UCL - Université catholique de Louvain

Faculté de Santé publique

MASTER EN SCIENCES DE LA SANTE PUBLIQUE

THE EFFECT OF PRE-EMPTIVE ORAL CHOLERA

VACCINATION ON DISEASE SEVERITY AND
FATALITY DURING AN OUTBREAK OF CHOLERA
IN SOUTH SUDAN

Finalité : spécialisée

Option : Erasmus Mundus Master in Public Health in Disasters

Mémoire présenté par

Bekolo Cavin Epie

Promoteurs:

Professor Debarati Guha-Sapir

Dr Joris van Loenhout

Année académique : 2014 - 2015

 CERTIFICATION

This thesis entitled:

THE EFFECT OF PRE-EMPTIVE ORAL CHOLERA

VACCINATION ON DISEASE SEVERITY AND
FATALITY DURING AN OUTBREAK OF CHOLERA
IN SOUTH SUDAN

is my own work.

All sources of information reported by others are indicated in the list of references
in accordance with the guidelines.

Signature:

Total word count:

I, _______________________________ approve this thesis for submission.

Supervisor’s signature

2
 Table of Contents
CERTIFICATION...................................................................................................................2
Table of tables......................................................................................................................4
Table of Figures...................................................................................................................4
ABSTRACT.............................................................................................................................5
INTRODUCTION...................................................................................................................6
Review of the literature of Cholera....................................................................................10
History of Cholera.........................................................................................................10
Causative Agent.............................................................................................................10
Global Cholera Epidemiology.......................................................................................11
Cholera in Africa...........................................................................................................12
Cholera in complex humanitarian emergencies.............................................................13
Pathophysiology, Diagnosis, Treatment and Prognosis of Cholera...............................13
Prevention of cholera.....................................................................................................18
Cholera Vaccines2..........................................................................................................18
AIMS AND OBJECTIVES...................................................................................................21
Research Questions/Hypotheses........................................................................................21
Aims and Objectives..........................................................................................................21
Expected results/implications............................................................................................21
METHODOLOGY................................................................................................................22
Setting............................................................................................................................22
Study design...................................................................................................................23
Participants.....................................................................................................................23
Data Collection..............................................................................................................24
Ethical consideration......................................................................................................24
Data analysis:.................................................................................................................24
RESULTS..............................................................................................................................26
Outbreak description..........................................................................................................26
Person.............................................................................................................................26
Place...............................................................................................................................26
Time...............................................................................................................................27
Clinical presentation and diagnosis...............................................................................28
Case management and outcome.....................................................................................28
Effect of oral cholera vaccine and other factors on severity of cholera.............................30
Effect of oral cholera vaccine and other factors on cholera case fatality..........................32
DISCUSSION........................................................................................................................34
Description of the outbreak................................................................................................34
Effect of oral cholera vaccine and other factors on severity of cholera.............................35
Effect of oral cholera vaccine and other factors on cholera case fatality..........................37
CONCLUSIONS AND RECOMMENDATIONS................................................................40

3
ACKNOWLEDGMENTS.....................................................................................................42
BIBLIOGRAPHY..................................................................................................................43

Table of tables

Table 1. Diagnostic tests........................................................................................................16

Table 2. Characteristics of WHO pre‐qualified oral cholera vaccines.................................20
Table 3. Baseline characteristics of cholera cases.................................................................29
Table 4. Logistic regression model of factors associated with severe diarrhoea...................31
Table 5. Logistic regression model of factors associated with cholera death........................32

Table of Figures

Figure 1. Map of South Sudan showing cholera affected counties between April and July
2014.......................................................................................................................................26
Figure 2. The epidemic curve................................................................................................27

4
 ABSTRACT
Background
The use of oral cholera vaccine (OCV) as a complementary control measure against cholera
is increasing worldwide and will continue to expand with the availability of the global OCV
stockpile. To encourage its use and acceptance, more knowledge about its effectiveness is
needed. We assessed the efficacy of OCV in reducing disease severity and cholera deaths.
Setting
The study was carried out in the Republic of South Sudan that experienced a large outbreak
of cholera in 2014 a few months after a pre-emptive vaccination campaign was carried out
amongst internally displaced persons and their host population by making use of the global
OCV stockpile for the first time.
Methods
A retrospective cross-sectional survey of health facility cholera cases occurring between the
17th and the 30th epidemiological weeks in 2014. Data about vaccination status and other
putative exposure variables were collected. Our outcome variables were severe cholera
defined as body water volume depletion of over 10% and case fatality defined as death of a
cholera case. We used multiple logistic regression to determine the likelihood of presenting
with severe disease or of dying from cholera among vaccinees compared to the
unvaccinated cases.
Results
Included were 4115 cases of whom, 1907 (46.4%) were women and 900 (22.0%) were

children below five years old. The total number of severe cholera cases was 1946
(48.3%) and death was observed in 62 cases (1.9%). There was strong evidence that
vaccination with two doses of OCV was associated with a 71% efficacy[adjusted odd ratio
(aOR) = 0.29, 95%CI: 0.14 - 0.62, p = 0.001] in reducing the odds of suffering severe
cholera but there was no evidence of a direct effect of OCV on case fatality rate
(aOR=4.50, 95%CI: 0.40-50.42, p=0.222). Severe cholera (aOR=5.64, 95%CI: 2.55-
12.47, p<0.001) and age≥50 years (aOR=3.45, 95%CI: 1.56 – 7.64, p=0.002) were
the main determinant of high cholera mortality.
Conclusion
We provide early evidence that oral cholera vaccination with Shanchol® is effective in
reducing the severity of breakthrough disease among vaccinees and thus can indirectly
reduce the risk of dying from cholera. Stakeholders should use this benefit of OCV as
argument to roll out OCV use where and when necessary.
Key words
Oral cholera vaccine, severe cholera, case fatality ratio

5
 INTRODUCTION
Cholera is a secretory diarrhoeal illness caused by ingestion of food or water
contaminated by the gram-negative bacterium Vibrio cholerae of the O1 and O139
serogroups. Secretion of the cholera enterotoxin leads to continued activity of
adenylate cyclase in intestinal epithelial cells and secretion of water and
accompanying salts into the gut lumen. Cholera is an extremely virulent disease that
affects both children and adults and can kill within hours if left untreated. The short
incubation period of two hours to five days, enhances the potentially explosive
pattern of outbreaks. Since 1817, the world has faced seven cholera pandemics; and
the problem remains one of public health significance worldwide. Every year, there
are an estimated 3–5 million cholera cases and 100 000–120 000 deaths due to
cholera[1]. The World Health Organisation (WHO) estimates that the officially
reported cases represent only 5–10% of the actual number occurring annually
worldwide.

The main reservoirs of V. cholerae are people and aquatic sources such as brackish
water and estuaries, often associated with algal blooms. Recent studies indicate that
global warming creates a favourable environment for the bacteria. Cholera
transmission is closely linked to inadequate environmental management. Typical at-
risk areas include peri-urban slums, where basic infrastructure is not available, as
well as camps for internally displaced people or refugees, where minimum
requirements of clean water and sanitation are not met. Cholera is thus a key
indicator of lack of social development. While cholera was largely eliminated from
industrialized countries by water and sewage treatment over a century ago, today it
remains a significant cause of morbidity and mortality in many areas of Africa and
Asia. The greatest reported burden of disease is in the African continent, particularly
in central sub-Saharan regions. In 2013, a total of 56 329 cases were reported from
Africa, a 52% decrease compared with 2012 (117 570 cases). In 2013, cases from
Africa represented 43.6% of the total of 129064 cases, a low proportion compared
with the high 93%–98% of total cases reported during the 2001–2009 period. This
reduced proportion of African cases has been observed since 2010 and is linked to a
significant decline in cases from this continent since 2011 as well as the still
ongoing outbreak in Hispaniola[2]. Cholera occurs in both endemic and epidemic
patterns. In endemic regions, the highest rates of infection are in children under the
age of 5 years. In non-endemic areas, adults are equally as affected as children.
6
Attack rates are high, although only a small proportion (~25%) of those infected will
become ill[1]. Among people who develop symptoms, 80% have mild or moderate
symptoms, while around 20% develop acute watery diarrhoea with severe
dehydration. A high bacterial load, low background immunity and presence of
malnutrition are risk factors for a severe clinical picture.

Cholera is an easily treatable disease. Up to 80% of people can be treated

successfully through prompt administration of oral rehydration salts. Very severely
dehydrated patients require administration of intravenous fluids. Such patients also
require appropriate antibiotics to diminish the duration of diarrhoea, reduce the
volume of rehydration fluids needed, and shorten the duration of V. cholerae
excretion. Mass administration of antibiotics is not recommended, as it has no effect
on the spread of cholera and contributes to increasing antimicrobial resistance. With
good or adequate fluid replacement (oral or intravenous), mortality from cholera is
reduced from 40% to about 1%[3]. Vibrio cholerae is autochthonous to aquatic
environments, hence, it cannot be eradicated but hydroclimatology-based prediction
and prevention is an achievable goal [4]. Access to clean water and adequate
sanitation remain the mainstays of preventing both endemic cholera and cholera
outbreaks, together with health education to promote the adoption of appropriate
hygiene practices.

