What did Ethiopia do?

Integrated Management of Childhood Illness

Integrated Management of Childhood Illness (IMCI) is an integrated approach to child health that was developed by the World Health Organization (WHO) and the United Nations Children’s Fund (UNICEF) in 1996.¹ This approach focuses on improving the ability of health care providers to diagnose and treat common illnesses in countries with high child mortality, strengthening health systems, and improving family and community health behaviors through education and outreach efforts.

The government of Ethiopia collaborated with WHO on the development of the strategy and early testing² and was among its earliest adopters, beginning the work to make IMCI a major pillar in its child health strategy in 1997.

Ethiopia’s IMCI initiative focused on prevention and control of diarrhea, acute respiratory infections, malaria, malnutrition, measles, and later, follow-up for HIV/AIDS.²

Beginning in May 2000, IMCI pilot testing took place in 27 facilities across Addis Ababa; Southern Nations, Nationalities, and Peoples' Region (SNNPR); and Tigray to develop strategies to ensure successful expansion and sustained quality.² National expansion of the program began in 2001, with support from WHO, UNICEF, and other partners. According to one key informant, the rollout was phased, with such factors as disease burden and patient flow determining the order in which regions – and individual facilities – received IMCI training.

During the first years of the study period, adoption of IMCI among facilities grew slowly. As of 2006, only 29 percent of districts had at least one health facility providing IMCI, and only 39 percent of hospitals and health centers evaluated had even one health worker trained in the program.³ A government evaluation identified budget constraints, lack of trainers, and compliance-related delays as barriers to a more rapid expansion of the program.

To increase the number of health care providers trained in IMCI, the government aimed to improve the availability of preservice training by recruiting more instructors and providing financial and material support to health education institutions.² Implementation of facility-based IMCI continued, and was accompanied by a community-based program in 2010.⁴

Despite these and other efforts to scale up IMCI in facilities, the 2014 Ethiopian Service Provision Assessment Plus (ESPA+) survey found ongoing gaps in the coverage and quality of IMCI services in facilities across Ethiopia, more than a decade after the program’s introduction.

Integrated Community Case Management

The biggest gap in IMCI was its exclusive focus on facility-based services. In a country where access to hospitals and health centers was still limited, this meant significant gaps in coverage.

The first major initiative by Ethiopia’s Federal Ministry of Health (FMOH) to address such gaps during the study period was the 2003 launch of the Health Extension Program, which sought to expand health services in Ethiopia’s rural and pastoralist communities. A new cadre of health extension workers (HEWs) began to engage in health promotion and disease prevention activities, and referred cases of illness to area health centers. In most of the country, HEWs were all female, although men were sometimes included in pastoralist regions.

This initially modest portfolio of responsibilities grew quickly. A 2005–2007 pilot project demonstrated the value of integrating rapid diagnostic testing (RDT) and treatment delivery for malaria into the HEWs’ routine activities. Health facilities in the pilot district had a nearly 46 percent lower case load of malaria than in the control district, and a significantly decreased risk of malaria-specific mortality.⁵

Large-scale HEW delivery of curative services at the community level began in 2007 after the FMOH permitted them to administer treatment for both malaria and diarrhea. However, provision of antibiotics for treatment of pneumonia by HEWs would not begin for another three years, despite growing global evidence that community health workers could effectively diagnose and treat the disease.⁶

With these measures in place, Ethiopia was now ready to formally adopt an international treatment protocol known as Integrated Community Case Management (iCCM), an approach endorsed by WHO and UNICEF^2,3,4 to extend proven methods for treating three major causes of childhood mortality – pneumonia, diarrhea, and malaria – to the rural areas of low-income countries.⁵ iCCM is referred to in some countries as community-based integrated management of childhood illness, or CB-IMCI.

Because Ethiopia had a strong community health platform and several components of iCCM had already been integrated, the FMOH planned to introduce a national iCCM program within the Health Extension Program, to be implemented by HEWs at households and health posts. According to a key informant, “Ethiopia chose to implement iCCM with the Health Extension Program because it did have that platform already.”

Ethiopia’s national iCCM program was officially launched in February 2010 within the Health Extension Program. Implementation was supported by different funding sources, including the Canadian International Development Agency, the Millennium Development Goals pooled funding mechanism, and other funding commitments from the FMOH and its development partners.

Implementation of iCCM was launched in phases, based on the strength of the Health Extension Program in different parts of the country. The number of health posts providing iCCM increased rapidly, from 4,510 in 2011 to 13,500 in 2013. During this three-year period, HEWs treated a total of 290,950 cases of malaria, 323,839 cases of suspected pneumonia, and 562,044 cases of diarrhea.⁷

The FMOH rolled out a new cadre of community health workers called the Health Development Army (HDA) in 2011 to address the high workload of HEWs and the belief that many Ethiopian families were “lagging behind” in adopting a healthy lifestyle. The HDA helped strengthen Health Extension Program activities, including iCCM, by increasing uptake of services and linking community members and HEWs.

A 2012 cluster-randomized study assessed the performance of iCCM in Oromia Region. Findings showed that although iCCM implementation was generally strong in intervention areas, the program did not significantly increase care-seeking and treatment for pneumonia, diarrhea, and fever in children under age five. In addition, U5M had decreased by 12.7 percent in the iCCM implementation area, but the difference in mortality decreases between iCCM and comparison areas were not statistically significant. The sample size may have been too small to detect such decreases.⁸

In March 2013, the FMOH launched the national Community-Based Newborn Care program in collaboration with several partners – UNICEF, Last 10 Kilometers, Integrated Family Health Program, and Save the Children. The program aimed to improve antenatal, intrapartum, postnatal, and newborn care by strengthening the primary health care unit approach and the Health Extension Program. The program expanded community-based maternal and newborn health services in several areas, including management of neonatal sepsis and other serious diseases at the community level.

By 2014, 86 percent of Ethiopia’s districts had iCCM services, with nearly 30,000 HEWs and 14,500 health posts participating in the program.⁹ That year, the ESPA+ survey findings showed high availability of iCCM, but it also identified ongoing gaps in capacity and provision of services, with health personnel addressing all danger signs in only 43 percent of cases observed at health posts. Only 25 percent of children with coughs or upper respiratory illness were referred to higher-level facilities, and only 11 percent received antibiotics in alignment with guidelines. Of children diagnosed with malaria, only 17 percent received antimalarial medications.¹⁰ The proportion of sick children receiving care, however, increased remarkably since 2000.

The percentage of parents and caregivers of children under age five seeking care for basic illness also improved during this period, although overall care-seeking remained low. Children with symptoms of respiratory illness brought to a health facility increased from 18 percent in 2000 to 34 percent in 2016, and children with diarrhea taken to a health facility increased from 14 percent to 46 percent.¹¹

In response to shortages of medicines and supplies, the iCCM program transitioned from use of a “push” system (driven by projected demand) to that of a “pull” system (driven by actual client demand). Following this transition, and the integration of child health into the country’s logistics and pharmaceutical system, HEWs requested drugs and other supplies from the health center every two months based on consumption data. According to a key informant, this adaptation minimized drug wastage, improved access to drugs at the health post level, and improved equity.

The iCCM program has enjoyed a high degree of sustainability, due in large part to its integration with the Health Extension Program.⁴ By 2016–2017, the program was in 99.4 percent of health posts in rural and pastoral areas across the country.¹²

Diarrhea

Ethiopia relied on a variety of interventions to reduce diarrheal deaths during the study period. Use of oral rehydration therapy and zinc increased over time, though coverage remained low. In addition, prevention efforts to address diarrhea incidence included improvements in water, sanitation, and hygiene (WASH) and introduction of the rotavirus vaccine. Improved nutrition and economic development also likely contributed to the decline in diarrhea-related deaths.

Oral Rehydration and Zinc Supplementation

In 1978, WHO adopted oral rehydration therapy as its primary recommended treatment for diarrhea.¹³ Before the study period, oral rehydration salts (ORS) were underused in Ethiopia. The 2000 Demographic and Health Survey (DHS) in Ethiopia reported that only 13.1 percent of Ethiopian children under age five with diarrhea received ORS, one of the lowest percentages among low-income countries. ¹⁴

In 2004, WHO and UNICEF jointly reported that complementing ORS with zinc supplementation for 10 to 14 days reduces both the duration and severity of acute diarrheal episodes.¹⁵

Ethiopia decided to incorporate ORS and zinc into facility-based IMCI in 2000 and iCCM in 2010. Health-facility personnel and HEWs received training in both treatments.

The 2011 DHS found that the proportion of children with diarrhea that received ORS from a packet was 26 percent – double the rate of 13 percent in 2000. Use of ORS was still low, however, and varied widely by region. Children in urban areas with diarrhea received ORS in 45 percent of cases, compared with 29 percent in rural areas.¹⁶

The 2014 ESPA+ survey reported that approximately eight in ten children with diarrhea were still being treated with antibiotics – a higher percentage than those receiving the prescribed course of ORS (66 percent of dehydrated children with diarrhea, 69 percent of non-dehydrated children) and zinc (8 percent of dehydrated children, 16 percent of non-dehydrated children).¹⁰

These low coverage levels were the result of several factors, including stockouts of essential medicines, low levels of training on diagnosis and treatment among health workers, a perception that diarrhea is not a serious condition, and the inconveniences of ORS and zinc – ORS has an unpalatable taste and zinc requires a two-week dosage regimen. In some areas, there was also a strong preference for antibiotics, which was not part of the national standard for treating uncomplicated cases of diarrhea.

Overall, the integration of ORS and zinc into facility-based ICMI and iCCM was successful. However, gaps in both acceptance and adherence (by providers and caregivers alike) limited the coverage of these interventions. Other limiting factors included the growing role of the private sector in diarrhea treatment, and the lack of involvement of the FMOH in this area.

In 2015, the FMOH partnered with DKT Ethiopia to distribute print materials promoting the use of ORS and zinc, including fliers and posters. That same year, the ministry began nationally televised promotions of ORS and zinc in multiple languages to increase knowledge and uptake of the treatments.

