirmed via Surat Edaran Bupati, to replicate clinical mentoring, Pokja and MOU in Bone.

Figure 7: Total number of replication districts and cities The extent to which each of these 35 districts and cities has replicated EMAS approaches and interventions varies. The full package of EMAS activities includes a number of inter-related interventions aimed at strengthening the quality of care within the facility and the referral system within the district. Table 2 shows the number of interventions that have been implemented in each of the replication districts and cities. Four districts/cities have implemented the ma jority of the package of interventions for both quality improvement and referral system strengthening- Sibolga, North Sumatra; Kota Tangerang and Kota Cilegon, Banten; and Kudus, Central Java. While no replication districts/cities in East Java have implemented referral system strengthening interventions, the seven replication districts/cities in the province have implemented nearly the full package of quality improvement interventions. Other replication districts/cities have, thus far, only implemented a smaller portion of the EMAS package. Table 2: Total # of Quality Improvement and Referral System Strengthening Interventions Replicated in Replication Districts/Cities Province Replication District/City # of Quality Improvement Interventions Replicated # of Referral System Strengthening Interventions Replicated 1 2 3 4 5 6 7 1 2 3 4 5 North Sumatra Banten Pak-Pak Bharat Binjai Serdang Bedagai; Labuhan Batu Utara; Batu Bara; Tapanuli Selatan; Nias Selatan Sibolga Kota Tangerang Selatan

West Java Central Java East Java Kota Tangerang Kota Cilegon Kuningan Garut; Kota Depok; Pangandaran Tasikmalaya Majalengka; Cianjua; Bekasi Kota Bogor Sukabumi; Purwakarta Pati; Batan; Pemalang; Kendal; Banjarnegara Kudus Kota Kediri; Bolongegoro; Lumanjang; Probolinggo; Gresik; Pacitan; Sumenep Figure 8 provides an overview of the extent to which individual interventions are implemented within the replication districts/cities. Of the 35 replication districts and cites, 71 percent implemented Clinical Performance Tools (25 districts/cities), 63 percent implemented Clinical Mentoring (22 districts/cities) and 46 percent implemented SijariEMAS (16 districts/cities). Among the commonly replicated interventions were also Clinical Dashboards and Death Audits, both implemented in 43 percent of districts (15 districts/cities) and Pokjas, implemented in 40 percent (14/37) of replication districts and cities. The Figure 8 also highlights the total number of facilities (represented by a black circle on the figure) within the replication district/city that have replicated the individual intervention (as applicable). A review of these interventions shows that Clinical Performance Tools, SijariEMAS, and Clinical Mentoring were implemented in the highest numbers of facilities. It should be noted that replication is still in the early stages. In many cases, replication that occurred during the period included in this review represents a portion of the replication that districts are planning in the coming months and through 2017.

# of districts replicating intervention # of facilities replicating intervention Figure 8: Total Replication Districts and Cities and Total Facilities Replicating Interventions, by intervention area 30 450 25 400 350 20 300 15 250 200 10 150 5 100 50 0 Clinical Perf. Tools Clinical Mentoring SijariEMAS Clinical Dashboard Death Audit Pokja Standard Register # of replication districts 25 22 16 15 15 14 10 9 4 3 2 1 # of replication facilities 432 168 224 89 15 78 69 78 1 Emerg. Trolley MOU MKIA Civic Forum Clinical Rotation 0 # of replication districts # of replication facilities

Similar to trends seen within district replication, puskesmas comprise the majority of facilities where replication has occurred in replication districts and cities (Figure 9). Overall, puskesmas comprise approximately two-thirds of the facilities where replication has occurred in non-emas districts and cities. Figure 9: Number of facilities implementing EMAS-supported interventions in replication districts/cities, by facility and intervention type 450 400 350 300 250 200 399 150 100 163 116 50 0 33 61 52 64 54 49 25 24 20 35 43 15 1 0 Hospital Puskesmas District priorities and budget availability influences what is replicated and where. Funds used to pay for replication come from a variety of sources, including district APBD budgets, Dekon funds allocated by the province and in some cases funds from hospitals (BLUD). It was not possible to disaggregate the funds spent directly on replicating EMAS interventions and approaches with a high-level of accuracy and therefore data on budget allocations have been omitted from this report. V. Reaching MOH Priority Districts The provincial level plays several roles in supporting replication including setting province-wide priorities for maternal and newborn health, providing financial resources, and facilitating mentoring across districts. The Indonesia MOH has also designated nine provinces (including 49 districts) as priority provinces. The six EMAS target provinces are included among the nine MOH priority provinces and 26 of EMAS target districts are among the 49 MOH priority districts. Table 3 provides an overview of the coverage of EMAS within these priority provinces and districts. EMAS has strong links at the province level and has put significant effort into promoting the replication of EMAS interventions and