There are two types of safe and effective oral cholera vaccines currently available on
the market. Both are whole-cell killed vaccines, one with a recombinant B-sub unit,
the other without. Both have sustained protection of over 50% lasting for two years
in endemic settings. Both vaccines are WHO-prequalified and licensed in over 60
countries. Dukoral® has been shown to provide short-term protection of 85–90%
against V. cholerae O1 among all age groups at 4–6 months following
immunisation. The other vaccine (Shanchol®) provides longer-term protection
against V. cholerae O1 and O139 in children under five years of age. Both vaccines
are administered in two doses given between seven days and six weeks apart. There
is evolving evidence that the vaccine also provides herd protection by interrupting
transmission such that high levels of immunisation could provide even higher
protection to populations at risk[5]. Cholera vaccination is a safe and an effective
additional tool that should be used under the right conditions to supplement existing
priority cholera control measures [6, 7]. In 2012, a technical working group took an

7
important step in the direction of coordinated global cholera management by
recommending the creation of a stockpile of 2 million doses of oral cholera vaccine
(OCV) for use in emergencies, following the adoption in 2011 of resolution WHA
64.15 by the 64th World Health Assembly[8]. Such a stockpile improved the rapid
access to OCVs for countries that may benefit from their use in cholera control. For
instance, in 2014 in South Sudan, in the context of an ongoing conflict and dire
humanitarian conditions of overcrowding and inadequate sanitation among IDPs
(Internally Displaced Persons), the Ministry of Health together with WHO and other
partners decided to implement a pre-emptive mass vaccination campaign using for
the very first time oral cholera vaccine (OCV) from the global stockpile to
complement other cholera prevention and control measures. The campaign that was
held between February and April 2014 attained an overall coverage of about 60% to
85% reaching 166 000 refugees and 40 000 host population who were at very high
risk of morbidity and mortality [9, 10].

However, vaccination was not the magical solution because on the 15 th May 2014,
South Sudan's Ministry of Health declared a cholera outbreak in Juba, Central
Equatoria State, after 18 suspected cholera cases and one death had been
reported[11]. Cholera is endemic in South Sudan, and major out-breaks had been
experienced in the past years. This latest outbreak occurred in a context
characterised by high rates of malnutrition due to unusually poor harvest, limited
humanitarian access owing to continuing conflict and associated population
movement, targeted destruction of healthcare facilities and killing of patients and
healthcare workers. The humanitarian situation and many other key indicators in the
world’s newest country were already dire and access to sparse healthcare facilities
was limited, with more than 80 percent of medical services provided by
international organisations[12].

In this particular context, one would expect that morbidity and mortality due to
cholera to be high owing to the complex emergency and limited humanitarian
assistance on one hand; but on the other hand we would expect to see a mitigating
effect as well following the pre-emptive vaccination and other control measures
provided by both local health facilities and international humanitarian agencies. To
assess the impact of this outbreak, we sought to retrospectively investigate what
factors in general affected morbidity and mortality, with a particular emphasis on

8
the likely contribution of the oral cholera vaccination and the relative performance
of international humanitarian organisations. We hope that the findings and lessons
learnt may be useful for informing stakeholders about what forms of interventions
worked better and on whom and where further resources should go to, given that the
risk of another outbreak is persistent as the crisis is ongoing.

9
 Review of the literature of Cholera
History of Cholera
In the Indian subcontinent, Visuchica a Sanskrit word describing a disease that
closely resembles a case of cholera has been mentioned in Sushruta Samita which is
estimated to have been written between ~400 and 500 BC[13]. When the Portuguese
landed in the south western or Malabar Coast of India in 1498, cholera existed in the
region. In 1503, about 20,000 men of Calicut died of an epidemic of the disease
having symptoms resembling cholera as described by Gasper Correa, an officer of
Vasco da Gama[14]. The modern history of cholera began in 1817 when a violent
epidemic of the disease broke out in the Ganges River Delta region of Bengal[15]. F.
Pacini in 1854 was the first scientist to isolate Comma bacillius, now known as
Vibrio cholerae followed by a similar discovery by Robert Koch in 1884[16]. John
Snow, a British physician, was the first to discover that cholera spread through
contaminated water. The ongoing seventh pandemic of cholera, began in Indonesia
in 1961 and spread through Asia to Africa, Europe, and Latin America, has been
reported in over 50 countries and affected over 7 million people [17].
Causative Agent
V. cholerae is a member of the Vibrionaceae family of curved, Gram-negative rods
that are found in coastal waters and estuaries[3]. These organisms grow best in the
presence of salt, although V. cholerae can grow in water of lower salinity when it is
warmer and contains sufficient organic nutrients and is often associated with
zooplankton and shellfish in water[18], and is capable of utilizing chitin as a carbon
and nitrogen source[19]. In water, they enter a viable but non culturable form, also
called active but non-culturable or conditionally viable environmental cells [20].
V. cholerae is classified into more than 200 serogroups based on the O antigen of
the lipopolysaccharide [21]; of these, only O1 and O139 serogroups cause epidemic
cholera. V. cholerae O1 is further classified into two biotypes, classical and El Tor
(3). There are two major serotypes, Ogawa and Inaba, which vary in prevalence over
time [22]. In 1992, V. cholerae O139 was first recognized in South Asia as a cause
of epidemic cholera[23]. This organism is derived from V. cholerae O1 El Tor by
lateral transfer of a genomic island substituting the O139 for the O1 antigen, but is
otherwise virtually identical to V. cholerae O1 El Tor [24]. Although classical V.
cholerae O1 caused the fifth and sixth pandemics (and presumably the earlier

10
pandemics), the seventh pandemic is due to the El Tor biotype, which has now
replaced the classical biotype.
Global Cholera Epidemiology
The global burden of cholera, as determined through a systematic review with
clearly stated assumptions, is high [25]. Every year, there are an estimated 3–5
million cholera cases and 100 000–120 000 deaths due to cholera [1]. The actual
numbers of cholera cases are therefore known to differ from those reported.
Differences can be explained by under-reporting due to fear of negative impact on
travel and trade, limitations in surveillance systems, inconsistencies in case
definitions and lack of laboratory diagnostic capacities may also contribute to under-
as well as over- reporting[2]. Cholera occurs in both endemic and epidemic patterns.
Cholera is endemic in many areas of Asia and Africa. About 1.4 billion people are at
risk for cholera in about 51 endemic countries. An estimated 2.8 million cholera
cases occur annually in such countries (uncertainty range: 1.4–4.3) and an estimated
87 000 cholera cases occur in about 18 non-endemic countries. The incidence is
estimated to be greatest in children less than 5 years of age. Every year about 91 000
people (uncertainty range: 28 000 to 142 000) die of cholera in endemic countries
and 2500 people die of the disease in non-endemic countries [25]. In Asia, cholera
occurs seasonally before and after the monsoon rains[3], and the incidence of
disease peaks in children younger than five years, and may occur in neonates[26,
27]. Cholera epidemics occur in a longer cycle superimposed on existing endemic
disease. This pattern relates to declining levels of population-level immunity from a
previous outbreak, overlaid on cycles of climate variability[28]. Hallmarks of the
epidemiology of cholera include: a high degree of clustering of cases by location
and season; highest rates of infection in children 1 to 5 years of age in areas of
endemic infection; antibiotic resistance patterns that frequently change from year to
year; clonal diversity of epidemic strains; and protection against the disease by
improved sanitation and hygiene and pre-existing immunity. Cholera has been
categorized as one of the “emerging and re-emerging infections” [29] threatening
many developing countries. Re-emergence has been noted in the Americas[30] with
the 2010 outbreak in Haiti being the most explosive in which there have been
731,880 cholera cases, of which 416,757 were hospitalized (57%) and 8741 died
(CFR,1.2%) as of February 2015[31]. Cholera has largely been eliminated in
industrialised countries but imported cases are being recorded. The United States

11
still has about 10 to 15 cases per year, particularly in the southernmost states,
because of undercooked seafood harvested in the Gulf of Mexico[32].

Cholera in Africa

During the current seventh cholera pandemic, Africa bore the major brunt of global
disease burden. More than 40 years after its resurgence in Africa in 1970, cholera
remains a grave public health problem, characterized by large disease burden,
frequent outbreaks, persistent endemicity, and high CFRs, particularly in the region
of the central African Great Lakes which might act as reservoirs for cholera. There,
cases occur year round with a rise in incidence during the rainy season. Elsewhere in
sub-Saharan Africa, cholera occurs mostly in outbreaks of varying size with a
constant threat of widespread epidemics. Between 1970 and
2011, African countries reported 3,221,050 suspected cholera cases to the World
Health Organization, representing 46 % of all cases reported globally. Excluding
the Haitian epidemic, sub-Saharan Africa accounted for 86 % of reported cases
and 99 % of deaths worldwide in 2011. In 2013, 43.6% of cases were reported from
Africa whereas between 2001 and 2009, 93% to 98% of total cases worldwide were
reported from Africa[33]. One source on country-specific incidence rates for Africa,
adjusting for underreporting, estimates 1,411,453 cases and 53,632 deaths per year,
respectively (50 % of 2,836,669 estimated cases and 58.6 % of 91,490 estimated
deaths worldwide)[25]. Within Africa, half of all cases between 1970 and 2011 were
notified from only seven countries: Angola, Democratic Republic of the Congo,
Mozambique, Nigeria, Somalia, Tanzania, and South Africa. In contrast to a global
trend of decreasing case fatality ratios (CFRs), CFRs have remained stable in Africa
at approximately 2 %[34]. Early propagation of cholera outbreaks depends largely
on the extent of individual bacterial shedding, host and organism characteristics, the
likelihood of people coming into contact with an infectious dose of Vibrio cholerae
and on the virulence of the implicated strain. Cholera transmission can then be
amplified by several factors including contamination of human water- or food
sources; climate and extreme weather events; political and economic crises; high
population density combined with poor quality informal housing and poor hygiene
practices; spread beyond a local community through human travel and animals, e.g.,
water birds. At an individual level, cholera risk may increase with decreasing
immunity and hypochlorhydria, such as that induced by Helicobacter pylori
12
infection, which is endemic in much of Africa, and may increase individual
susceptibility and cholera incidence[34]. There is some evidence that HIV infection
may predispose to cholera. In a case-control study from sub-Saharan Africa there
was a trend towards HIV-positive patients having a recent history of suspected
cholera [35]. The mechanism is likely to be due to depletion of secretory
immunoglobulin A, the most important antibody for mucosal immunity, due to its
ability both to neutralise cholera toxin and block binding receptors used by the
bacteria as they attach to the epithelium.