At the end of the study period, in 2016, the availability of ORS and zinc still varied widely by region.¹⁷ Overall usage of both treatments remained low, again with significant disparities among regions.¹⁸ The 2016 DHS found that among children with diarrhea, 35 percent received zinc, and only 17 percent received both ORS and zinc.¹⁸

Rotavirus Vaccine

WHO released a recommendation for universal rotavirus vaccination in 2009.¹⁹ Following this announcement, Gavi began offering financial support to 33 countries to help introduce the vaccine.²⁰

In 2008, Ethiopia accounted for 6 percent of rotavirus-related diarrhea deaths worldwide and was the fifth-largest contributing country.²¹ The FMOH conducted WHO-sponsored rotavirus surveillance from 2007 to 2009 at the Tikur Anbessa (Black Lion) Specialized Hospital in Addis Ababa and confirmed the presence of rotavirus in 27 percent of all children under five hospitalized with acute diarrhea.²²

The FMOH’s immunization technical working groups, which included organizations such as WHO, UNICEF, the US Centers for Disease Control and Prevention (CDC), the Clinton Health Access Initiative, and the US Agency for International Development (USAID), used local surveillance evidence and global data to consider introduction of the rotavirus vaccine in Ethiopia.

In 2009, the FMOH dispatched Ethiopia’s Vaccine Management Assessment (VMA) to evaluate and improve the country’s vaccine infrastructure, including storage facilities. cold-chain equipment, and transportation links. Due to these efforts, Ethiopia’s vaccine storage capacity was already sufficient to accommodate introduction of the rotavirus vaccine by the time the government applied to Gavi in 2011 for financial support of a rotavirus vaccine.

The funding was approved in September 2011 and called for a March 2012 rollout of the vaccine on a two-dose schedule, with government co-funding.²³ The introduction of the vaccine was delayed by more than a year and a half, due to a global shortage of the preferred two-dose treatment, and to the competing priorities of other vaccination campaigns (particularly for meningococcal meningitis) taking place in Ethiopia at the time.²⁴

On November 7, 2013, Ethiopia became the 17th Gavi-eligible country to introduce a national rollout of the rotavirus vaccine. The vaccine was distributed in all regions except the Somali Region, which was grappling with a polio outbreak.²⁵ It was introduced in the Somali Region the following year in August 2014, after the polio vaccine campaign.²⁶

National coverage rates lagged behind the FMOH’s “best-case scenario” target, which was 90 percent rotavirus immunization coverage by 2014–2015.²⁷ WHO and UNICEF reported that 56 percent of Ethiopian children received both doses of the vaccine in 2014, and 79 percent in 2015. The coverage rate for rotavirus vaccine in the first year after rollout, however, was comparable to that of other routine childhood vaccines. By 2015, rotavirus-2 coverage rates surpassed those of other vaccines, despite having been introduced more recently.²⁸

The 2016 DHS showed a large regional variation in coverage, similar to differences seen for other routine vaccines in Ethiopia. Coverage was lowest in the primarily pastoralist region of Afar, where only 23 percent of children received both doses of the rotavirus vaccine. It was highest in Addis Ababa, with a coverage rate of 92 percent. The DHS also revealed a notable coverage gap between the first and second doses – 64 percent for the first dose, but only 56 percent for the second.¹⁸

Water, Sanitation, and Hygiene

Finally, in addition to the rotavirus vaccine, ORS, and zinc supplementation, Ethiopia addressed diarrhea-related illnesses through water, sanitation, and hygiene (WASH) interventions such as improved access to clean water and the introduction of policies to reduce the incidence of open defecation.

Ethiopia’s WASH improvements helped reduce the incidence of diarrhea and, therefore, diarrhea-related U5M. WASH policies and programs, such as community-level awareness campaigns, were instrumental in reducing the national open-defecation rate from 80 percent in 2000 to 27 percent in 2015.²⁹

Meanwhile, access to safe sources of drinking water increased, reaching 65 percent of households in the country in 2016, though large differences between urban and rural areas persist.¹⁸

The incidence of diarrhea dropped in all regions during the study period, and U5M due to diarrhea decreased by more than half, from 424 deaths per 100,000 people in 2000 to 191 in 2017.³⁰ As care-seeking for diarrhea and coverage of ORS and zinc ultimately remained low during the study period, WASH efforts likely contributed to this reduction in disease burden and mortality.

A statistical analysis conducted by the Institute for Health Metrics and Evaluation (IHME) found that 3 percent of the reduction in under-five deaths between 2000 and 2017 was attributable to WASH .¹⁵⁶   A very small reduction is attributed to rotavirus vaccine (less than 1 percent) – the vaccine was administered at national scale during just the last three years of the period assessed in this analysis (2000 to 2017) .¹⁵⁶

Malnutrition

General Overview

Research has estimated that half of all under-five deaths globally occur in children with malnutrition.³¹ Severe acute malnutrition may directly cause U5M, and children suffering from malnutrition are also more vulnerable to incidence of and death from other childhood illnesses such as diarrhea and pneumonia.³²

Food insecurity is an ongoing concern in Ethiopia, where a famine in the mid-1980s killed an estimated 1 million people. Even today, amid rapid economic growth and poverty reduction, the country still lives with the threat of famine, and nearly 8 million people require some form of food assistance.³³

The country’s vulnerability to mass hunger derives in part from the fragility of its agricultural sector, which consists mainly of subsistence farms;³⁴ 95 percent of Ethiopia’s farmland is divided into holdings of 25 hectares or less.³⁵

Ethiopia stands out as a country that made major commitments to improving nutrition with a holistic, multisectoral approach. It developed a National Nutrition Strategy in 2008, and then a National Nutrition Plan in 2013. Both plans sought to address the top three nutrition priorities in the country: food insecurity, undernutrition and malnutrition, and severe micronutrient deficiencies (especially of iron, iodine, and vitamin A).³⁶

While global IMCI protocols do not typically include treatment of malnutrition, Ethiopia made sure to include treatment of severe acute malnutrition in its IMCI program – a step that, among other things, ensured that malnutrition care was included in preservice training for health workers.

Treatment of severe acute malnutrition also became a central feature of the Health Extension Program. HEWs began providing vitamin A supplementation and deworming, community management of acute malnutrition, salt iodization, and zinc supplementation.³⁷

By January 2008, 165 hospitals and health centers in Ethiopia provided in and outpatient treatment for malnutrition. At that time, management of these services was shared by the FMOH, UNICEF, and international nongovernmental organizations (NGOs). That year, in response to drought emergencies in Oromia and SNNP, the FMOH also rapidly expanded its Community-Based Management of Acute Malnutrition program, which it had introduced on a pilot basis in 2000.³⁸

The FMOH aimed to first establish outpatient therapeutic programs at 1,239 health posts in 100 drought-affected districts (woredas) in Oromia and SNNP, with financial and technical support from UNICEF. Establishment of outpatient therapeutic programs at health posts improved service coverage from 38 percent to 65 percent in affected areas. In all, 27,739 children were admitted to these programs in the 455 health posts that established programs between July and October 2008.³⁹

Of these children, 78 percent were cured of severe acute malnutrition, surpassing international Sphere performance standards of 75 percent cured. The child mortality rate during this period was 0.7 percent, well below the Sphere standard of under 10 percent mortality. This monitoring demonstrated the feasibility of using HEWs to treat severe acute malnutrition.³⁹

Following the national expansion of iCCM, severe acute malnutrition cases were managed free of charge at the community level across Ethiopia, improving access to these services. However, a study of children admitted to outpatient therapeutic programs in 94 health districts in SNNP in 2011 noted that upon discharge from the program, only 33 percent of children admitted with severe acute malnutrition met the recovery criteria of 15 percent weight gain or resolution of edema. Fourteen weeks after admission to the outpatient therapeutic program, 40 percent of children were still classified as having severe acute malnutrition, 37 percent were classified with moderate acute malnutrition, and 23 percent had attained normal nutritional status.⁴⁰

In 2015, the government of Ethiopia declared the failure of the spring rains (known as Belg), which affected farmers and pastoralists across the country. After an assessment in October 2015 found that the harvest did not meet expectations due to El Niño weather patterns, the government worked with UNICEF to increase the number of health facilities capable of treating severe acute malnutrition to 14,778 facilities by December 2015. Severe acute malnutrition admissions increased accordingly, with nearly 300,000 cases treated in 2015; that year UNICEF reported that 88 percent of severe acute malnutrition cases were successfully treated.⁴¹

Treatment of severe acute malnutrition in Ethiopia continues to be available at both the facility and community levels, with high sustainability due to its integration with the national facility-based IMCI and iCCM programs. The 2016 Service Availability and Readiness Assessment found that 84 percent of public health facilities offered diagnosis and treatment of malnutrition in children. At the community level, 82 percent of health posts offered this service.¹⁷

Within the study period, there was a decline in the rates of stunting (57.7 percent in 2000 to 38.4 percent in 2016) and underweight (41.2 percent in 2000 to 23.3 percent in 2016), with relatively high rates persisting for both conditions. There was a smaller reduction in rates of wasting (12.2 percent in 2000 to 9.8 percent 2016), although baseline rates for that condition were fairly low to begin with.³⁷ 

Malnutrition is a byproduct of extreme poverty as well as nutritional insecurity, and the national government sought to address it as such. In 2005, it implemented the Productive Safety Net Programme, which focuses on chronically food-insecure districts in six regions of the country – Afar, Amhara, Oromia, SNNP, Somali, and Tigray. This initiative offered jobs in conservation and infrastructure projects, with payment made in either cash or food, with the labor requirement waived for low-income households that cannot provide any workers.

Assessments have shown that the program’s benefits include improved food security, likely contributing to the improved nutritional status of children under five, and to a reduced vulnerability to some of the major causes of childhood mortality.

[For more information on malnutrition and its effects, see Exemplars in Global Health: Stunting in Ethiopia].

