The new WHO decision-making framework on vaccine use in acute humanitarian emergencies: MSF experience in Minkaman, South Sudan

Introduction The main causes of death during population movements can be prevented by addressing the population’s basic needs. In 2013, the World Health Organization (WHO) issued a framework for decision making to help prioritize vaccinations in acute humanitarian emergencies. This article describes MSF’s experience of applying this framework in addition to addressing key population needs in a displacement setting in Minkaman, South Sudan. Case description Military clashes broke out in South Sudan in December 2013. By May 2014, Minkaman, a village in the Lakes State, hosted some 85,000 displaced people. MSF arrived in Minkaman on 28 December 2013 and immediately provided interventions to address the key humanitarian needs (health care, access to drinking water, measles vaccination). The WHO framework was used to identify priority vaccines: those preventing outbreaks (measles, polio, oral cholera vaccine, and vaccine against meningococcal meningitis A (MenAfrivac®)) and those reducing childhood morbidity and mortality (pentavalent vaccine that combines diphtheria, tetanus, whooping cough, hepatitis B, and Haemophilus influenzae type B; pneumococcal vaccine; and rotavirus vaccine). By mid-March, access to primary and secondary health care was ensured, including community health activities and the provision of safe water. Mass vaccination campaigns against measles, polio, cholera, and meningitis had been organized. Vaccination campaigns against the main deadly childhood diseases, however, were not in place owing to lack of authorization by the Ministry of Health (MoH). Conclusions The first field use of the new WHO framework for prioritizing vaccines in acute emergencies is described. Although MSF was unable to implement the full package of priority vaccines because authorization could not be obtained from the MoH, a series of mass vaccination campaigns against key epidemic-prone diseases was successfully implemented within a complex emergency context. Together with covering the population’s basic needs, this might have contributed to reducing mortality levels below the emergency threshold and to the absence of epidemics. For the WHO framework to be used to its full potential it must not only be adapted for field use but, most importantly, national decision makers should be briefed on the framework and its practical implementation.


Background
By late 2016, more than 40 million people worldwide had been internally displaced, mainly as a result of conflicts. This is the highest number ever reported [1]. Overcrowding, high population density, rudimentary or inappropriate shelters, poor access to water and sanitation, and poor nutritional status due to food insecurity, combined with limited access to preventive and curative health care, increase the risk of the spread of infectious diseases. This can result in higher morbidity and mortality, especially in camp settings [2]. Improvements in the response to complex emergencies have focused mainly on the prevention and control of malnutrition and diseases with epidemic potential, leading to a decline in morbidity and mortality in camp settings when effective assistance is provided [3]. However, the main causes of death (typically diarrheal diseases, respiratory infections, measles, and malaria where endemic) are still common and can be easily prevented or treated [4].
Emergency humanitarian responses to situations involving displaced people are focused on preventing excess mortality by addressing essential needs based on the "MSF Top Ten Priorities" (Table 1) framework drawn up by Médecins Sans Frontières (MSF) in the late 1990s [5] or SPHERE standards [6].
In 2013, the World Health Organization (WHO) issued a framework for decision making to help prioritize vaccinations in acute humanitarian emergencies [7]. In addition to vaccination against measles (a major cause of death in humanitarian disasters [8] and part of a standard emergency response [5,9]), the tool proposes a systematic framework to assess: 1) the risk of vaccine-preventable, outbreak-prone diseases such as polio, cholera, meningococcal meningitis, and hepatitis E; and 2) the main deadly childhood diseases, by including new and underutilized vaccines, notably pentavalent (diphtheria, tetanus, whooping cough, hepatitis B, Haemophilus influenzae type B), pneumococcal, and rotavirus vaccines. The decisionmaking process consists of three steps: 1) epidemiological risk assessment to determine and grade the risk of each vaccine-preventable disease (VPD); 2) assessment of the main characteristics of the vaccines (suitability, availability, affordability, etc.) and their amenability to mass vaccination campaigns; and 3) assessment of contextual considerations.
During the recent large-scale displacement in Minkaman, South Sudan, MSF used the WHO decision-making framework to define priority vaccinations in addition to addressing essential needs of the population. To our knowledge, this is the first utilization of the WHO framework. Here we describe our experience of applying it during an acute emergency and suggest some recommendations for future use.