Cholera in complex humanitarian emergencies

Preventable infectious diseases have been consistently the major causes of morbidity
and mortality during most CHE in Africa and Asia[36]. The increased rates of
infectious diseases can be attributed largely to mass population displacements, with
the associated problems of poor water supplies, inadequate sanitation,
overcrowding, malnutrition, disruption of public health programs, and limited access
to basic health care. The four infectious diseases most commonly associated with
morbidity and mortality during CHE are diarrhoeal diseases, acute respiratory tract
infections, measles and malaria. Diarrhoeal diseases have been identified as the
most lethal public health threat to refugees and IDPs [37]. Approximately 90% of
the deaths among Rwandan refugees in and around Goma, Zaire in July–August
1994 were attributed to diarrhoeal illnesses caused by Vibrio cholerae 01 and
Shigella dysenteriae type 1[38]. Major outbreaks of these diseases have occurred
among refugee and IDP populations in Somalia, Sudan, Ethiopia, Tanzania, Angola
and DR Congo [36-38].

Pathophysiology, Diagnosis, Treatment and Prognosis of Cholera

The infectious dose of V. cholerae O1 has been estimated to be 10 5 −108 in human

volunteers, but may be as low as 103 in the presence of achlorhydria (rice-water
stools contain 1011 to 1012 V. cholerae organisms per litre) [39]. In the upper small
intestine, bacteria replicate rapidly and adhere to the epithelial cell surface, having
penetrated the outer mucus layer using pili. The bacteria then produce vast local
quantities of cholera toxin. This 84,000-kDa protein is made up of 5 binding (B)
units that latch the toxin to GM1 ganglioside receptors, allowing intracellular
placement of the single active (A) unit, an adenosine diphosphate (ADP)-
13
ribosylating toxin, which irreversibly stimulates the intracellular enzyme adenylate
cyclase[40]. Adenylate cyclase breaks down ADP to cyclic adenosine
monophosphate (cAMP), a powerful intracellular signalling molecule. Ultimately,
increased chloride secretion and inhibition of sodium chloride resorption leads to
massive volumes of diarrhoea. There is also inhibition of water resorption in the
colon, augmenting the effect. Loss of large volumes of salt and water leads to
systemic volume depletion. Hypokalaemic metabolic acidosis is driven by
potassium and bicarbonate loss, which can affect muscle strength further, leading to
exhaustion as patients increase their respiratory rate to compensate. There is no
breach of mucosal integrity. Thus, protein and blood are not lost in the diarrhoea.
Cholera diarrhoea is watery and replete with infectious bacteria. This quickly leads
to secondary cases among carers, particularly if infection control is difficult to
sustain. Contamination of water supplies and the environment can quickly occur,
leading to further outbreaks.

Key diagnostic factors

 copious watery diarrhoea (common)

 Copious watery, bloodless 'rice-water' diarrhoea is specific for cholera in

the proper clinical setting.
 Diarrhoea >1 litre/hour is almost pathognomonic of cholera if sustained
(>20 mL/kg during a 4-hour observation period)[41]

 evidence of volume depletion (common)

 Patients with severe volume depletion in the context of a diarrhoeal illness are
5.5 times more likely to have cholera than other diarrhoeal illnesses [42].
 Mild (<5% volume depletion): alert but tachycardic, dry mucous membranes,
small postural blood pressure (BP) drop (<20 mmHg), able to drink fluids; in
young children, anterior fontanelle palpable but not sunken.
 Moderate (5% to 10% volume depletion): irritability, sunken eyes, dry mouth,
significant (>20 mmHg) postural drop in BP, mildly decreased skin turgor, thirst
and may drink fluids.[3].
 In severe (>10% volume depletion): patients will be lethargic or comatose
with circulatory collapse (thready pulse, low BP with a systolic BP <80
14
 mmHg), as well as sunken eyes, absent tears, dry mucous membranes, poor
(>2 seconds) capillary return, and poor skin turgor. Muscle cramping and
weakness due to electrolyte losses and ion shifts (particularly potassium and
calcium) are common. In children, depletion of glycogen stores and
inadequate gluconeogenesis can lead to severe hypoglycaemia, manifest by
altered consciousness, seizures or even coma. “Cholera sicca” is an unusual
form of the disease in which fluid accumulates in the intestinal lumen, and
circulatory collapse and even death, can occur before the passage of the first
loose stool. Less than 20% of people get serious V. cholerae diarrhoeal
infection, and only those who are severely dehydrated will develop
circulatory collapse. In the short term (up to 24 hours) this is a reversible
situation with fluids alone, but after 24 to 48 hours the complications of
prolonged hypotension will almost inevitably lead to death.
 In controlled laboratory conditions, healthy volunteers are at risk of more severe
disease if they are group O[43].

Other diagnostic factors

 age <5 years (common)

o In endemic areas, adults often have pre-existing immunity to circulating
strains of Vibrio cholerae. Therefore, new outbreaks often affect the
youngest first. Because diarrhoea due to other agents is common in
young children, the occurrence of severe diarrhoea in this age group is
not specific for cholera.
o During outbreaks, cholera can affect all ages. Even though rates are
higher in children, more adults become ill because, population-wise,
there are more adults at risk. Among returning travellers, adults may be
affected with an incubation period of up to 5 days.
 Breastfeeding has been shown to protect babies from occurrence of cholera.
Cholera is uncommon in infants and children under 2 years of age who are
breastfed.
 Ingestion of shellfish (common): Shellfish are a common source of
contamination with cholera.
 Family history of recent, severe, cholera-like illness (common): Cholera often
occurs in family clusters, due either to secondary cases or to a common source

15
 outbreak. In Bangladesh, 50% of household contacts develop symptoms within
48 hours [44].
 Vomiting (common): Can be an early sign of cholera and is extremely common
among children, but is non-specific.
 Fever , abdominal pain, lethargy or coma are uncommon

Laboratory diagnosis

Where laboratory facilities are available, V. cholerae infection can be confirmed by

isolation of the organism from stool on selective media, followed by biochemical
tests, as well as serogrouping and serotyping with specific antibodies.

Table 1. Diagnostic tests1

Test
Full Blood Count
Serum electrolytes
Serum urea and creatinine
Serum lactate
Arterial blood gases
electrocardiogram
Dark field/phase-contrast
microscopy
Rapid dipstick testing of
stool
Gram stain of stool
Stool culture of liquid stool,
faecal suspension or rectal
swab
Result
May have elevated haematocrit or neutrophil count
possible potassium derangement or acidosis
Possible elevated urea and creatinine
elevated
acidaemia
Sinus tachycardia, bradycardia, prolonged PR, flat T waves
Large quantity of curved bacilli
Positive
Small, curved gram-negative rods
yellow, shiny colonies, 2 to 4 mm diameter; selective
growth of Vibrio cholerae on thiosulphate citrate bile salts
sucrose (TCBS) agar

Serogroup confirmation positive for O1 or O139

using antisera
Antibiogram (sensitivities) antibiotic resistance or sensitivity indicated
ELISA assay of stool Positive
Polymerase Chain Reaction Positive
(PCR) assays of stool
Loop-mediated isothermal Positive
amplification (LAMP) assay
of stool

Differential Diagnosis

1
Adapted from BMJ Best Practice
16
Other infectious diarrhoea such as E. coli, Campylobacter, viral diarrhoea; food
poisoning, VIPoma, Tubulovillous adenoma.

World Health Organization (WHO) standard clinical case definition

A case of cholera should be suspected when:

 In an area where the disease is not known to be present, a patient aged 5

years or older develops severe volume depletion or dies from acute watery
diarrhoea
 In an area where there is a cholera epidemic, a patient aged 5 years or older
develops acute watery diarrhoea, with or without vomiting.

Standard laboratory techniques are then used to confirm Vibrio cholerae in stool
samples from suspected cases and allow antibiotic sensitivities to be ascertained.
Once a cholera outbreak has been confirmed, the WHO clinical case definition is
sufficient to diagnose subsequent new cases.

Treatment of cholera

Efficient treatment resides in prompt rehydration through the administration of oral

rehydration salts (ORS) or intravenous fluids, depending of the severity of cases. Up
to 80% of patients can be treated adequately through the administration of ORS.
Very severely dehydrated patients are treated through the administration of
intravenous fluids, preferably Ringer lactate. Patients with severe cholera typically
require an average of 200 mL/kg of isotonic oral or intravenous fluids in the first 24
hours of therapy, and may require more than 350 mL/kg [1, 45]. Appropriate
antibiotics can be given to severe cases to diminish the duration of diarrhoea, reduce
the volume of rehydration fluids needed and shorten the duration of V. cholerae
excretion. For children up to five years, supplementary administration of zinc has a
proven effective in reducing duration of diarrhoea as well as reduction in successive
diarrhoea episodes. In order to ensure timely access to treatment, cholera treatment
centres should be set up among the affected populations whenever feasible.
Employing the current standard of care, the mortality of severe cholera can be
reduced to less than 0.2%, even in resource-limited settings[3]. In an outbreak,
clinicians and public health officials often need to manage many patients at

17
the same time. Critical response features include establishing cholera treatment
centres; training staff in case recognition and management; providing safe water and
sanitation and health promotion.