Vitamin A Deficiency

Vitamin A deficiency is a widespread malnutrition problem among children in lower- and middle-income countries. This condition contributes to U5M due to an increased risk of illness and death from causes such as diarrhea and measles. Routine vitamin A supplementation for children ages 6 to 59 months is therefore recommended in Ethiopia and other countries where deficiencies are prevalent.⁴²

Though Ethiopia’s vitamin A supplementation coverage rate was high in the late 1990s, it dropped to 3 percent in 2000 and 2001 following the reported deaths of children who had choked as a result of the whole vitamin A capsule being accidentally put into their mouths.⁴³

In April 2004, the FMOH initiated the Enhanced Outreach Strategy in areas vulnerable to drought and food shortages. As a result of this campaign, vitamin A supplementation coverage increased to more than 80 percent in the targeted regions by 2006.

The ministry expanded the program to all districts in Ethiopia in 2007, targeting all children in Ethiopia between the ages of 6 and 59 months for vitamin A supplementation. A program-coverage survey conducted in 2008 reported that the national coverage of supplementation among children in this age range had risen to 94 percent, with only minor regional variability (from 82 percent in Dire Dawa to 99 percent in Tigray).⁴³

The Enhanced Outreach Strategy continued in three highly pastoralist regions with low coverage of health services.⁴⁴ Despite these efforts, coverage of supplementation remained fairly low. Nationally, coverage decreased from 53 percent in 2011 to 45 percent in 2016.⁴⁵

A statistical analysis conducted by the Institute for Health Metrics and Evaluation (IHME) found that 29 percent of the reduction in under-five deaths between 2000 and 2017 was attributable to reductions in child growth failure, while another 3 percent of the reduction in under-five deaths was attributable to reductions in vitamin A and zinc deficiency.¹⁵⁶   

Malaria

Insecticide-Treated Bed Nets

Malaria is endemic to all regions of Ethiopia except Addis Ababa,⁴⁶ yet it causes only a small portion of U5M in Ethiopia – 0.4 percent in 2015, down from 2.4 percent in 2000.⁴⁷ Transmission varies greatly, even within endemic regions, and is mostly associated with geographies at lower altitude.

In 2000, the government of Ethiopia signed the Abuja Declaration on malaria, committing to a range of interventions called for by the Roll Back Malaria initiative, including the provision of insecticide-treated mosquito nets (ITNs) and other protective measures within five years to at least 60 percent of all people at risk of contracting the disease.⁴⁸

To prepare for such a broad distribution of ITNs, the government submitted a proposal to the Global Fund to Fight AIDS, Tuberculosis and Malaria to obtain 6 million long-lasting ITNs for free distribution.⁴⁹ Additional partners such as UNICEF, the World Bank, and the Carter Center also provided substantial support for procurement of ITNs.⁵⁰

Although use of ITNs was an area of emphasis in the FMOH’s first National Five-Year Strategic Plan for Malaria (2001–2005), the 2005 DHS found that just 3.3 percent of households owned an ITN, and only 6 percent owned a bed net of any kind.^51,52

As part of a new initiative to increase ITN coverage, the FMOH initiated free distribution of long-lasting ITNs in 2005. This campaign employed three main delivery strategies: the Enhanced Outreach Strategy, routine distribution through health facilities, and door-to-door distribution by health extension workers.

The country’s second National Five-Year Strategic Plan for Malaria (2006–2010) was released in April 2006. It set a new goal of providing, on average, two ITNs per household in 90 percent of target areas by 2007.⁵¹

As a result of these efforts, 23.8 million ITNs were distributed by the end of 2007. Between 2005 and 2011, 43.1 million ITNs were given to people living at altitudes of 2,000 meters or below – the topography where people are most at risk for malaria outbreaks.⁵¹

In 2011, the FMOH’s ITN distribution strategy was expanded to include free provision of ITNs in all malaria-endemic areas.⁵³ Despite these efforts, coverage gaps remained. In 2015, the Ethiopia National Malaria Indicator Survey reported that only 64 percent of households in endemic areas below 2,000 meters in altitude owned even one ITN.⁵⁴

The strategy for distribution of ITNs was to use a continuous rolling campaign approach, with households receiving new ITNs every three years – the estimated life span of the nets. This strategy ensured lasting coverage of households in endemic areas.⁵⁵

Indoor Residual Spraying

WHO has recommended indoor residual spraying for control and elimination of malaria since well before the study period. In 2006, it released a position statement recommending indoor residual spraying as a strategy for achieving malaria-related Millennium Development Goals by 2015.⁵⁶

In Ethiopia, deltamethrin was introduced as the insecticide of choice in the mid-2000s due to growing levels of resistance to DDT (dichloro-diphenyl-trichloroethane) in most areas of the country.⁵³ In addition to its own funding efforts, the government received partial funding for indoor residual spraying activities from the Global Fund in 2003, and from the President’s Malaria Initiative in 2007.⁵⁷

The National Five-Year Strategic Plan for Malaria Control in Ethiopia (2001–2005) established a goal of reaching 60 percent indoor residual spraying coverage in epidemic-prone areas by 2005. However, only 30 percent of these areas were estimated to have been sprayed in 2007, due to a lack of funding. After the government received additional funding from the Global Fund, coverage levels rose to 47 percent in 2011 in all areas with malaria transmission.⁵⁸ 

Respiratory Infections

Pneumococcal Conjugate Vaccine

In 2007, Ethiopia’s immunization Inter-Agency Coordination Committee – a group consisting of FMOH directorates and international development partners such as WHO, UNICEF, the Clinton Health Access Initiative, and USAID – began discussing pneumococcal conjugate virus (PCV) as a means of preventing pneumonia.⁵⁹

Approximately 80 percent of the burden of pneumococcal disease in the country was caused by the ten serotypes contained in the 10-valent version of PCV, known as PCV-10.⁶⁰ This led the government to select PCV-10 over the more limited 7-valent form of the vaccine (PCV-7). After a period of intensive study and preparation, the FMOH announced its intention to introduce PCV-10 by January 2010.⁵⁹

At the time of the vaccine’s introduction, lower respiratory infections accounted for 15 percent of all deaths among children under five in Ethiopia.⁴⁷ Based on slightly higher FMOH estimates of deaths from pneumonia, introduction of PCV in Ethiopia was estimated to have the potential to avert 35,000 deaths in one birth cohort over a period of five years.⁶⁰

The Inter-Agency Coordination Committee then established a pneumococcal vaccine introduction task force to prepare the vaccine introduction plan, as well as a request to Gavi for financial support of the program. Gavi approved the request in July 2010. Under its terms, the government was expected to provide co-financing of about US$13.4 million for PCV doses, syringes, and other vaccination supplies, with Gavi providing approximately US$341 million.⁶¹

At the time of Ethiopia’s request for support, an even more versatile form of the vaccine – PCV-13 – was available through Gavi, and several experts preferred it over PCV-10 due to its coverage of additional strains, but selection of PCV-13 was expected to delay introduction. The government stuck with PCV-10, opting to save additional lives through earlier implementation rather than delaying the introduction of a sufficiently effective vaccine.⁶²

Following a comprehensive FMOH-led advocacy and communication campaign, PCV-10 was introduced countrywide in Ethiopia in November 2011. It was added to the Expanded Programme on Immunization (EPI) infant vaccine schedule, to be administered at 6 weeks, 10 weeks, and 14 weeks of age alongside the pentavalent vaccine. As a catch-up strategy unique to this particular vaccine, all children in the country under the age of one year were eligible to receive PCV. The vaccine was administered as a routine vaccination at health facilities and community health posts, as well as through dedicated outreach campaigns.⁶³

The country’s EPI Comprehensive Multi-Year Plan (CMYP) for 2011–2015 set targets for complete triple-dose PCV coverage: 88 percent in 2011, 92 percent in 2012, 95 percent in 2013, and 96 percent in 2014 and 2015.⁶⁴

WHO and UNICEF found that actual outcomes fell substantially short of these targets, reporting estimated coverage levels of 12 percent in 2011, 30 percent in 2012, 48 percent in 2013, 55 percent in 2014, and 68 percent in 2015.⁶⁵

The 2015 DHS reported still lower coverage levels: 67 percent of children ages 12 to 35 months received a first dose of PCV and 61 percent got a second dose, but only 49 percent received all three doses of the vaccine.¹⁸

Three-dose PCV coverage was higher in children residing in urban areas (73 percent) compared with those living in rural areas (46 percent). Its regional footprint varied from a high of 91 percent coverage in Addis Ababa to 18 percent in Afar. Despite the availability of free immunizations, there was a significant wealth gap in the completion of all three doses – 76 percent in the highest-income quintile and 36 percent in the lowest.¹⁸ 

Other Vaccination Campaigns

Vaccination campaigns across Ethiopia’s vast territory have faced several challenges, including warfare and droughts. Despite these challenges, however, the country has made significant achievements. A 2012 EPI cluster study found that 97.1 percent of urban health facilities and 89.2 percent of rural health facilities, including health posts, surveyed provided routine immunization services on a regular basis, though key informants mentioned difficulties in obtaining accurate coverage data.⁶⁴

Measles Vaccine

The campaign against measles represents a notable example of how Ethiopia has overcome many obstacles to vaccination coverage. National implementation of the measles vaccine began in 1980 with the launch of the EPI. As one of six vaccines initially included in the program, the measles-containing vaccine (MCV1) was part of the routine immunization schedule to be given as a single dose at or after nine months of age.

In 2000 – just before the study period began – the national immunization coverage rate was 27 percent, according to the DHS.¹⁸ In 2002, Ethiopia sought to improve the stagnant coverage levels by conducting supplemental immunization activities. With support from WHO and UNICEF, these were carried out in a phased approach, focusing on areas that surveillance data had identified as especially high risk for measles outbreaks.