Setting and context
After decades of civil war, South Sudan is still undergoing a political-military crisis with severe humanitarian consequences. South Sudan has some of the poorest health indicators in the world and is prone to outbreaks of communicable diseases. In addition, the health system has limited capacity to respond to emergencies without external aid; 80% of health care is provided by non-governmental organizations [10]. In 2013, almost half of the population lived below the poverty line, and only 57% of people had access to an improved water source [11].
In December 2013, armed conflict led to a massive population displacement in South Sudan. Many people fled fighting around the city of Bor in Jonglei State towards Minkamana village in South Sudan (Lakes State) located approximately 20 km south of Bor, on the river Nile. Minkaman itself was not directly affected by the violence ( Fig. 1, map). Before the crisis, it had an estimated population of 7000 inhabitants. The influx of newly displaced people reached 85,000 in 5 months [12]. The displaced population was widely spread in a swamp-like area along the Nile and had limited access to clean water and sanitation, and the existing infrastructure was unable to meet the needs of the increasing population.

Description of the intervention
The MSF team arrived in Minkaman on 28 December 2013. After meeting with the local authorities and other humanitarian actors present, the team immediately launched an emergency intervention focused on healthcare provision and access to safe drinking water. MSF did not prioritize food, distribution of non-food items, or delivery of shelter, as these were planned by other partners.
A chronogram of MSF's intervention is shown in Fig. 2. Primary health care was organized as soon as the team arrived, with two outpatient clinics being opened. Curative care was soon expanded, with hospitalization capacity and a nutritional program added. To address the high burden of watery diarrhea, several oral rehydration points were set up in the community and lasted until the end of March 2014, when the number of cases of watery diarrhea declined.
The water treatment station, water trucking, and bucket chlorination were functional within 3 days of MSF's arrival. By 3 January 2014, five liters of treated water/per person/per day was being provided to the displaced population, an acceptable quantity in the initial phase of an emergency [8]. After a pipeline designed for adaptation to population movements was set up, the amount of water distributed doubled. The pipeline also reduced dependence on trucks (a costly solution unable to reach all sites in the rainy season). A community-based surveillance system for mortality was established. The internally displaced population camp was divided into eight sections ( Fig. 1, map) covered by 60 community health workers (CHW) who collected weekly information about numbers of deaths from the heads of households and community leaders (without verbal autopsy). CHW also conducted active case finding for malnutrition (based on mid-upper arm circumference (MUAC) screening) and directed and referred sick patients to the health facilities.
By July 2014, more than 30 non-governmental organizations and several UN agencies were present in the area.
Consequently, MSF decided to hand over its activities and close the Minkaman project on 15 October 2014.

Prioritizing vaccines and use of WHO framework
We used the above mentioned WHO framework to define priority vaccinations to address the risk of vaccinepreventable outbreaks and VPD causing the highest morbidity and mortality. We used the three-step approach shown in Fig. 3. We assessed the general risks first and then the risk for each specific VPD. The major general risk factors were overcrowding and extremely precarious access to water and sanitation, in addition to high birth rates in a Table 1 The Top Ten Priorities for Refugee Health 1.Initial assessment Quantitative and qualitative information on background to the displacement, population, risk factors related to the main diseases and requirements in terms of resources through observation, interviews, sample surveys, mapping.
Usually approximate, results may need to be corroborated later.
2.Measles immunization Displacement, overcrowding and poor hygiene are factors that encourage emergence of large scale epidemics.
Mass vaccination of children from 6 months to 15 years should be a priority during the first week.

3.Water and sanitation Prevention of diarrhoeal diseases and survival
Ensure immediate provision with temporary water supply until more permanent solutions (wells) can be found Indicators in regard to water supply and latrines must be monitored.

4.Food and nutrition
Malnutrition is often associated with displacement Provision of food ration to cover daily minimum needs Feeding programs for specific groups are supplementary feeding for moderately malnourished and therapeutic or intensive feeding for the severely malnourished.   rounds. In order to optimize the campaigns, we combined polio with measles vaccine in the first round, and MenAfriVaq® vaccine was administered together with the second dose of oral cholera vaccine (OCV).
Lastly, we looked at contextual factors that appeared favorable for vaccine use, including the feasibility of accessing the vaccines and implementing sequential campaigns in the closed setting of Minkaman. A proposal was made to the Ministry of Health (MoH) to provide oral cholera, MenAfriVac®, pentavalent, PCV, and rotavirus vaccines to the population in a series of vaccination campaigns combining different vaccines ( Table 2). MSF conducted preventive vaccination campaigns against measles, polio, cholera, and meningitis A from January to March 2014. Measles vaccine was administered concomitantly with polio vaccine to 13,256 children aged under 5; 54,415 people were vaccinated against cholera, and 32,681 people aged 1-30 years were vaccinated against meningitis A. To monitor and evaluate the intervention, we conducted a cross-sectional population-based survey between 3 and 9 May 2014 to estimate vaccination coverage and examine mortality rates. Two stage cluster sampling was used.