Prevention of cholera

Measures for the prevention of cholera mostly consist of providing clean water and
proper sanitation to populations who do not yet have access to basic services. Health
education and good food hygiene are equally important. Communities should be
reminded of basic hygienic behaviours, including the necessity of systematic hand-
washing with soap after defecation and before handling food or eating, as well as
safe preparation and conservation of food. Appropriate media, such as radio,
television or newspapers should be involved in disseminating health education
messages. Community and religious leaders should also be associated to social
mobilization campaigns.

In addition, strengthening surveillance and early warning greatly helps in detecting

the first cases and put in place control measures. Conversely, routine treatment of a
community with antibiotics, or mass chemoprophylaxis, has no effect on the spread
of cholera, can have adverse effects by increasing antimicrobial resistance and
provides a false sense of security.

Cholera Vaccines2
Vaccines provide immediate, short‐term protection that can be implemented while
the interventions to improve access to safe water and sanitation are also put into
place. There are two WHO‐prequalified Oral Cholera Vaccines (OCV), Dukoral®
and Shanchol®. Both are whole‐cell, killed vaccines of V. cholerae O1.
Shanchol® also includes the O139 serogroup and, unlike Dukoral®, does not
contain the recombinant cholera toxin B subunit which makes it less expensive to
produce and easier to administer. The table below summarises their major
characteristics.
OCVs have been used in mass‐vaccination campaigns since 1997 as an additional
tool for cholera control to supplement existing priority cholera control measures.
Mass vaccination with OCVs acts directly, by protecting people receiving the
vaccine, and indirectly, by reducing the risk of further contamination of the
environment by those infected with V. cholerae. There is also evidence that OCVs
18
confer significant herd protection [5].
Most mass vaccination campaigns with OCV have so far been pre‐emptive and have
taken place before any potential upsurge in cholera transmission or detection of
outbreaks in targeted populations and areas at risk, including during humanitarian
crises. On some occasions, mass vaccination campaigns with OCV have been
organized on a reactive basis, as part of the response to a cholera outbreak which
had already commenced, to reduce mortality and limit the spread of the disease.
Overall, more than 1.6 million doses of WHO pre‐qualified OCV have been
deployed in mass vaccination campaigns since 1997. Pre‐emptive campaigns have
been conducted in endemic settings in Beira, Mozambique (2003/04), and Zanzibar
(2009). Pre‐emptive use in refugee camps and post‐ crisis situations have been
reported from Uganda (1997), Darfur, Sudan (2004), Aceh, Indonesia (2005), Haiti
(2012), South Sudan (2013) and Thailand (2013). The global OCV stockpile was
used for the first time in 2014 in South Sudan as another pre-emptive measure.
Reactive mass vaccination campaigns were also organized as part of the response to
a cholera outbreak in the Federated States of Micronesia (2000), Mayotte, France
(2000), Marshall Islands (2000), Guinea (2012) and Solomon Islands (2012).
The biggest challenges for implementing an OCV mass vaccination campaign
include:
 The availability of doses of vaccines to cover the target population (including
registration of the vaccine in the country).
 Time required for the vaccines to reach beneficiaries from the producer
 Access to the target population (infrastructure, climate, security)
 Population stability to ensure administration and completion of the 2 doses
 Logistics (transport, cold chain, etc.) of a sufficient scale to support the volume of
vaccines
 The availability of human and financial resources to conduct the campaign

19
Table 2. Characteristics of WHO pre‐qualified oral cholera vaccines2

2
http://www.who.int/cholera/vaccines/OCV_in_humanitarian_emergencies_15Jan2014.pdf?ua=1
20
 AIMS AND OBJECTIVES
Research Questions/Hypotheses

 Did OCV have an effect on morbidity and/or mortality in infected

individuals and to what extent?
o Vaccination is a protective factor and is expected to reduce morbidity
and/or mortality amongst the vaccinated compared to the
unvaccinated population.
 Do international humanitarian organisations perform better than local health
facilities in the management of cholera?
o Even though international agencies are expected to better handle
cholera cases given the expertise and resources they have, cholera
treatment is relatively simple so that even local health facilities (with
limited resources) can perform equally well.

Aims and Objectives

 The ultimate objective is to reduce morbidity and mortality in cholera
outbreaks in unstable settings.
 Specific objectives are:
o To describe the epidemiology of the cholera outbreak in South Sudan
o To assess the effect of oral cholera vaccination on disease severity
o To compare local health facilities with international organisations on
the outcome of case management

Expected results/implications
 Vaccination has been proven to be effective after pre-qualification and licensure
and is thus expected to reduce morbidity as well as mortality. Further rounds of
OCV campaigns may be recommended to at-risk groups (identified in this study)
or the general population of South Sudan as the crisis is protracting.
 NGOs are expected to deliver services that yield results to considerably mitigate
the health impact of cholera and other disasters. Their expertise and resources
like that of MSF should be deployed to strengthen local health facilities in South
Sudan that are currently collapsing.

21
 METHODOLOGY
Setting
The Republic of South Sudan became the world’s newest nation and Africa’s 55th
country on July 9, 2011, following a peaceful secession from the Sudan through a
referendum in January 2011. The country began its life with significant natural
resources and a binding narrative of triumph and freedom, but with a legacy of over
50 years of conflict and an extremely low level of physical, human and institutional
development. The neglect and destruction of war has left the country with virtually
no infrastructure, agriculture at bare subsistence level, dismal human development
indicators, distortions associated with its high dependence on oil revenues and weak
formal institutions of government. The country’s GDP per capita in 2013 was $1081
and the incidence of poverty is more than 57.2% in 2015. Life expectancy at birth
stood at 55 years in 2013. It has an estimated population of 10.9 million (48.1% are
females and 15.8% are U5)[46] and an area covering 644,329 sq. km. South Sudan
is comprised of 10 states: Central Equatoria, Eastern Equatoria, Jonglei, Lakes,
Northern Bahr El Ghazal, Unity, Upper Nile, Warrap, Western Bahr El Ghazal and
Western Equatoria. South Sudan borders the Central African Republic in the west,
Ethiopia in the east, Sudan in the north, and Uganda, Kenya and the Democratic
Republic of Congo to its south. The country is very young with two-thirds of the
population under the age of 30. Almost 83% of South Sudanese resided in rural
areas before the outbreak of the recent conflict, which has displaced nearly 2 million
people. Only 27% of the population aged 15 years and above is literate, with
significant gender disparities: the literacy rate for males is 40% compared to 16%
for females. The infant mortality rate is 105 (per 1,000 live births), maternal
mortality rate is 2,054 (per 100,000 live births), and only 17% of children are fully
immunized. Fifty-five percent of the population has access to improved sources of
drinking water. Around 38% of the population has to walk for more than 30 minutes
one way to collect drinking water. Some 80% of South Sudanese do not have access
to any toilet facility. Large parts of the country can remain isolated for up to six
months of the year due to the rainy season and poor road conditions which render
access close to impossible.
The 2005 Comprehensive Peace Agreement (CPA) was followed by a fragile
interim process towards independence of South Sudan, culminating in the January
2011 referendum. However, neither the peace agreement nor formal succession had
put an end to instability and violence. Yet, on 15 December 2013, conflict erupted in
22
Juba, the capital of South Sudan. By March 2014, the United Nations Office for the
Coordination of Humanitarian Affairs (UNOCHA) estimated that 1 million people
had been displaced, including 800 000 internally displaced persons (IDPs) and 200
000 as refugees in neighbouring countries. The IDP camps were overcrowded,
shelter was makeshift, with limited access to food, water, sanitation and healthcare.
Cumulatively these components magnified risk factors for a cholera outbreak as
cholera is endemic in country. In January 2014, the Ministry of Health together with
WHO and other partners decided to implement a pre-emptive mass vaccination
campaign using OCV (Shanchol®) from the global stockpile to complement other
cholera prevention and control measures. For fully protective immunity, 2 doses of
OCV were required, given 2 weeks apart. The target population was all persons >1
year of age, except pregnant women. Coverage ranged from <60% to 85% according
to administrative data and post-vaccination surveys respectively reaching 166 000
refugees and 40 000 host population at a cost of about 104000USD [9, 10]. Key
challenges encountered included: continued hostility impacted the humanitarian
operations and access to the displaced people; inaccurate population registration and
estimates; increased logistical requirements for cold chain and storage; increased
costs due to requirements for human resources and need for internal air transport,
given the security context; lack of space in the camps to set up fixed vaccination
sites due to overcrowding and structure of the camps; curfew and access constraints
in the UN compounds restricted movement for outside personnel and led to
operational delays and short working days.
Despite these efforts, on the 15 th May 2014, South Sudan's Ministry of Health
declared a cholera outbreak in Juba, Central Equatoria State[11] As of 12 October
2014, a total of 6,141 cholera cases including 139 deaths (2.26%)3. It is unknown
what impact did OCV have in the prevention and control of the cholera outbreak in
South Sudan. Our study focuses on investigating the effect of OCV on severe
disease and lethality.
Study design: retrospective cross sectional study collecting data both exposure and
outcome measures of interest at only one period in time.
Participants: cholera patients reported by health facilities in affected areas of South
Sudan.
Data Collection: We made use of the 2014 South Sudan cholera cases dataset (a
line listing which is the property of Ministry of Health of The Republic of South
3
http://www.who.int/hac/crises/ssd/sitreps/south_sudan_cholera_situation_12october2014.pdf?ua=1
23
Sudan and WHO). A registry was developed to compile data collected from various
treatment facilities across the country between April and June 2014. Data collected
from health-facility cases included: socio-demographics like age, gender and place
of residence, time of onset and time at arrival to health facility, OCV status, clinical
signs and symptoms, laboratory tests, type of treatment, duration in hospital and
outcome. Complementary information related to the same outbreak and the setting
was obtained online from the websites pertaining to the various stakeholders that
were active on the ground during the outbreak.
Ethical consideration: Permission to use the dataset and IRB clearance from MOH
South Sudan and WHO is pending. Though individual consent was not necessary
because we were using secondary data, identity of patients however would not be
disclosed. Use of complementary material/input will be acknowledged accordingly.
Data analysis:
Data analyses were performed using Stata® 13.1 (StataCorp LP, Texas 77845,
USA). Microsoft Excel® 2013 was used to plot the epidemic curve. The data set
was checked for logical inconsistencies, illegal codes, omissions and improbabilities
by tabulating, summarising, describing and plotting variables. Missing observations
were systematically excluded. Cases reported from the community were excluded
from analyses, because a complete count of community cases and deaths was not
available for comparison with hospital cases.
Outcome variables: Case fatality rate (mortality) defined as the proportion of deaths
among all the cholera cases at a given time and place; weekly incidence rate defined
as the number of new cholera cases occurring during a week at a given place;
severity of cholera infection (morbidity) was defined as a function of blood volume
depletion or degree of dehydration. Mild and moderate cases of dehydration on
entry into the health facility were considered as simple cholera cases while cases
presenting with severe dehydration were classified as having severe cholera or
cholera gravis.
Explanatory variables: Taken into account were variables related to socio-
demographics (age, sex, residence, reporting health facility, date of onset, date of
arrival at health facility); clinical assessment (nature of diarrhoea, vomiting, degree
of dehydration, other associated signs and symptoms); treatment-related variables
(oral rehydration, fluids, antibiotics); doses of OCV received; laboratory variables
(rapid diagnostic test, rectal swab/culture results); treatment outcome (discharged