Additional phases in 2004 and 2005 targeted the remaining areas of the country, raising the maximum age for vaccination from 6 months to 15 years. Case-based surveillance conducted since 2002 had indicated that the burden of measles cases was shifting from children under five to older children. This was most likely due to the country’s dedicated immunization activities for children under five. The country continued to conduct follow-up campaigns every two to three years, with support from Gavi.⁶⁶

As is the case for many vaccines implemented in Ethiopia, estimates of coverage for measles immunization vary. WHO and UNICEF estimated MCV1 coverage to be 44 percent in 2006 and 65 percent in 2015.⁶⁷

The 2016 DHS found greater gaps in coverage, reporting that 54 percent of children ages 12 to 23 months had received the measles vaccine, an increase from 27 percent in 2000.¹⁸ While coverage gaps persisted, the increase in coverage since 2000 was significant, contributing to reductions in disease burden.

Geographic differences were pronounced, with coverage ranging from 93 percent in Addis Ababa to 30 percent in Afar. Though equity graphs show improvement in coverage across all wealth quintiles from 2000 to 2016, large gaps between these groups persist. The 2016 DHS found that coverage of the measles vaccine was only 43 percent in the lowest wealth quintile, but it reached 74 percent in the highest quintile.¹⁸

Measles surveillance indicated that despite increases in coverage, the incidence rate of confirmed measles cases in Ethiopia increased, from 0.6 per 100,000 under-5s in 2005 to 11.1 in 2014.⁶⁸ However, it is likely that this change primarily reflects the strengthening of the measles surveillance system and its improved ability to detect and report cases.

A clearer sign of the effects of Ethiopia’s measles campaigns is the simple fact that the proportion of measles cases in children under five declined sharply in a short time – from 56 percent in 2008 to 30 percent in 2014.⁶⁸

Additionally, the rate of deaths due to measles among children under five decreased by 92 percent between 2000 and 2017.³⁰ This reduction represented 19 percent of the total reduction in mortality among children under five during this period .¹⁵⁶

Meningococcal Vaccine

Another significant vaccination achievement during the study period was Ethiopia’s effort to reduce meningococcal meningitis, a disease that killed thousands of children per year in Ethiopia during the 1990s.

In response to that epidemic, WHO created the Epidemic Meningitis Vaccines for Africa project. In September 2010, a new vaccine against meningitis – MenAfriVac – became available, and was quickly introduced in Burkina Faso, Mali, and Niger, with other high-incidence countries soon to follow.⁶⁹

In Ethiopia, WHO conducted a meningitis risk assessment in January 2012, identifying five regions at especially high risk for meningitis outbreaks. The remaining six regions were considered at moderate to low risk.⁷⁰

In August 2012, the government applied to Gavi for support of a phased introduction of the meningococcal vaccine beginning the following year. The FMOH decided to use this staggered approach based on the WHO risk assessment research, and the increased risk of meningitis in the western part of Ethiopia, which bordered South Sudan, a country vulnerable to outbreaks. Gavi required no co-funding by the government for the campaign, which was estimated to cost US$78.4 million.

As planned, the FMOH introduced the meningococcal vaccine in 2013, with areas already identified to be at high risk for meningitis outbreaks prioritized for vaccine introduction, and medium-risk and low-risk areas receiving the vaccine in later phases. 

The FMOH used the HEWs and Health Development Army of the Health Extension Program and social mobilization networks to engage and inform communities about the vaccine. Promotional messages were also disseminated via national and regional radio and television channels.

Phase 1 delivered a single dose of meningococcal vaccine to nearly 19 million people between the ages of 1 year and 29 years, achieving 92.4 percent coverage (see Figure 2).⁷¹

Phase 2, which began in October 2014, had an estimated coverage rate of 93.5 percent in targeted areas. In 2015, Phase 3 attained the nearly identical coverage rate of 93 percent.⁷¹

As a result of these and related interventions, the rate of death due to meningococcal meningitis among children under five declined over the study period by 68 percent, though the proportion of U5 deaths caused by the disease remained constant at less than 1 percent.³⁰

Haemophilus Influenzae Type B and the Pentavalent Vaccine

In 1998, WHO recommended introduction of the Haemophilus influenzae type B (Hib) vaccine into national immunization programs of countries with a high Hib burden .⁷²

Ethiopia was one of those countries. The Tikur Anbessa (Black Lion) Specialized Hospital in Addis Ababa began collecting and reporting sentinel surveillance data on bacterial etiologies of pediatric meningitis in 2001. In 2004, surveillance data showed 107 suspected meningitis cases, and that Hib had been present in 44 percent of them.⁷³

In 2003, the Ethiopian government decided to introduce new vaccines against Hib and hepatitis B into the EPI. The government received funding from Gavi in 2005 to introduce a pentavalent vaccine that protects against five major diseases – diphtheria, tetanus toxoids, whole-cell pertussis, hepatitis B, and Hib.

The government and its partners quickly turned to the expansion of the country’s vaccine storage space at all levels to support new vaccine introductions. As part of this effort, approximately 1,400 refrigerators procured with support from UNICEF and the government of Ireland were distributed to regions in 2005, doubling cold storage space at the regional level.⁷⁴

The FMOH intended to introduce the pentavalent vaccine in January 2006. However, the global supply of the pentavalent vaccine was still somewhat limited. Faced with the choice of waiting for the pentavalent vaccine or opting for an immediate rollout of a quadrivalent vaccine that did not cover hepatitis B, the ministry chose to wait.

The first shipment of pentavalent vaccine arrived in Ethiopia in December 2006,⁷⁵ and was introduced into the EPI in March 2007, replacing the previous DTP vaccine. Following its inclusion in the immunization schedule, the vaccine was administered at public and NGO-supported health facilities free of charge.²²

FMOH estimated that coverage of the pentavalent vaccine (all three doses) reached 81 percent by 2008 and 96 percent by 2015. However, as with some other vaccines, figures from WHO and UNICEF paint a more modest picture, indicating a coverage level of 52 percent in 2008 and 73 percent in 2015 – and a decline of 12 percentage points between the first and third doses (from 85 percent to 73 percent).⁷⁶

According to the 2016 DHS, the coverage rate for all three doses was even lower, at 53 percent. It also found that almost 80 percent of urban children had received all three doses, whereas only half of those living in rural areas did. In addition, the DHS reported stark regional differences in coverage, with a low of 20 percent in Afar and a high of 96 percent in Addis Ababa.¹⁸

The FMOH sought to address the decline in coverage between doses through community-level tracking, with HEWs identifying and registering unvaccinated children in their communities. However, a key informant said the level of decline in coverage between doses are still “not acceptable,” and present a continuing challenge.

Overall, across all vaccines, results from a statistical analysis conducted by the Institute for Health Metrics and Evaluation (IHME) showed that 21.4 percent of the reduction in under-five deaths between 2000 and 2017 can be attributed to vaccine interventions, including vaccines against measles (10.7 percent), DTP3 (4.1 percent), PCV (3.5 percent), Hib (3.1 percent reduction), and rotavirus (<1 percent).¹⁵⁶

HIV

Prevention of Mother-to-Child Transmission

In 1999, the FMOH estimated that mother-to-child transmission of HIV contributed to up to 25 percent of all new HIV infections in Ethiopia. An estimated 250,000 children were living with HIV in 2000.⁷⁷

In 2001, the FMOH prepared a five-year national strategic framework (2001–2005) for HIV/AIDS. That same year, the FMOH released the country’s first guidelines on the prevention of mother-to-child transmission (PMTCT) of HIV in Ethiopia. Several partners were involved in developing the guidelines, including WHO, UNICEF, and USAID. These guidelines adopted WHO’s four-pronged approach:⁷⁸

• Primary prevention of HIV infection.
• Prevention of unintended pregnancies among HIV-infected women.
• Prevention of HIV transmission from infected women to their infants.
• Treatment, care, and support of HIV-infected women, their infants, and their families.

The government of Ethiopia received US$370 million from the Global Fund between 2004 and 2015 to support progress toward HIV-specific objectives, including PMTCT.⁷⁹ One of the early steps the government took toward those objectives was to increase the number of health facilities providing PMTCT services. In 2005, 129 health facilities in Ethiopia provided PMTCT. This number increased more than tenfold by 2010, to 1,352 facilities.⁸⁰

In 2005, IntraHealth International established the Mothers’ Support Group (MSG) program, which focuses on reducing mother-to-child transmission by providing support for pregnant and postpartum women who are HIV-positive. Two of the program’s seven main objectives included increasing access and use of PMTCT services. MSG established health center-based peer support groups consisting of trained “mentor mothers,” pregnant women, postpartum women, program graduates, and community members.⁸¹

The program began with the establishment of three sites in 2005 and grew to 85 sites across six regions and city administrations by 2008. The FMOH took over the MSG program in 2013 and began slowly expanding it across the country.^82,83  

This extended reach was reflected in the increase in women receiving PMTCT. From mid-2004 to mid-2005, an estimated 1,314 women received PMTCT services in Ethiopia, increasing to 6,990 by mid-2010. Overall PMTCT coverage remained low, but was edging upward, from 5 percent in 2006–2007 to 8 percent in 2009–2010.

Following international guidelines, Ethiopia adopted Option B+ in 2013. The Option B+ treatment strategy initiates lifelong antiretroviral therapy (ART) for all pregnant women who are HIV-positive, regardless of clinical stage or CD4 white blood cell count.