approaches, especially in the GOI priority districts. Of the 23 priority districts not originally targeted by EMAS, 16 have already replicated EMAS interventions using government funding sources. All priority districts in North Sumatra, West Java and Central Java are EMAS target districts or EMAS replication districts. Table 3: Coverage of EMAS within GOI Priority Districts Province Total # of GOI Priority Districts # of Priority Districts Supported directly by EMAS # of Priority Districts Replicating EMAS using Government Funds # of Priority Districts where no EMAS Intervention has been implemented North Sumatra 9 4 5 0 Banten 4 2 0 2 West Java 10 5 5 0 Central Java 12 7 5 0 East Java 8 5 1 2 South Sulawesi 6 3 0 3 Total 49 26 16 7 In the provinces where priority districts have not yet begun replicating EMAS, efforts have been underway to encourage uptake by both the PHO and by EMAS staff through province-wide promotion activities. Yet, the decision to replicate EMAS interventions lies with the district government. Discussions regarding replicating EMAS in the two remaining priority districts have been underway for some time in Banten province, but detailed plans have not yet materialized. Similarly, in South Sulawesi replication plans have been made, but had not yet been implemented as of the writing of this report. In East Java, the PHO reached out to the two remaining districts regarding their interest to replicate EMAS, but confirmation was not received until DeKon budgets were already issued. Overall however, PHOs show strong support for the replication of EMAS. In North Sumatra, the PHO has stated that EMAS mentors should be used for replication purposes, managed by Tim 21. In West Java, the PHO already manages replication schedules, provides mentors to replicate interventions in new districts, leads the facilitation of replication activities and pays directly for mentor fees and transport. In Central Java, the PHO has issued a decree regarding how the mentoring teams should be managed and has provided DeKon funds to support replication. Finally, in East Java, the PHO established the PENAKIB team which handles the replication of activities in six districts. South Sulawesi has had more challenges in terms of obtaining significant support from the PHO to replicated EMAS-supported interventions. To date, no provincial funds have been allocated. However, the PHO has provided additional funds to support more frequent MPAs within the province. In addition to budget availability, the ability to replicate activities in non-emas districts relies on the readiness and availability of mentor teams from other districts to provide support for implementation. The project has put significant efforts into ensuring a sufficient mentor assets capable of providing mentoring support both within and outside of their districts. While replication within EMAS -supported

districts is generally managed by the district and with assets based within the distri ct, replication in district and cities that were not supported by EMAS is managed by the Provincial Health Office, who draws upon the mentoring assets developed within EMAS-supported districts to support the implementation of interventions in replication districts/cities. The exact process and mechanisms for managing the mentoring process has been defined for each province and described in other EMAS project documentation. VI. Replication Beyond EMAS Provinces The most significant levels of replication have occurred within EMAS provinces. However, as a result of various national meetings where districts had an opportunity to share their experiences implementing EMAS approaches in their districts, some non-emas provinces and districts in attendance showed interest in replicating elements of the project. Some of these provinces and districts visited EMASsupported districts to learn more about the project, while others requested follow up discussions with EMAS staff. To date, eleven districts/cities in provinces have replicated EMAS-supported interventions: West Papua (Unicef-funded), Padang, West Kalimantan, Riau, West Sumatra, Kepulauan Riau, Bengkulu, Yogyakarta, DKI Jakarta, Ternate, and Tidore. Unfortunately, detailed information regarding the extent of or type of replication that occurred is not available. VII. Conclusion Although expansion and replication of approaches beyond the original target facilities and districts was not an explicit goal of the project, the design of EMAS has resulted in consi derable buy in and ow nership of the project approach. In its final year, EMAS is leaving behind essential human assets capable of mentoring others in project approaches, PHOs and DHOs who are equipped to plan, budget and facilitate the expansion of project approaches, and packages of tools, materials and other resources that make replication possible. Because the project directly supports and compliments existing MOH strategies and programs, leverages existing budgets and targets implementers and those accountable for improving health systems, EMAS project approaches have been replicated and funded using local government sources within 20 of the 30 districts and cities targeted by EMAS and adopted in an additional 35 districts and cities outside of EMAS target districts. Currently, EMAS interventions and approaches have been implemented or replicated in 42 of the 49 GOI priority districts. In addition to the replication that has already occurred, EMAS expects continued momentum to lead to additional replication of project interventions and approaches. In the six provinces that have been directly supported by the project, PHOs are leading the development of plans to replicate EMAS in new facilities and districts. Support for continued replication has also come from the national level. In 2015, the MOH Direktor Jeneral directed non-emas districts to replicate EMAS approaches in MOH priority districts using Dekon funds. In addition, EMAS was successful in integrating the package of EMAS interventions into national policies (Collaborative Improvement Guidelines) which are expected to result in further replication.