24
alive, dead, still admitted, escaped, readmitted); and length of hospital stay. For
some variables it was necessary to categorise further, as displayed in Table 3. Age
was considered as an a priori confounder for morbidity and mortality while
treatment was thought to interact (effect modification) with clinical status to
influence mortality because treatment administered depended on the state of
dehydration.
Summary statistics were presented as proportions for categorical variables and as
means (standard deviations) or medians (IQR-Interquartile Range) for continuous
variables. Pearson chi-squared test or Fisher exact tests were used where appropriate
to assess for the association between categorical variables. The student t-test was
used to test for the mean difference between two subgroups for continuous
variables.
A univariable logistic regression model was set up for each association between
exposure variables with severe cholera and then with death. Crude odd ratios, their
95% confidence intervals and p-values were obtained.
Factors associated with severe cholera or death at 95% significance level, were
included in their respective and appropriate multiple logistic regression model to
pick up variables independently associated with severe cholera and/or death. Where
the model could not converge due to collinearity, the variable(s) responsible was
(were) taken out of the model. The final model for screening for determinants of
disease severity excluded treatment variables arising while the patient is already in
care, since they were thought to be dependent on the clinical picture. The likelihood
ratio test (LRT) was used to decide between which of the model with or without
interaction terms (effect modification) was appropriate. The model without
interaction terms was better and was therefore retained as our final model. The
corresponding adjusted odd ratios, their 95% confidence intervals and p-values in
the final models were reported.

25
 RESULTS
Outbreak description
A total of 4144 cholera cases arising between the 23 rd April 2014 and the 20th July
2014 were reported of whom 29 were reported from the communities. Our results
are based on cases treated in health facilities and on the complete number of
observations available for a given variable within this period of the outbreak.
Person
Of those affected, 1907 (46.4%) were women and 900 (22.0%) were children below
five years old (Table 3). Their median age was 20 (IQR: 5-32) years. The first case
was a 28-year-old internally displaced man who was partially protected by a single
dose of oral cholera vaccine. Seven more cases had received one dose of OCV while
75 (1.8%) did receive the two recommended doses of OCV.
Place
The index case was reported from the MSF clinic in Juba 3 UN House IDP camp.
Almost 90% of the cases were observed from two out of nine reporting counties:
2007 cases (50.5%) from Juba County and 1614 (39.2%) from Torit County (Figure
1). Attacks rates could not be computed due to incomplete ascertainment of
numerators and denominators. Local health facilities treated a little above three-
quarters of the cases 3142 (76.4%) while the others were treated in structures run by
MSF.

26
Figure 1. Map of South Sudan showing cholera affected counties between April and July 2014

Time
The outbreak occurred in the 17th epidemiological week in 2014 and it is described
up to the 30th epidemiological week when it was almost over. The epidemic curve
depicts a propagated epidemic with two prominent waves; the first wave occurring
before the 25th epi-week and the second thereafter (Figure 2). The second wave
witnessed most of the cases 2380 (57.8%). Early presentation to a medical
professional was common as the mean time from onset of disease to presentation at
the health facility was 0.4 days with just 206 (5%) seeking care more than one day
after of onset.

Alive
Died
140

120

100
Number of cases

80

60

40

20

0
21/04/2014
23/04/2014
25/04/2014
27/04/2014
29/04/2014
01/05/2014
03/05/2014
05/05/2014
07/05/2014
09/05/2014
11/05/2014
13/05/2014
15/05/2014
17/05/2014
19/05/2014
21/05/2014
23/05/2014
25/05/2014
27/05/2014
29/05/2014
31/05/2014
02/06/2014
04/06/2014
06/06/2014
08/06/2014
10/06/2014
12/06/2014
14/06/2014
16/06/2014
18/06/2014
20/06/2014
22/06/2014
24/06/2014
26/06/2014
28/06/2014
30/06/2014
07/02/2014
07/04/2014
07/06/2014
07/08/2014
07/10/2014
07/12/2014

18/7/20014
14/7/2014
16/7/2014

20/7/2014

17 18 19 20 21 22 23 24 25 26 27 28 29
Date of onset

Figure 2. The epidemic curve

27
Clinical presentation and diagnosis
All cases presented with diarrhoea as cholera is a diarrhoeal disease but the nature of
the stools was different. Overall, there was an equal proportion between cases with
clear watery (49.2%) and those with “rice-water” stools (50.8%).Diarrhoea was
accompanied by vomiting in 3625 (91.4%) of cases. Other symptoms recorded
included: body weakness in 1838 (48.2%), abdominal cramps in 1530 (40.1%)
cases, leg and arm cramps in 1312 (34.4%) cases, headache in 80 (2.1%) patients
and a combination of these in 1349 (35.4%) persons. About a quarter of patients
1017 (25.2%) presented with mild dehydration, 1067 (26.5%) were assessed for
moderate dehydration while severe dehydration was observed in 1946 (48.3%).
Results of cholera rapid diagnostic tests were available for 258 cases of which 101
(39.1%) were positive. Similarly, vibrio culture results were available for 31 cases
of which 22 (71.0%) were positive. Only a few positive tests are necessary to
confirm an outbreak.

Case management and outcome

A majority of cases 3978 (96.7%) were admitted for treatment. Patients received
various treatments including oral rehydration therapy in 1943 (52.4%) cases,
intravenous fluids in 2845 (76.8%), antibiotics in 2070 (55.9%) or a combination of
these in 2513 (67.8%) of them. After a median hospital stay of one (IQR: 0-2) day,
2931 (87.9%) were discharged alive, 297 (8.9%) were still undergoing treatment, 15
(0.4%) were re-admitted, 30 (0.9%) were reported to have left care facility against
medical advice and 62 were recorded dead. Thus the case fatality ratio was 1.9%.

28
Table 3. Baseline characteristics of cholera cases

Characteristic Cases (%) Characteristic Cases (%)