The 2014 ESPA+ survey reported that availability of PMTCT services at facilities offering antenatal care (ANC) was high across all facility levels. All referral hospitals offering ANC offered HIV testing and counseling services for pregnant women, and 93 percent also offered PMTCT through provision of ART to HIV-positive pregnant women. Availability of PMTCT services was similarly high at general and primary hospitals.¹⁰

The Joint United Nations Programme on HIV and AIDS (UNAIDS) estimated that coverage of HIV-positive pregnant women in Ethiopia who receive ART for PMTCT reached 70 percent in 2014. This progress in coverage was accompanied by an impressive reduction in estimated new HIV infections in children younger than age 14, from 15,000 in 2000 to 3,300 in 2015.⁸⁴ An estimated 3,700 new cases of pediatric HIV were averted in 2015 due to PMTCT. Although one source reported the rate of mother to child transmission at 16 percent,⁸⁵ other sources indicated better success, with one study reporting rates of 5.9 percent at selected facilities in Northwest Ethiopia.⁸⁶

Early Infant Diagnosis

Ethiopia’s initial PMTCT guidelines included a recommendation of antibody testing of HIV-exposed infants at 18 months of age. ICAP at Columbia University (formerly the International Center for AIDS Care and Treatment Programs) initiated a pilot program for early infant diagnosis at 16 Ethiopian health facilities.⁸⁷

Following implementation of this pilot project, a workshop attended by the FMOH, implementing partners, and other stakeholders was held to guide development of the national early infant diagnosis program in 2006. The FMOH worked with partners such as ICAP to develop the National Early Infant Diagnosis Implementation Plan in 2007. In preparation for scale-up of the program, two laboratories (one in Addis Ababa and one in Hawassa) were renovated by the Ethiopian National Laboratories, CDC, Johns Hopkins University Technical Support for the Ethiopian ART Initiative, PEPFAR, and the Clinton Health Access Initiative.

In 2007, the FMOH released its Guidelines for Paediatric HIV/AIDS Care and Treatment in Ethiopia. These guidelines urged early infant diagnosis as a primary component of care for HIV-exposed infants to ensure timely access to care for HIV-infected infants.⁸⁸

Scale-up of the national early infant diagnosis program began in January 2008. Initial scale-up was planned to expand from two to six referral laboratories and more than 1,000 health facilities in the country.⁸⁷ At the time of scale-up, dried blood spot samples were transported by laboratory couriers in Addis Ababa and by clinical mentors in other areas.⁸⁹ According to a key informant, the pace of the initial scale-up was limited by availability of PCR (polymerase chain reaction) machines, which were available only at the central and regional laboratories and a very limited number of health facilities.

By 2009, early infant diagnosis programs were in place at 58 hospitals and 23 health centers across the country. At that time, 257 health workers had received training on dried blood spot sample collection and early infant diagnosis protocols. Though still a tiny number, this represented a national increase of 300 percent.⁸⁹

Antiretroviral Therapy

One key informant who had been involved in the HIV treatment effort reported that the government of Ethiopia launched fee-based ART in 2003. Before that year, coverage of the treatment was very low, with drugs available only at a small number of private facilities.

The HIV epidemic reached its peak in Ethiopia in the early to mid-2000s, resulting in a very high demand for ART. In 2005, the national government initiated a national free ART program, including pediatric treatment. In preparation for this, the government signed memoranda of understanding with the Global Fund for the supply of ART drugs and with PEPFAR for technical support.

The FMOH drew upon local experts and several international partners, including research institutions such as the Johns Hopkins University and the University of Washington, to develop national standard training materials for health care providers. These partners collaborated in weekly meetings to customize international ART training materials to fit the Ethiopian context – and were later also involved in provider training and the overall implementation of the ART program.

To improve access in rural areas, ART services were decentralized in 2006 to health centers and hospitals spanning all three levels of the public health system. From an initial eight facilities, Ethiopia’s ART program grew to 60 facilities within three years. By 2013, 1,047 HIV care clinics were providing comprehensive care, including ART,⁹⁰ for the more than 48,000 Ethiopian children under the age of 15 who had tested positive for HIV.^91,92

A 2017 study assessed the performance of Ethiopia’s ART program from 2005 to 2015. It found that coverage among HIV-positive children under age 15 had grown significantly, from 1 percent in 2005–2006 to 25 percent in 2014–2015, reflecting progress but also an ongoing gap in service provision.⁹³

By 2016, 61 percent of adults, but only 35 percent of children, living with HIV were receiving ART – a troubling disparity that represents an ongoing challenge for Ethiopia’s health system. Yet overall, HIV treatment for the youngest patients had come a long way – a development perhaps illustrated most clearly by the fact that the number of AIDS-related deaths among Ethiopian children age 14 or younger declined 79 percent from 2002 to 2016.⁹⁴

Neonatal Mortality

Historically the FMOH was not as strongly committed to newborn health as it was to overall child health, an inclination perhaps at least partially explained by the ingrained tendencies of a country where neonatal death was heartbreakingly common. As one key informant said,

In 2003, a Lancet series on child survival emphasized that any substantial reduction in U5M would be impossible without addressing newborn care.⁹⁵ Following the release of this series, the FMOH began to elevate neonatal issues as a health priority, drawing upon the support of partners such as the Ethiopian Pediatric Society, WHO, and UNICEF.

However, progress toward reducing neonatal mortality was slow, particularly compared to progress in reducing U5M. Between 2000 and 2015, Ethiopia’s neonatal mortality rate declined from 44 deaths per 1,000 live births to 26, a decline of 41 percent, compared to a 56 percent reduction in overall U5M during the same period.³⁰

The country’s gains in this area are attributable to multifaceted intervention programs covering the three periods of risk: antenatal, delivery, and postnatal stages. The following section will discuss each of these stages in turn.

Antenatal Care

In 2000, the DHS reported that only 27 percent of Ethiopian women giving birth over the preceding five years had attended any ANC visits, and only one in ten had attended the four visits that WHO recommends.

The survey showed large demographic differences in the use of ANC – 63 percent of mothers in urban areas had attended at least one ANC visit, whereas only 22 percent of rural mothers had done so. Among mothers with secondary or higher education, 72 percent had attended at least one ANC visit, compared with only 22 percent of women with no formal schooling.⁹⁶

Ethiopia adopted strategies to both increase access to ANC and improve the quality of interventions offered, including treatment for hypertension and eclampsia as well as maternal tetanus vaccination.

Improving Access to Antenatal Care

Throughout the study period, the government tried a variety of means to increase ANC attendance. These efforts generally resulted in significant increases in access to service, though attendance rates remained low.

With the establishment of the Health Extension Program in 2003, Ethiopia gained an important new asset for improving ANC access and use. Health extension workers (HEWs) were trained to provide the full round of four consultations, conduct home visits, refer pregnant women with health problems to facilities,^97,98 and act as promoters of ANC in their communities.⁹⁹

In 2005, the FMOH introduced health care financing reforms, with the specific aim of improving equity in the availability of health services. Maternity and family care services, including ANC, were made free of charge at health posts, health centers, and primary hospitals.¹⁰⁰

The 2011 DHS found that use of ANC services in Ethiopia had increased somewhat. The proportion of expectant mothers attending at least one ANC visit had increased from 27 percent in 2000 to 43 percent in 2011.

Attendance at all four visits had also increased, though the rate still remained low – 10 percent in 2000 and 19 percent in 2011. The increased role of HEWs had only a limited effect, at least with regard to ANC attendance; only 9 percent of mothers reported receiving ANC from a HEW.¹⁶

Although ANC attendance had not yet reached the levels the government hoped for, access was improving markedly. A 2007 assessment found that 83 percent of health posts were capable of providing ANC,⁹⁸ and a 2014 study reported high availability of ANC services nationally, with 87 percent of all facilities assessed offering services and 74 percent offering services at least five days per week.¹⁰ As a further promising sign, availability of ANC services was generally high across regions.¹⁰

The 2016 DHS showed continued increases in ANC uptake, though still fairly low use of services despite better access. Among women who had given birth in the preceding five years, 64 percent received some form of ANC from a skilled provider, and 33 percent had attended four or more visits.¹⁸ Both figures increased across all wealth quintiles from 2000 to 2016, though gaps in equity between highest and lowest quintiles persisted.

One key informant explained that early use of ANC remains affected by cultural and social norms, since many women consider it taboo to share news of a pregnancy until it is visible. A 2015 study of women attending ANC clinics in Tigray confirmed that timely initiation of care remains a challenge – only 41 percent of pregnant women booked timely ANC sessions, and the median duration of pregnancy at the first visit was five months.¹⁰¹

Hypertension and Eclampsia

The importance of expanding ANC access is visible across several of the diseases and conditions that contribute to U5M in Ethiopia. One example is hypertensive disorders such as eclampsia – a serious and sometimes fatal pregnancy-related disorder characterized by convulsions in mothers suffering from an elevated blood pressure condition known as preeclampsia.

An extensive 2008 FMOH study of emergency medical care found that preeclampsia (and eclampsia) were present in only 1 percent of infant deliveries, but these conditions accounted for 5 percent of all birth-related complications in Ethiopia, and 16 percent of maternal peripartum deaths related to quality of obstetric care.¹⁰²

The assessment found that antihypertensive drugs were available at 75 percent of health centers and 99 percent of hospitals. Among specific hypertensive medications, methyldopa – the first-choice drug for hypertension at the time of the assessment – was available at 97 percent of all facilities.¹⁰²

The consensus on what constituted optimal treatment, however, was shifting. In 2011, WHO recommended a different medication – magnesium sulfate – to prevent eclampsia in women with severe preeclampsia, as well as for treatment of women with eclampsia.¹⁰³

The Ethiopian medical community was already familiar with magnesium sulfate’s potential in this area. National treatment guidelines had featured it since 2004, and since 2009 its broader introduction had been championed by an impressive multinational coalition, including the FMOH, the Ethiopian Society of Obstetricians and Gynecologists, UNICEF, the Pharmaceuticals Fund and Supply Agency, and Emory University.¹⁰⁴

Despite this high-caliber support, magnesium sulfate remained under review for addition to Ethiopia’s National Essential Medicine List until 2012 and was finally included in 2014. As of that year, 22 percent of facilities carried magnesium sulfate – double the rate seen in 2008, but still very low.¹⁰⁵

Widespread use of magnesium sulfate did not begin until after the study period.¹⁰⁴ In 2016, a national assessment found stockouts of magnesium sulfate across all facility levels, primarily due to inadequate supplies at regional stores and inadequate transport of supplies.¹⁰⁶ This assessment showed that some form of antihypertensives was available at 92 percent of surveyed health facilities, and that methyldopa was still available at 74 percent.¹⁰⁶