Discussion
The MSF intervention in Minkaman followed MSF's "Ten Top Priorities", which have been guiding its emergency interventions since the late 1990s. In addition, we used the new WHO framework for prioritizing vaccines in acute emergencies as a complementary tool, with the aim of rapidly reducing the risk of vaccine-preventable outbreaks and reducing the morbidity and mortality due to the main vaccine-preventable diseases during a period of extreme vulnerability.
The use of oral cholera vaccine has already been discussed in South Sudan, and several pre-emptive campaigns were planned in the same period [14]. Similarly, the preventive use of MenAfriVac® was quickly approved, and the vaccine could be delivered at the same time as the second dose of OCV.
We successfully organized a series of vaccination campaigns targeting outbreak-prone diseases (measles, polio, cholera, and meningococcal meningitis) in a complex emergency. This was the first time that the MenAfriVac® vaccine had been used in a pre-emptive mass campaign in a complex emergency, and the first time that it was delivered concomitantly with OCV. Although acceptable, the OCV coverage was lower than in some of the other campaigns taking place at the same time [14], probably at least partially owing to high mobility of the population; high population mobility can also partially explain the low coverage for other antigens. We demonstrated that, despite logistic constraints, conducting several mass vaccination campaigns in displacement situations is feasible, even within a highly mobile population as is the case in South Sudan. However, we were unable to obtain authorization from the MoH for use of vaccines against the main deadly childhood diseases (pentavalent, pneumococcal, and rotavirus vaccines). None of these vaccines had been introduced in a routine immunization program at the time in South Sudan, although preparations were ongoing for the introduction of pentavalent vaccine into the Expanded Program on Immunization. This might have contributed to the reluctance of health authorities, which prioritized routine activities over emergency mass campaigns. However, pentavalent and pneumococcal vaccines have both been previously used in complex emergency situations n South Sudan, although in a setting of Sudanese refugees in Yida [15]. It is also likely that MSF became less inclined to pursue negotiations once authorization had been obtained for organizing the prevention of outbreak diseases and the situation in Minkaman stabilized.
The use of the framework is complex and was not easy to interpret during our Minkaman intervention. The first part (epidemiological risk assessment) provides some quantitative guidance for decision making (Fig. 3), but step 2 and particularly step 3 are qualitative, which makes summarizing findings into decisions difficult. To be widely used as guidance in emergency situations by response teams, the tool needs to be reviewed and simplified. However, work is underway to improve the framework and provide implementation guidance [16], which will probably address this point. Currently, national authorities and partners in the

Conclusion
In addition to mass vaccination campaigns targeting diseases with epidemic potential (measles, polio, meningitis and cholera), we propose a series of campaigns with new and underutilized vaccines (pentavalent, pneumococcal and rotavirus) targeting the most common childhood vaccine preventable diseases AND follow up with routine vaccination activities. We believe such vaccination campaign achieving high coverage in a displaced population can have a very important impact on childhood morbidity and mortality. The 3 rounds of campaigns necessary are feasible in this setting with logistic and human resources available.
field are not necessarily aware that the framework exists, which compromises its wider uptake, particularly during acute emergencies with limited time for introducing new strategies. WHO and other health actors must proactively introduce the framework to national health authorities at local, regional, and national level and conduct real-time practical exercises with the MoH and health partners if it is to be used to its full potential. Closer collaboration between humanitarian actors such as MSF and WHO will be helpful when lobbying national authorities. Lastly, more flexibility from the MoH is needed when some of the vaccines have not yet been introduced in the EPI, to enable implementation of preventive campaigns in humanitarian emergencies.

Conclusion
We describe the first field use of a new WHO framework for prioritizing vaccines in acute emergencies. Although we were unable to implement the full package of priority vaccines, we successfully implemented a series of mass campaigns against key epidemic-prone diseases in a complex emergency context. Together with covering the population's basic needs, this might have contributed to maintaining mortality levels below the emergency threshold and to the prevention of epidemics. For the WHO framework to be used to its full potential, it has to be adapted for field use, but most importantly its existence must be disseminated to national decision makers.