Gender Degree of dehydration
Female 1907(46.4) Mild 1017 (25.2)
Male 2207(53.6) Moderate 1067 (26.5)
Total 4110 (100) Severe 1946 (48.3)
Total 4030 (100)
Age (years) Rapid diagnostic test
<5 900 (22.0) Not done 3835 (93.7)
5-49 2879 (70.4) Positive 101 (2.5)
≥50 312 (7.6) Negative 157 (3.8)
Total 4091 (100) Total 4119 (100)
Oral cholera vaccine (doses) Vibrio Culture results
0 4019 (98.0) Not done or reported 3832 (99.2)
1 8 (0.2) Positive 22 (0.6)
2 75 (1.8) Negative 9 (0.2)
Total 4102 (100) Total 3863 (100)
Epi-week of onset Hospitalisations
17-24 (First wave) 1735 (42.2) No 135 (3.3)
25-30 (second wave) 2380 (57.8) Yes 3978 (96.7)
Total 4115 (100) Total 4113 (100)
Reporting health facility Hospital stay (days)
Local centres 3142 (76.4) <3 1599 (77.1)
MSF centres 972 (23.6) ≥3 476 (22.9)
Total 4114 (100) Total 2075 (100)
County Treatment
Ikotos 3 (0.1) Oral rehydration salt 1943 (52.4)
Juba 2077 (50.5) Intravenous fluids 2845 (76.8)
Kajo Keji 60 (1.5) Antibiotics 2070 (55.9)
Kopoeta North 51 (1.2) Multiple treatment 2513 (67.8)
Lopa-Lafon 94 (2.3)
Magwi 164 (3.9)
Mundri East 3 (0.1)
Torit 1614 (39.2)
Yei 48 (1.2)
Total 4114 (100)
Delay at presentation (days) Outcome
1 3889 (95.0) Discharged alive 2931 (87.9)
2+ 206 (5.0) Dead 62 (1.9)
Total 4095 (100) Still hospitalised 297 (8.9)
Nature of diarrhoea Escaped 30 (0.9)
Clear water 1970 (49.2) Readmitted 15 (0.4)
Rice water 2037 (50.8) Total 3335 (100)
Total 4007 (100)
Vomiting
No 341 (8.6)
Yes 3625 (91.4)
Total 3966 (100)
Other symptoms
Absent 2344 (61.5)
Present 1469 (40.4)
Total 3833 (100)
29
Effect of oral cholera vaccine and other factors on severity of cholera
The total number of severe cholera cases (severe dehydration) was 1946 (48.3%).
In univariable analysis, compared to unvaccinated cases, there was no evidence
that there was a 47% reduction (OR=0.63, 95%CI: 0.15-2.63, p=0.5) in the odds of
presenting with severe cholera among the partially vaccinated cases but there was a
very strong evidence that there was a 80% reduction (OR=0.20, 95%CI: 0.11-0.38,
p<0.001) in the odds of presenting with severe cholera among the completely
vaccinated cases (Table 4). There was no evidence of non-linearity therefore
suggesting a dose response effect between OCV doses and disease severity [χ 2 (1) =
0.19, p=0.66]. Other factors associated with a reduction in the likelihood of
presenting with sever cholera included: the female gender (OR=0.84, p=0.007), the
under-5-year-olds (OR=0.58, p<0.001), cases affected during the second wave
(OR=0.18, p<0.001), cases arising outside the county of Juba, cases reported by or
seeking care at MSF centres (OR=0.16, p<0.001) and cases not presenting with
multiple symptoms (OR=0.06, p<0.001). Factors associated with an increased risk
of presenting with severe disease included: late presentation after one day of onset
(OR=2.12, p<0.001), cases with “rice-water” diarrhoea (OR=10.24, p<0.001) and
with vomiting (OR=10.12, p<0.001).
After controlling for confounding in a multivariable analysis, relative to the
unvaccinated cases, there was still no evidence of an apparent 43% reduction of risk
of severe cholera in partial vaccinees but there was still a strong evidence that cases
who received two doses of OCV did experience a 71% reduction (aOR=0.29,
95%CI: 0.14-0.62, p=0.001) in the odds of developing a severe disease (Table 4).
There was no evidence that MSF centres were unlikely to report severe cases
(aOR=0.83, 95%CI: 0.61-1.12, p=0.230). Centres outside Counties of Juba, Kajo
Keji and Yei were likely to report or to receive severe cases. The effect of age and
sex could no longer be retained. Other factors that independently were linked to a
reduction in disease severity included: presentation in the second half of the
epidemic (aOR=0.28, 95%CI: 0.21-0.37, p<0.001) and the absence of multiple
symptoms (aOR=0.15,95%CI: 0.12-0.19, p<0.001).Other factors that consistently
were associated with an increased risk of severe cholera were: late presentation

30
 (aOR=2.65, 95%CI: 1.70-4.13, p<0.001), rice-water stool (aOR=4.25, 95%CI:3.32-
5.43, p<0.001) and vomiting (aOR=7.32, 95%CI: 4.79-11.18, p<0.001).

Table 4. Logistic regression model of factors associated with severe diarrhoea

Characteristic Severe Cases (%) OR (95%CI)* P-value aOR (95%CI)* P-value
Oral cholera vaccine
(doses)
0 1926 (48.9) 1 1
1 3 (37.5) 0.63 (0.15 – 2.63) 0.524 0.57 (0.08 – 4.03) 0.572
2 12 (16.2) 0.20 (0.11 – 0.38) <0.001 0.29 (0.14 – 0.62) 0.001

Gender
Female 859 (46.0) 0.84 (0.74 – 0.95) 0.007 0.86 (0.71 - 1.03) 0.099
Male 1084 (50.3) 1 1

Age (years)
<5 332 (37.8) 0.58 (0.50 – 0.68) <0.001 1.12 (0.89 - 1.40) 0.342
5-49 1443 (51.1) 1 1
≥50 156 (51.7) 1.02 (0.81 – 1.30) 0.849 1.27 (0.89 - 1.82) 0.183

Epi-week at
presentation
17-24 (First wave) 1239 (71.5) 1 1
25-30 (second wave) 707 (30.8) 0.18 (0.15 – 0.20) <0.001 0.28 (0.21 - 0.37) <0.001

Reporting health
facility
Local centres 1773 (57.8) 1 1
MSF centres 173 (18.0) 0.16 (0.13 – 0.19) <0.001 0.83 (0.61 – 1.12) 0.230

Delay at
presentation (days)
1 1801 (47.3) 1 1
≥2 133 (65.5) 2.12 (1.57 – 2.85) <0.001 2.65 (1.70 – 4.13) <0.001
Nature of diarrhoea
Clear water 418 (22.0) 1 1
Rice water 1508 (74.3) 10.24 (8.84 -11.87) <0.001 4.25 (3.32 – 5.43) <0.001

Vomiting
No 34 (10.0) 1 1
Yes 1898 (52.9) 10.12 (7.06-14.51) <0.001 7.32 (4.79 – 11.18) <0.001

Other symptoms
Present 1697 (72.9) 1 1
Absent 186 (12.9) 0.06 (0.05 – 0.07) <0.001 0.15 (0.12 – 0.19) <0.001

* OR: crude odd ratio, aOR: adjusted odd ratio, CI: confidence interval

31
 Effect of oral cholera vaccine and other factors on cholera case fatality
In a univariable logistic regression model, there was no evidence of an apparent
28% reduction (OR=0.72, 95%CI: 0.10-5.24, p=0.743) in mortality associated with
receiving two doses of OCV compared to zero dose of OCV (Table 5). Compared
with local centres, MSF centres did perform better overall and could reduce the CFR
by 77% (OR= 0.23, 95%CI: 0.09-0.58, p=0.002). Elderly persons of age 50 years
and above were at a risk of mortality three and half times greater than the 5-49 years
age group (OR=3.55, 95%CI: 1.81-6.95, p<0.001). Patients affected by the second
wave had a 1.84 times the odds of mortality than those affected by the first wave of
the epidemic. Severe cholera cases witnessed a more than three and half times the
risk of death relative to simple cholera cases (OR=3.62, 95%CI: 1.86-7.04,
p<0.001). Cases admitted to treatment centres had an 89% reduction in the risk of
death compared to persons seen in the outpatient setting.

Table 5. Logistic regression model of factors associated with cholera death

Characteristic Deaths (CFR) OR (95%CI)* P-value aOR (95%CI)* P-value
Oral cholera vaccine
(doses)
0 60 (1.9) 1 1
1 0 (0.0) NA* NA*
2 1 (1.3) 0.72 (0.10 – 5.24) 0.743 4.50 (0.40 – 50.42) 0.222
Age (years)
<5 14 (2.0) 1.44 (0.77 – 2.71) 0.604 1.91 (0.90 – 4.05) 0.092
5-49 34 (1.4) 1 1
50+ 12 (4.9) 3.55 (1.81 – 6.95) <0.001 3.45 (1.56 – 7.64) 0.002
Epi-week at
presentation
17-24 (First wave) 23 (1.3) 1 1
25-30 (second wave) 39 (2.4) 1.84 (1.10 – 3.10) 0.021 0.64 (0.17 – 2.37) 0.501
Reporting health facility
Local centres 57 (2.4) 1 1
MSF centres 5 (0.6) 0.23 (0.09 – 0.58) 0.002 1.04 (0.29 – 3.74) 0.948

Clinical status
Simple cholera 11 (0.7) 1 1
Severe cholera 43 (2.5) 3.62 (1.86 – 7.04) <0.001 5.64 (2.55 – 12.47) <0.001

Hospitalisations
No 11 (12.6) 1 1
Yes 51 (1.6) 0.11 (0.06 – 0.22) <0.001 0.05 (0.02 – 0.14) <0.001

* OR: crude odd ratio, aOR: adjusted odd ratio, CI: confidence interval,
32
 NA: not available

In a multiple logistic model, persons who received two doses of appeared to be

associated with an increased risk of mortality but there was no evidence to support
this harmful effect (aOR=4.50, 95%CI: 0.40-50.42, p=0.222). The greatest risk
factor for death was severe disease which was associated with more than five and
half folds, the odds of death compared with mild and moderate disease (aOR=5.64,
95%CI: 2.55-12.47, p<0.001). The elderly also suffered significant high mortality
compared to the young adults (aOR=3.45, 95%CI: 1.56 – 7.64, p=0.002). The
relative effects of the second wave of the epidemic and that of type of reporting
health facility were confounded by the effect of place of residence and degree of
dehydration. The protective effect of hospitalisation was highly substantial in
reducing deaths by about 95% (aOR=0.05, 95%CI: 0.02-0.14, p<0.001).

33
 DISCUSSION
Our study sought to investigate the effect of pre-emptive oral cholera vaccination on
the severity and mortality of cholera and to examine the performance of local health
facilities in cholera case management with respect to international humanitarian
organisations during the cholera outbreak in South Sudan. There was strong
evidence that vaccination with two doses of OCV was associated with a 71%
reduction in the odds of suffering severe cholera but there was no evidence of a
direct effect of OCV on case fatality rates. Meanwhile, there was no significant
difference in the outcome of cases managed by local health facilities compared to
MSF centres.