Availability of medications was high, but quality of care was more variable. Provision of antihypertensives was documented in only 48 percent of preeclampsia and eclampsia cases, and only 26 percent of facilities reported providing any parenteral anticonvulsants such as magnesium sulfate in the preceding three months, likely reflecting missed opportunities.¹⁰⁶

Maternal Tetanus Vaccination

A worldwide Maternal and Neonatal Tetanus Elimination Initiative was launched in 1999 by UNICEF, WHO, and United Nations Population Fund (UNFPA), with the goal of reducing maternal and neonatal tetanus cases. The initiative’s 2000 strategy document identified 59 countries, including Ethiopia, that had not yet achieved universal neonatal tetanus elimination.¹⁰⁷

In fact, a 1999 WHO study had estimated that Ethiopia accounted for nearly 5 percent of all neonatal tetanus deaths globally, with approximately 18,000 cases, resulting in more than 13,000 deaths.⁶⁴

Though immunization with at least two injections of the tetanus toxoid vaccine had been integrated into ANC services for pregnant women in Ethiopia, coverage remained poor, likely due to low use of antenatal care and related health services. The 2000 DHS reported that 73 percent of women who had given birth in the preceding five years did not receive any injection with the tetanus toxoid vaccine during pregnancy. Only 17 percent had received the recommended minimum of two injections.⁹⁶

By this time, the FMOH had already begun collaborating with EPI partners to begin implementing supplementary immunization activities to administer tetanus toxoid vaccine in high-risk areas, in addition to continued provision during routine ANC visits. The goal was to administer three doses of the vaccine to all women of childbearing age in these areas, with the long-term aim of eliminating maternal and neonatal tetanus in Ethiopia.

To achieve this, the FMOH worked with Save the Children and UNICEF to implement a social mobilization program promoting tetanus toxoid immunization.¹⁰⁸ More than 15 million women of reproductive age in 59 high-risk zones were vaccinated during three rounds of tetanus toxoid supplementary immunization activities between 1999 and 2009. These activities achieved impressive coverage levels, ranging from 76 percent to 94 percent.²²

WHO recommends validation of maternal and neonatal tetanus elimination status once a country has implemented all planned activities and claims to have achieved elimination. In 2011, WHO and UNICEF conducted a pre-validation survey in Ethiopia. It concluded that all regions in the country, except Somali, had indeed eliminated maternal and neonatal tetanus.^107,109 A national validation occurred in 2016, after the end of the study period, when the Somali region was found to have eliminated the disease.¹⁰⁸

Delivery Services, Access, and Uptake

In 2000, the DHS reported that in the preceding five years, only 5 percent of infants had been born in a health facility. Furthermore, the likelihood of delivering in a facility varied greatly across socioeconomic categories – 32 percent of women in urban areas delivered in facilities, compared with only 2 percent of women in rural areas, and 41 percent of women with secondary school education delivered in facilities, compared with 7 percent with primary-level education and 2 percent with no education.⁹⁶

Ethiopia implemented strategies to improve the capacity of facility-based providers to prevent and manage complications during labor and delivery and to increase access to skilled birth attendants, primarily through primary health care facilities.

Sepsis Prevention

One potential consequence of low facility use is an increased risk of unsanitary birth conditions. In 2000, 9 percent of Ethiopia’s deaths within the first month of life were due to sepsis and 4 percent to tetanus, according to IHME.³⁰ Clean birth, delivery, and postnatal practices have been associated with decreases in neonatal mortality due to both of these causes. WHO established the following “six cleans” for healthy childbirth procedures:

• Handwashing of the birth attendant prior to birth
• Clean birth surface
• Clean perineum
• Cutting the umbilical cord using a clean implement
• Clean cord tie
• Clean cloth for drying¹¹⁰

Following the release of the Millennium Development Goals in 2000, the government of Ethiopia began to prioritize efforts to increase skilled delivery at birth to decrease maternal and child mortality.

The FMOH included clean and safe delivery services at homes and health posts as a primary component of the services that HEWs would provide when the Health Extension Program was introduced in 2003. The ministry’s Health Sector Strategic Plan III (2005–2010) also defined clean and safe delivery as a key high-impact health intervention to be integrated into service delivery systemwide. The government aimed to increase clean delivery births from 10 percent to 50 percent within five years.¹¹¹

Active Management of the Third Stage of Labor

Progress toward increasing clean deliveries could be seen in Ethiopia’s adoption of global standards to improve active management of the third stage of labor (AMTSL). In 2003, the International Confederation of Midwives and the International Federation of Gynecology and Obstetrics issued Preventing Postpartum Hemorrhage: Managing the Third Stage of Labor, declaring that all birth attendants should have the skills, knowledge, and critical judgment to carry out AMTSL.¹¹²

This protocol consists of three primary steps: administration of oxytocin or another uterotonic drug within one minute after birth, controlled cord traction, and uterine massage after delivery of the placenta as appropriate.¹¹³

An essential tool for AMTSL is a partograph – a single sheet of paper that captures such essential data as cervical dilation, fetal heart rate, and duration of labor. WHO produced a model partograph for improved management of labor in 1994 as part of the Safe Motherhood Initiative and recommended it for identifying slow labor progress and initiating appropriate interventions as needed.¹¹⁴

The FMOH collaborated with the Ethiopian Society of Obstetricians and Gynecologists and IntraHealth International in 2003 to launch a one-year project focused on reducing maternal morbidity and mortality through universal application of AMTSL to prevent postpartum hemorrhage.¹¹⁵ The following year, the FMOH and the Ethiopian Society of Obstetricians and Gynecologists released guidelines on AMTSL procedures.

A study in late 2005 assessed the use of AMTSL for facility-based deliveries in seven countries, including Ethiopia. It found correct use of AMTSL in only 4.5 percent of deliveries at the Ethiopian hospitals included in the study.¹¹⁶

A policy analysis conducted as part of the study found that while standard treatment guidelines included the definition of AMTSL, they provided conflicting guidance on how to provide it. The post-study analysis also found that while Ethiopia’s national guidelines featured AMTSL, the protocol was not included in the preservice curriculum of doctors, nurses, and midwives.¹¹⁶

The FMOH coordinated large-scale training of health care providers on AMTSL, and in 2006 USAID’s Prevention of Postpartum Hemorrhage Initiative found some signs of progress. A high proportion of observed deliveries provided controlled cord traction (70 percent) and uterine massage (72 percent). Complete and correct use of AMTSL, however, was observed in only 29 percent of deliveries.¹¹⁷

In 2010, the FMOH released the Management Protocol on Selected Obstetric Topics, which established AMTSL as standard care to be received by every woman delivering at a health facility.¹¹⁸

Despite this work, gaps in knowledge of key tools and training coverage remained. For example, partograph use remained low, as shown by a study conducted in early 2012. All providers knew what a partograph was, but only 57 percent used it and only 20 percent knew the correct function of a partograph alert line. Providers who reported awareness of the partograph but never used it commonly claimed that the records took too much time, were too complicated to administer, or were another provider’s responsibility. Some also reported not knowing enough about the forms to fill them in correctly.¹¹⁹

The 2014 ESPA+ survey reported continuing low levels of instruction in AMTSL; only 10 percent of interviewed providers reported receiving in-service training in the protocol.¹⁰

Yet there were signs of growing access to facility-based care, including in remote areas far from major hospitals. In 2013, the FMOH launched the Community-Based Newborn Care program as part of the Health Extension Program in collaboration with UNICEF, Last Ten Kilometers, the Integrated Family Health Program, and Save the Children.

The community-based program sought to increase the prevalence of safe and clean delivery practices at health centers and health posts. It operated in two phases, beginning in 2014.¹²⁰ By 2015, community-based newborn care services were provided in 70 percent of health posts in Ethiopia.¹²¹

Preterm Births

The value of facility-based delivery becomes especially apparent when the final stage of labor comes too soon. In 2000, nearly a third (32 percent) of all deaths among early neonates (age 0 to 6 days) in Ethiopia were related to preterm birth, according to an estimate from the University of Washington’s Institute for Health Metrics and Evaluation (IHME).³⁰

A decade later, preterm births accounted for 10 percent of all births in Ethiopia, according to UNICEF¹²² – slightly lower than the average for lower-income countries (12 percent).¹²³ In 2010, the FMOH released its Management Protocol on Selected Obstetrics Topics, a set of guidelines that recommends corticosteroids at tertiary referral hospitals in cases of preterm labor, except when there is evidence of chorioamnionitis.¹¹⁸

Antenatal corticosteroids (ACS) have been used since 1972 to accelerate fetal lung development and reduce risk of respiratory distress syndrome in pregnancies where a woman is considered to be at high likelihood of delivering significantly preterm. This practice has long been used in high-income countries prior to more recent introduction in low- and middle-income countries during the study period.¹²⁴

In Ethiopia, the use of ACS on a larger scale began in 2014 following the release of national guidelines recommending their use for mothers at risk of imminent preterm birth between 28 and 34 weeks of gestational age at all levels of the health system, and at all facility types except health posts. The guidelines included information to help determine whether a baby was preterm and whether a mother had begun labor; these determinations were among the main challenges that hindered the use of ACS in clinical trial settings.¹²⁵

In 2015, WHO recommended the administration of ACS worldwide for women at risk of preterm labor between 28 and 34 weeks of gestation under the proper medical circumstances.¹²⁶

Although national guidelines had permitted use of ACS across all levels of the public health care system, a 2016 FMOH assessment found that only 9 percent of health facilities provided ACS in the preceding three months. Of the 91 percent of facilities that had not provided ACS during that period, 36 percent cited a lack of corticosteroids as a reason for not administering ACS, and 23 percent cited insufficient training.¹⁰⁶

 

Skilled Birth Attendants

In addition to encouraging facility-based delivery and improving procedures, another international criterion for healthy childbirth practices is the presence of a skilled birth attendant. Here too, Ethiopia clearly had a lot of room for improvement in 2000. That year, only 10 percent of births were assisted by either a health professional or a trained traditional birth attendant.⁹⁶

In 2004, the WHO Regional Office for Africa convened a meeting to develop a “Road Map” for accelerated reduction of maternal and newborn morbidity and mortality on the continent. The 11 countries in attendance – including Ethiopia – created a framework for the development of midwifery care standards.¹²⁷

The FMOH’s Health Sector Strategic Plan III (2005–2010) called for an additional 1,300 nurse midwives per year, over a period of six years. This strategy aimed to improve the ratio from one midwife per 13,388 women of reproductive age to one per 6,759.¹¹¹ It also set a target to increase deliveries attended by skilled birth attendants from 12 percent to 32 percent.¹¹¹

A major factor in Ethiopia’s effort to encourage facility-based delivery and skilled birth attendance was the arrival in 2004 of the Health Extension Program’s first class of HEWs, who were trained initially to assist in deliveries. Over the study period, it became clear that this was not an optimal role for the HEWs, so their primary delivery-related task shifted from assisting directly in childbirth to increasing facility-based delivery through promotion and accompanying expectant mothers to health facilities.