Description of the outbreak

We reported 4144 cases between April and July 2014 with a case fatality rate of
1.9%. However, the epidemic did continue for a couple of months and when WHO
stopped reporting, as of 14 December 2014, the cumulative total stood at 6,421
cases including 167 deaths (CFR 2.6%) from five states and 16 counties (2.26%)4.
The difference is obviously explained by the fact that in our study, we excluded
community cases and deaths and the reporting period was shorter, thus resulting in
an underestimation of the overall picture of the outbreak.
The proportion of women affected (46.4%) was lower than that of men but this was
partly a reflection of South Sudan demography in which women represent about
48.1%[46] of the general population. In the literature, the pattern is that while boys
tend to have slightly higher incidence rates among infants and young children, this
trend is entirely reversed among older children and adults where the rate is higher
for females. It is known that with the exception of diarrheal disease in pregnant
women, which can be more severe due to the major immunological changes during
pregnancy, any differences between males and females in cholera incidence and
disease progression appear to be related to gender norms rather than to biology[47].
The counties of the three Equatoria states in general and of those of Central
Equatoria State with Juba as the headquarters in particular were the hardest hit. The
conflict that led to the humanitarian emergency prior to the outbreak erupted and
concentrated in Juba which is also the most densely populated city. Local health
facilities reported more cases than MSF centres but this partly reflects the number
4
http://reliefweb.int/disaster/ep-2014-000064-ssd
34
and bed capacity of the former, and in addition the affected population is unlikely to
radically change their health-seeking behaviour and will prefer to go to local health
facilities they have been familiar with.
The outbreak occurred between April and October corresponding to the beginning
and the end of the season with high rainfall, an essential condition necessary for
transmission of the water-borne disease. The propagated pattern of the epicurve
suggested a person-person transmission of the disease which is also a characteristic
of cholera. More cases were recorded during the second wave of the epidemic,
possibly due to more rains and thus increased transmission but probably due to
improved reporting owing to increased awareness and case finding usually observed
when an outbreak has fully established.
This outbreak was characterised by early detection of the first case and by early
presentation (mean time of 0.4 days) to health facility for treatment within the first
day of onset of symptoms. It is possible that the disease surveillance (early warning)
system was performing at its best at that time when the dire humanitarian conditions
in IDP camps predicted an imminent outbreak, thus compelling the health
authorities to implement three rounds of oral cholera vaccination campaigns that
were associated with community mobilisation to improve awareness about the
disease[10]. It is however unknown yet to what extent early detection, early
presentation and OCV contributed to the containment of the spread of the epidemic
in South Sudan.

Effect of oral cholera vaccine and other factors on severity of cholera

Disease may occur in previously vaccinated individuals. Such breakthroughs are
either primary – due to vaccine failure – or secondary. In such cases, the disease is
usually milder than in the non-vaccinated [48]. In this study, OCV appeared to have
a dose response effect in mitigating disease severity analogous to findings from a
recent study of the protective effectiveness of OCV on preventing clinically-
significant disease in India [49]. This study however failed to find significant
differences in disease severity between the vaccinated and unvaccinated.
On one hand, the apparent protective (43% reduction) effect of a single dose of
OCV against severe disease among the partially vaccinated could not be supported
by evidence from the data because of very small number of cases in this group of

35
vaccinees. We were therefore under-powered to detect the effect of a single dose,
but the dose response effect as shown by the statistical test for linearity hints that a
single dose may provide some protection. The protective efficacy of 71% of the
vaccine in reducing disease severity among the completely vaccinated was strongly
supported by the data. Similarly, initial outbreak investigation reported that ten
contacts of the index case were fully vaccinated and none become a secondary case 5.
These findings are in line with current recommendation of two doses of the vaccine
to achieve its desired efficacy in disease prevention. Therefore, two doses of OCV
may also be required to mitigate the severity of cholera, in vaccinees who develop
clinical disease. Communities need to be persuaded to accept and adhere to multiple
vaccination campaigns in order to achieve an optimum coverage of OCV2 given that
OCV1 may provide little or no protection at all. Our study may be the first
observational study to demonstrate the beneficial effect of OCV (Shanchol) on the
severity of cholera, however, the mitigating effects of vaccines on infectious
diseases has been extensively described[48]. Milder disease in vaccinees was also
reported for rotavirus vaccine against rotavirus diarrhoeal disease [50], for pertussis
vaccine [51] for varicella vaccine[52] and BCG vaccine [53].
Though men and children under-5 years old seemed to become seriously ill in
univariable analysis, disease severity was neither gender- nor age-specific after
controlling for confounders.
Though more cases were recorded during the second than the first wave of the
epidemic, the second wave witnessed less severe illness even after correcting for
delays in seeking care at health facilities. Late presentation to treatment centres was
rare (5%) but was obviously significantly associated with severe illness. It is
possible that in addition to control measures and perhaps improved healthcare
seeking behaviours over time, the virulence of the strain responsible for the outbreak
must have reduced over time as well.
The nature of diarrhoeal stool was linked to severity as patients with the
characteristic “rice-water” stool experienced more severe disease than those with
“clear-watery” stool. Cholera is a secretory diarrhoea mediated by the cholera toxin
and its action on enterocytes enzymes. The presence of free enterocytes in the
almost clear liquid stool give the appearance of rice grains during the early acute
phase of the disease6. But again, after correcting for early presentation and other
5
http://reliefweb.int/report/south-sudan/cholera-outbreak-juba-central-equatoria-state-situation-
report-sitrep-no-1-1700
6
Copyright 2004, Medical Ecology.org
36
factors, the strength of the association between rice stool and severe disease though
falling from about ten to four remained significant. It is plausible that cases with
rice-water stool may have a higher bacterial load and multiplication in the
enterocytes that are subsequently shedded into the intestinal lumen. Another reason
for a milder disease in cases with purely clear-watery diarrhoea is that, these cases
were merely suspect cholera cases in a context of an outbreak since not all cases had
a confirmatory laboratory test to rule out other causes of secretory diarrhoea.
Vomiting when associated with diarrhoea will definitely lead to more body fluid
loss and by interfering with oral fluids intake will certainly result to severe
dehydration. In this study, vomiting was prominent and was the factor with the most
significant deleterious effect on disease severity. Other symptoms like body
weakness, leg and arm cramps are merely symptoms of severe cholera owing to
electrolytes losses than factors contributing to severe cholera.
There were significant differences in the number of severe cases reported by
different health centres in different areas of the country. These differences are due to
complexities in the completeness, timeliness and willingness of reporting by
different health facilities and further compounded by distances from the
communities to the health facilities and the relative performance of the health
information system set up at different levels of the health system. Though small
numbers of cases were reported from some counties and thus posing a challenge in
handling them statistically, overall the place of residence was a factor strongly
associated with the disease.
Our data suggest that rapid diagnostic tests had low sensitivity. This finding should
be taken with a lot of precaution because laboratory testing during an outbreak is not
systematic and is based on convenience sampling of a few cases. WHO suggests that
all samples tested positive with the RDT are re-tested using classic laboratory
procedures for confirmation. Not all cases fitting the WHO clinical case definition
need to be tested. Once an outbreak is confirmed, a clinical diagnosis using WHO
standard case definition is sufficient, accompanied by sporadic testing at regular
intervals[54].

Effect of oral cholera vaccine and other factors on cholera case fatality
The CFR of 1.9%, though an underestimation of the overall picture, is still
considered to be high given that the benchmark for an optimal management is <1%.