The HEWs had a powerful incentive to perform this work – the quantity of home deliveries in their community was one of the basic criteria used to evaluate their performance. As a key informant from the FMOH put it, “If there are a number of home deliveries, that counts a lot against the health extension workers.”⁹⁸

The Health Management Information System (HMIS) began tracking the proportion of births attended by health extension workers at health posts – a figure that was reported on a monthly basis.¹²⁸ According to Ethiopia’s Health Sector Transformation Plan (2015–2020), HMIS data showed an increase in clean delivery due to the expansion of health centers, the strengthening of health center and health post networks, and an increased focus on provision of skilled care at birth.⁴

The FMOH’s Reproductive Health Strategy (2006–2015) called for increasing births attended by skilled health personnel at home or in a facility to 60 percent by increasing the number of trained health workers and improving access to facilities.¹²⁹

Prescribed steps toward this goal included equipping one health post per 5,000 people to provide essential obstetric and newborn care, and one health center per 25,000 to provide basic emergency obstetric and newborn care services. The strategy also included a plan to amend the midwifery curriculum so that new graduates would satisfy the requirements of “skilled birth attendants.”¹²⁹

In 2011, the FMOH initiated the Accelerated Midwifery Programme as part of the national Health Resource Strategy, with the support of UNFPA. This initiative trained clinical diploma-level nurses in a one-year midwifery program, using a curriculum centered on international standards established by WHO and the International Confederation of Midwives.

The program trained 4,461 new midwives over a three-year period, achieving 95 percent of its target.¹³⁰ By 2014, the number of midwives in Ethiopia had increased to 9,244.¹³¹

The FMOH had identified another, often overlooked, element of healthy childbirth practices: the availability of transportation to a hospital or birth facility.

The use of ambulances for deliveries was significantly associated with reduced pregnancy-related deaths; in districts found to have above-average ambulance use, the maternal mortality rate was 149 deaths per 100,000 live births, compared with 350 per 100,000 in those with below-average use.¹³² In response to this need, the FMOH invested about US$50 million to distribute 1,250 ambulances countrywide.¹³²

Emergency Obstetric and Newborn Care

In addition to making emergency medical transportation more readily available, Ethiopia took steps to improve the emergency care that infants and young children receive within hospitals and other health centers.

In 2008, the FMOH, UNICEF, WHO, UNFPA, and the Ethiopian Society of Obstetricians and Gynecologists began planning the largest national assessment of emergency obstetric and newborn care services yet conducted in Ethiopia, covering 806 hospitals, health centers, and higher-level clinics, in both the public and private sectors.¹³³

Just over a decade earlier, in 1997, WHO, UNICEF, and UNFPA introduced emergency obstetric and newborn care (EmONC) as a strategy to reduce both maternal and newborn mortality, particularly in low-resource settings.

EmONC comprises two subcategories: basic emergency obstetric and newborn care (BEmONC); and comprehensive emergency obstetric and newborn care (CEmONC). BEmONC consists of seven key services, or “signal functions,” used to classify health facilities:

• Administration of parenteral antibiotics
• Administration of parenteral anticonvulsants
• Administration of parenteral uterotonics
• Removal of retained products
• Assisted vaginal delivery
• Manual removal of the placenta
• Newborn resuscitation

CEmONC includes the seven signal functions in addition to blood transfusion and the capability to perform surgery (particularly delivery by cesarean section).¹³⁴

The 2008 assessment found large gaps in readiness. Of the hospitals studied, 51 percent were classified as fully functioning CEmONC centers; another 14 percent only met BEmONC criteria. Among health centers, 1.3 percent were classified as fully functioning BEmONC, and less than 1 percent as CEmONC.¹³³

In addition, WHO recommends at least five EmONC facilities per 500,000 people, but Ethiopia was found to have only 0.6 per 500,000 – and the shortfall was markedly worse outside of Addis Ababa and the surrounding center of the country.¹³³

In response, the FMOH partnered with local and international stakeholders in 2010 to develop a standardized, competency-based BEmONC training curricula for in-service training of midwives and nurses. In 2013, the FMOH began managing an initiative to upgrade the capability of health centers to provide BEmONC.

A 2015 assessment of 134 health centers found that their delivery rate for births in their catchment areas increased from 24 percent of local births in 2013 to 56 percent in 2015. However, gaps in coverage and readiness remained. Facilities in the review met only 16 percent of their communities’ anticipated need for BEmONC.¹³⁵

The following year, the 2016 Ethiopia EmONC Assessment – which collected data from 3,804 facilities across the country – reported progress across all EmONC indicators between 2008 and 2016. For example, the proportion of fully functioning EmONC facilities had risen from 11 percent in 2008 to 40 percent in 2016.¹⁰⁶

The assessment showed significant room for improvement, with only 18 percent of anticipated EmONC needs being met in the reviewed facilities. Additionally, this proportion varied greatly by region, ranging from just 3 percent in Gambella to 83 percent in Addis Ababa.¹⁰⁶

In 2014, the ESPA+ survey found that availability of normal delivery services was very high for health centers (99 percent) and all hospital types (ranging from 98 percent for primary hospitals to 88 percent for referral hospitals). Nearly all public facilities (99 percent) provided normal delivery services, defined as uncomplicated vaginal delivery for low-risk women.¹⁰ Furthermore, the FMOH was building more facilities to accommodate more births. By 2015, it had constructed 3,547 health centers throughout the country.⁴

Despite the widespread – and growing – availability of health centers and public facilities, the 2016 DHS reported that only 33 percent of Ethiopian women who had given birth in the preceding five years had done so at a health facility, with regional variation reaching as high as 97 percent in Addis Ababa and only 15 percent in Afar.¹⁸ While this represented a major increase since 2000’s rate of 5 percent, it was still low – and lower still in the rural areas where most Ethiopians live. In spite of HEWs’ efforts to promote facility-based delivery, only 26 percent of women in rural areas delivered at a facility, compared with 86 percent in urban areas.¹⁸

Equity plots show that facility-based delivery increased in all wealth quintiles from 2000 to 2016. However, progress was greatest in the wealthiest quintile, and the equity gap between the wealthiest and all other quintiles widened notably over the study period.

Cesarean Sections

The global standard for optimal national cesarean section rates is 10 to 15 percent. A lower rate indicates an unmet need among women requiring operative care to avoid maternal and newborn health conditions that can lead to mortality for both.

Ethiopia’s rate has remained far lower than this recommended range. The 2000 DHS reported that fewer than 1 percent of all births in Ethiopia in the five years preceding the survey had been delivered by cesarean section. Rates varied by region, with a high of 7.9 percent in Addis Ababa and a low of 0.1 percent in Amhara.⁹⁶

The low rates of cesarean section deliveries reflects longstanding shortages of surgical personnel in Ethiopia. As explained by one key informant,

The Health Sector Development Plan IV (2010–2015) set a target to achieve a national cesarean section rate of 7 percent by 2015.¹³⁶ An important component of this effort was a program that the FMOH and the Ministry of Education had launched the previous year – the Integrated Emergency Surgical Officers program, which created a cadre of midlevel, non-physician surgical providers capable of performing emergency surgical, gynecological, and obstetric procedures.

From the start, one of the primary objectives of the program was to improve access to cesarean sections. A key informant noted, “Initially the idea was to improve surgical access for obstetric care, and mainly what is meant by that is C-sections.”

A study conducted in 2014 assessed the clinical performance of emergency surgical officers working in eight hospitals in the SNNP Region of Ethiopia. Nearly two-thirds of all emergency operations performed at these hospitals between September and November 2014 were cesarean sections, all of which were performed by the integrated emergency surgical officers. Cesarean sections accounted for 12.5 percent of all deliveries during this period, within the globally accepted range at the time.¹³⁷

Efforts such as the Integrated Emergency Surgical Officers program have improved access to cesarean sections at health facilities, but national and some regional rates still fall short of global standards due to low use of facilities during delivery. The 2016 DHS reported that only 1.9 percent of live births at health facilities in the preceding five years were delivered by cesarean section.

A notably wide gap had opened up between urban and rural populations, with cesarean section rates of 10.6 percent among women in urban areas and only 0.9 percent in rural areas. In Addis Ababa, the rate was 21.4 percent. The economic element of this disparity was further highlighted by the high rate of delivery by cesarean section in Ethiopia’s private facilities, at 23 percent.¹⁸ Rates this far above the recommended range are indicative of excessive procedures; at this frequency, researchers have found no additional reductions in maternal or neonatal mortality rates.

The FMOH reported slightly higher cesarean section rates in 2016 – 3 percent for the country as a whole, and 35 percent for Addis Ababa.¹³⁸

Postnatal Care

The 2000 DHS found that 90 percent of mothers who had a live birth in the five years preceding the survey did not receive postnatal care services. The problem was especially acute in rural areas, where only 6 percent of mothers received postnatal care, compared with 74 percent in Addis Ababa.⁵²

Addressing Cultural Barriers

According to key informants, low use of postnatal care by rural women in Ethiopia may have been at least partially due to cultural beliefs and traditional practices, such as a 40-day confinement period intended to protect the mother and newborn from malevolent spirits.¹³⁹

The lack of postnatal care had a measurable impact on U5M. At the start of the study period in 2000, 34 percent of deaths among early neonates (age 0 to 6 days) in Ethiopia were attributable to birth asphyxia, according to 2017 estimates from IHME.¹⁴⁰

The FMOH’s Reproductive Health Strategy (2006–2015) aimed to reduce maternal and neonatal mortality by ensuring access to a core package of services, particularly in rural areas of the country with limited access to health facilities. This core package included postnatal care,¹²⁹ which the FMOH recommends at intervals of six hours, six days, and six weeks after birth.

As of 2016, however, 81 percent of Ethiopian mothers had not received any postnatal care.¹⁴¹ Attendance at the sixth-day visit is especially low, as it is typically not culturally acceptable for women to leave the home at that time. One key informant said, “In many parts of the country, they tend to keep them in [an] isolated room. . . . Coming out and going to facility for a checkup [during the isolation period] is unthinkable.”

To increase uptake of postnatal care in consideration of beliefs and practices, HEWs were instructed to make postnatal visits with all new mothers in a community, ideally within four hours of delivery. The health workers were trained to evaluate newborns for a variety of danger signs, including jaundice, fever, hypothermia, eye infections, respiratory distress, bleeding, or inadequate breastfeeding.¹⁴² In addition, HEWs conducted follow-up postnatal visits at two days, six days, and six weeks after delivery.⁹⁸

Neonatal Resuscitation

In 2010, the Maternal and Child Health Integrated Program evaluated care in 19 hospitals in Ethiopia with the highest birth volumes. It found that only 28 percent of surveyed providers had sufficient knowledge of neonatal resuscitation techniques, and only 18 percent had ever resuscitated a newborn. In addition, while all of the facilities stated that they performed resuscitation, only 77 percent had the equipment needed to do so effectively.¹⁴³

In 2010, the FMOH collaborated with the American Academy of Pediatrics, Global Development Alliance, and other partners to create a three-week training program that included Helping Babies Breathe,¹⁴⁴ an intervention that was developed under the leadership of the American Academy of Pediatrics to train birth attendants across the world in basic resuscitation and immediate care of newborns. The program was launched in Tanzania in 2009 and was then introduced worldwide.¹⁴⁵

A post-training study found that instruction in the Helping Babies Breathe intervention significantly improved neonatal resuscitation knowledge among participants, with 68 percent more likely to identify infants requiring resuscitation, 75 percent more likely to correctly choose to begin ventilation in an apneic baby, and 79 percent more likely to appropriately attempt to improve mask seal when the bag-mask ventilation does not produce adequate chest rise.¹⁴⁶

The reach of this training program was limited, however. In 2014, the ESPA+ survey reported that 43 percent of providers in health posts reported receiving in-service training on neonatal resuscitation, a rate that dropped to 17 percent of surveyed providers in other facilities.¹⁰

Regarding birth asphyxia, IHME estimated a decrease in the rate of death due to birth asphyxia, as well as declines in the estimated number of deaths – 27,726 deaths in neonates younger than 7 days old were due to birth asphyxia in 2015, compared with 34,266 in 2000.¹⁴⁰

A 2016 assessment of more than 3,800 facilities across Ethiopia found that 70 percent were supplied and staffed for resuscitation, an increase of 47 percentage points from the 2008 baseline assessment. In addition, 71 percent of health centers had equipment for resuscitation in 2016, a substantial increase from 38 percent in 2008.¹⁰⁶

Even more impressive was the elimination of the training gap that had been identified as a barrier in 2008. By 2016, 98 percent of facilities reported having at least one staff member capable of providing neonatal resuscitation.¹⁰⁶

Neonatal Intensive Care

For a large part of the study period, the availability of intensive care for neonates was very limited in Ethiopia, with a scarcity of space available for neonatal intensive care units (NICUs). Only a handful of hospitals in larger cities had such facilities. According to one key informant, hospitals elsewhere typically turned away sick newborns requiring intensive care.

Following the 2003 Lancet series on child survival, the sponsoring of a handful of small hospitals by partners to build NICUs, and advocacy by partners, the FMOH began to prioritize NICUs as a strategy to decrease neonatal mortality.

As a result of advocacy by UNICEF and the Ethiopian Pediatric Society for prioritization of neonatal health in Ethiopia, the FMOH introduced the country’s first NICU initiative as part of the Health Sector Development Plan IV (2010–2015).¹⁴⁷

This initiative included a package of interventions to reduce neonatal mortality and morbidity through improved management of neonatal complications, enhanced infrastructure, more accessible equipment, and more robust referral linkages among facilities. It established three levels of NICUs, establishing standards corresponding to the size and capabilities of various types of health facilities.¹⁴⁸

In 2014, the FMOH worked with partners such as the Ethiopian Pediatrics Society and WHO to develop protocols and training materials for NICUs. By mid-2015, 100 percent of target hospitals had received NICU training, and 49 percent of the targeted number of NICUs were operational.¹⁴⁸

Despite these and other interventions, the 2016 DHS found that only 19 percent of mothers who gave birth within the preceding two years had received postnatal care – only a slight increase from the 2000 DHS.

For women who had given birth outside of facilities, the rate was worse still – just 1.9 percent had received postnatal care, compared with 42 percent for those who had delivered at a facility. Only 27 percent of newborns born within the preceding two years had at least two signal functions of postnatal care performed within two days of birth. Regional differences in coverage were also seen, ranging from 9 percent in Oromia Region to 55 percent in Addis Ababa.¹⁸

Kangaroo Mother Care

Kangaroo mother care (KMC) – skin-to-skin contact between the mother and her newborn as soon as possible following birth – began in Bogota, Colombia, in 1979 as a simple, low-cost intervention for low-weight infants.¹⁴⁹ Ethiopia’s largest public hospital, the Tikur Anbessa (Black Lion) Specialized Hospital in Addis Ababa, introduced KMC in 1996 after one of the hospital’s physicians learned about the practice in Mozambique.¹⁵⁰

As an early adopter, Tikur Anbessa participated in a large multi-center study that included facilities in Mexico, Brazil, India, and Italy, as well as Ethiopia. This study found KMC to be an effective intervention for newborns, and the FMOH soon funded a randomized controlled trial to examine the effectiveness of KMC over a 12-month period during 2001 and 2002.^151,152

The study found lower mortality rates among infants receiving KMC (23 percent) than among those who received the conventional method of care (38 percent). Over 90 percent of mothers of infants included in the study reported that they were happy to care for their babies using the KMC method. Findings from the study provided the FMOH with evidence that KMC was effective, safe, feasible, and acceptable at health facilities in Ethiopia .¹⁵³

KMC was incorporated into NICU services and trainings, and the practice expanded across Ethiopia along with the NICU network. Due to this integration, KMC was expected to be provided at all NICUs. A key informant said,

Despite this progress, a 2014 Save the Children assessment found large gaps in KMC readiness among hospitals and health centers in the Oromia and SNNPR regions. While all of the hospitals in the survey reported providing KMC services within the preceding three months, the study showed that only 14 percent of eligible babies born in these hospitals had been enrolled in KMC, and fewer than 20 percent of delivery attendants in the facilities were found to be trained in KMC.¹⁵⁰

These gaps were also reflected in the 2014 ESPA+ survey, which found that although 67 percent of facilities (excluding health posts) providing normal delivery services also provided KMC, only 15 percent of providers reported receiving in-service training in the practice.¹⁰

One key informant identified the lack of adequate facilities as an ongoing challenge to the provision of KMC in Ethiopia:

Neonatal Sepsis

In 2000, an estimated 9 percent of deaths in neonates were caused by sepsis.³⁰ Though the FMOH promoted clean delivery practices during the study period, management of neonatal sepsis was limited.

In September 2012, the FMOH agreed to introduce community-based management of newborn sepsis through the Health Extension Program. The following year, the ministry announced new national guidelines for addressing the condition, as well as the new Community-Based Newborn Care collaboration to improve antenatal, intrapartum, postnatal, and newborn care – including management of neonatal sepsis.¹⁵⁴

By 2015, community-based newborn care services were provided in 70 percent of health posts, expanding provision of neonatal sepsis treatment in Ethiopia.¹²¹ In November of that year, an assessment found high rates of adherence to protocols for less complicated cases – 95 percent of infants classified as having a local bacterial infection were treated with amoxicillin.¹²⁰

Quality of care, however, was lower for more complex cases including management of possible bacterial sepsis of newborns. Community-based newborn care protocols for HEWs’ treatment of neonates with severe bacterial infection included pre-referral doses of amoxicillin and gentamycin, but only 41 percent of people with very severe disease seen by HEWs at health posts received both antibiotics.¹⁵⁵

Only 6 percent of infants with very severe infections were referred to a higher-level facility without a pre-referral dose of antibiotics .¹²⁰

Decomposition

A decomposition analysis undertaken in collaboration with the Institute for Health Metrics and Evaluation (IHME) at the University of Washington identified the top interventions and risk factors that contributed to reductions in under-five deaths in Ethiopia. These interventions and risk factors can be seen in the bottom two bars of the visualization below.

• Some of the largest contributors to reduction in under-five mortality were health system interventions, which were responsible for 51 percent of the reduction in under-five deaths from 2000 to 2017.
• Within this, vaccines (especially measles first dose vaccine, DTP3, and Hib vaccine) were a significant factor, and were attributed to 21.4 percent of the reduction.
• Beyond health system interventions, reductions in child growth failure were found to be responsible for 29 percent of the reduction in under-five deaths over this time period.¹⁵⁶