37
The rate is the highest since 2006 when the country recorded a CFR of 2.9%.
Between 2007 and 2013, yearly outbreaks of cholera or acute watery diarrhoea have
been reported in South Sudan with CFRs ranging from 0.07% to 1.8% 7. The lowest
rates were reported during the years when no outbreaks were not confirmed as
cholera but perhaps as acute watery diarrhoea. The latest outbreak in February 2015
witnessed the highest CFR of 7% and was thought to be related to a point source
exposure involving 43 cases and three deaths[55]. Overall, the CFRs in South Sudan
are similar to those reported in the whole African Region with an average of about
2%[34] and in conflict settings in the same region[36-38].
Our data failed to demonstrate a direct effect of OCV on case fatality. Only one case
among the vaccinees was accounted for in the logistic regression model which
therefore lacked the power to detect a statistically significant effect. Death in
cholera is as a result of severe volume depletion that eventually leads to
cardiovascular collapse and renal failure in the absence of treatment. In this study,
severe volume depletion was the main determinant for mortality. Severe dehydration
is therefore on the causal pathway between vaccine action and death. In order
words, vaccine prevent deaths by preventing clinical disease or by protecting against
severe disease.
To prevent severe volume depletion and thus death, timely and adequate rehydration
is known as the single most important and easy-to-deliver intervention in all
settings[1]. Our data rightly isolated hospitalisation of cases as the strategy that
provides the ideal conditions conducive for an effective case management. This does
not imply that all cases should be hospitalised in order to achieve the best outcome
possible but in a conflict setting like South Sudan where movement is restricted due
to insecurity, in-patient care for all cases could be an option to consider. While
hospitalisation for better case management directly reduces CFR by correcting
volume depletion in cholera patients; OCV indirectly reduces mortality by
preventing severe volume depletion in cholera patients. Therefore OCV reduces
mortality by reducing the burden of (severe) disease or morbidity.
Cholera is distinctive among diarrhoeal diseases in that mortality is high among
patients of all ages [25]. Our findings suggest that the elderly cases may be
particularly vulnerable to death. Given that our data failed to find an association
between age and severity of disease, it is likely therefore that an increased risk of
death among the elderly was due to their management. Challenges in treating the
7
http://www.who.int/hac/crises/ssd/en/as
38
elderly could be related to co-morbidities which get less attention given the urgency
to treat cholera or might get worse as a result of treatment, for example, over-
zealous use of fluids to treat a volume-depleted cholera case may worsen an existing
heart failure[56]. There are further challenges associated with clinical assessment of
the degree of dehydration at presentation and during follow-up of an elderly person,
for example, skin turgor may not be a good indicator to capture a state of
dehydration in elderly person with an inelastic skin tissue with wrinkles owing to
age. Another practical difficulty with the elderly compared to younger adults is
access to a secure intravenous line necessary for adequate volume repletion. We
cannot tell whether these difficulties were observed or there were instances of
deficiencies or neglect with respect to the elderly. Nonetheless, by aggregating data
as under-fives and over-fives may fail to recognise the specific needs or
vulnerabilities of older people. We hope that through proper training and
supervision of health care workers, these practical difficulties associated with the
care of the old ones can be overcome. We equally draw the attention of the
humanitarian action to cases aged 50 years who are also a vulnerable group like the
under-5s and (pregnant) women during a humanitarian crisis.
With reference to the famous Goma cholera outbreak in 1994, characterised by very
high mortality amidst controversies that international relief organisations
particularly MSF failed in their mission to reduce deaths among Rwandan
refugees[57] and claims by other organisations like The International Centre for
Diarrhoeal Disease Research, Bangladesh (ICDDR, B) of a higher case management
efficiency than local treatment facilities[58]; we sought to investigate existence of
such views two decades later in South Sudan. Though centres operated by MSF
seemingly performed better than local centres, but after controlling for place of
residence and disease severity, there was no real difference between locals and
international agencies. Therefore such opinions that are based on crude CFR alone,
are likely to be spurious and misleading. The methods of case management of
cholera in treatment facilities remains the same, irrespective of reason or magnitude
of an epidemic. Deaths from cholera can only occur either from lack of treatment or
inadequate treatment[57]. Today, there has been improvements in the standards of
delivery of relief actions across disaster affected areas so much so that were do not
expect remarkable discrepancies in outcomes at the end of a cholera outbreak. A
cholera case fatality rate of less than 1 % can be achieved even in rural makeshift
treatment centres, as has been shown in other settings[59, 60].
39
Our study had some limitations. The overall picture of the outbreak is truncated and
incomplete because our data does not take into account community cases, outcome
of cases that were still undergoing treatment in care facilities, cases occurring after
data must have been collected and other missing observations due to recall bias. The
consequence to our estimates is an underestimation of CFRs, inability to compute
attack rates and the lack of power to detect the effect of some exposure variables on
CFR. We could not assess the effect of the various types of treatment on CFR
because of variation in adherence to standards across centres.

CONCLUSIONS AND RECOMMENDATIONS

Despite the above limitations, we can however draw the following conclusions and
pave the way forward:
 Two doses of OCV can mitigate the severity of a breakthrough disease
among the vaccinees thus reducing the burden of disease. Communities
should be mobilised to understand the necessity for a second dose of the
same vaccine.
 We could not observe an effect of OCV on cholera case fatality rates but by
reducing severe cholera which precedes death, we would expect that OCV
indirectly reduces cholera mortality. Vaccine promoters and activists should
use the benefit of OCV in reducing disease burden and mortality to roll out
OCV use where and when necessary, in addition to the prevention of illness
in OCV-vaccinated individuals.
 In an ongoing conflict setting (in South Sudan), hospitalisation substantially
reduced mortality from cholera. Health workers should consider monitoring
most cases as in-patients irrespective of clinical status as an option to save
more lives given the prevailing insecurity that may hamper out-patient
treatment.
 The elderly cholera cases face significant mortality therefore, they should be
considered as a vulnerable group during an outbreak of cholera and be given
appropriate and comprehensive attention.

40
 The burden of disease was greatest in the first half of the outbreak so control
measures need to be put in place as soon as possible during subsequent
outbreaks. It is difficult to organise relief operations as soon as a disaster
strikes but pre-positioning of stocks in epidemic-prone areas may be a
preventive approach.
 Patients arriving to treatment centres two or more days after onset of cholera
are at a greater risk of severe cholera. Raising awareness, case finding and
contact tracing, and setting up of cholera treatment facilities closer to
affected populations may reduce the delay in seeking care.
 Cases presenting with vomiting or “rice-water” are likely to suffer severe
illness. Rice water may suggest “true” cholera cases and should be
monitored closely. Patient who are vomiting should be admitted for
rehydration with intravenous fluid since oral treatment may be ineffective.
 Cholera can be managed optimally at any care facility irrespective of level of
referral or whether it is operated by local or international organisations.
However, coordination of operations around all treatment facilities would be
necessary to guarantee minimum standards.

41
 ACKNOWLEDGMENTS
I would like to sincerely thank all those who have helped in any manner to shape
this project and make it successful.

Acknowledgement of academic support

Project development
Jose M Rodriguez-Llanes thought through, defined and proposed suitable research
areas, project supervisor and collaborators.
Contact, input, and support
Dr Patrick OTIM was our resource person in South Sudan helping us to secure
approval from MoH of South Sudan and WHO in order to use the cholera dataset
and providing us with additional information concerning the outbreak.
Esmeralda Gerritse who initially did some ground work on the dataset also helped to
locate the dataset and other documentation related to South Sudan and the outbreak.
Main research work
Prof Debarati Guha-Sapir approved the research project, reshaped the research plan
and scope, suggested further readings and contact persons, oversaw the entire
progress and gave her approval for eventual publication in a peer reviewed journal.
Dr Joris van Loenhout helped in the projection conception, liaison with collaborators,
suggestion of important reading materials, extensive revision of all sections of the thesis,
proof-reading, entire project supervision and approval for submission and defence.
The MoHSS, WHO, MSF and other humanitarian organisations working in South Sudan,
the health workers and allied staff, the affected population and all others who help in the
collection, transfer and management of data and its eventual approval for their use.
Pascaline Wallemacq for her GIS input in mapping affected areas
Tefera Delbiso provided some statistical advice

Acknowledgement of other support

I am particularly thankful to The EMMPHID Program for the financial support to cover all

42
expenses for this course. I would like to thank the staff of UCL, my family, my
friends and class mates for their support, collaboration and encouragement.
Finally, my gratitude to the Almighty God for his graces all along.

BIBLIOGRAPHY
o
1. Cholera Fact sheet N 107 [http://www.who.int/mediacentre/factsheets/fs107/en/]
2. WHO: Cholera, 2013. Releve epidemiologique hebdomadaire / Section d'hygiene
du Secretariat de la Societe des Nations = Weekly epidemiological record / Health
Section of the Secretariat of the League of Nations 2014, 89(31):345-355.
3. Sack DA, Sack RB, Nair GB, Siddique AK: Cholera. Lancet 2004, 363(9404):223-
233.
4. Jutla A, Whitcombe E, Hasan N, Haley B, Akanda A, Huq A, Alam M, Sack RB,
Colwell R: Environmental factors influencing epidemic cholera. The American
journal of tropical medicine and hygiene 2013, 89(3):597-607.
5. Ali M, Sur D, You YA, Kanungo S, Sah B, Manna B, Puri M, Wierzba TF, Donner
A, Nair GB et al: Herd protection by a bivalent killed whole-cell oral cholera
vaccine in the slums of Kolkata, India. Clinical infectious diseases : an official
publication of the Infectious Diseases Society of America 2013, 56(8):1123-1131.
6. Cholera vaccines: WHO position paper-Recommendations. Vaccine 2010,
28(30):4687-4688.
7. Ivers LC, Hilaire IJ, Teng JE, Almazor CP, Jerome JG, Ternier R, Boncy J, Buteau
J, Murray MB, Harris JB et al: Effectiveness of reactive oral cholera vaccination in
rural Haiti: a case-control study and bias-indicator analysis. The Lancet Global
health 2015, 3(3):e162-168.
8. WHO: Meeting report of the WHO Technical Working Group on creation of an oral
cholera vaccine stockpile. In. Geneva; 2012: 37.
9. Porta MI, Lenglet A, de Weerdt S, Crestani R, Sinke R, Frawley MJ, Van Herp M,
Zachariah R: Feasibility of a preventive mass vaccination campaign with two doses
of oral cholera vaccine during a humanitarian emergency in South Sudan.
Transactions of the Royal Society of Tropical Medicine and Hygiene 2014,
108(12):810-815.
10. WorldHealthOrganisation: Oral cholera vaccine campaign among internally
displaced persons in South Sudan. Releve epidemiologique hebdomadaire / Section
d'hygiene du Secretariat de la Societe des Nations = Weekly epidemiological
record / Health Section of the Secretariat of the League of Nations 2014,
89(20):214-220.
11. MoHSouthSudan: Cholera Outbreak in Juba, Republic of South Sudan. In. Edited
by Services PH. Juba: MOHSS; 2014.
12. South Sudan Crisis: Key facts [http://www.msf.ie/node/17101]
13. Bhishagranta: An english translation of the Sushruta Samhita. In., vol. 3. Varanasi:
The Chowkhamba
Sanskrit Series Office; 1963: 352-356.
14. Peters J: Early history of Asiatic cholera as Known to Europeans A.D. 1500–1800.
In: Asiatic Cholera. edn. Edited by EC W. New York: William Wood; 1885.
15. R; P: Cholera. In. Geneva: World Health Organisation; 1959.
16. Bentivoglio M, Pacini P: Filippo Pacini: a determined observer. Brain research
bulletin 1995, 38(2):161-165.
17. Gleick P: The World's Water 2008-2009: The Biennial Report on Fresh Water
Resources. In. USA: Inland Press; 2008.
18. Colwell RR: Global climate and infectious disease: the cholera paradigm. Science
(New York, NY) 1996, 274(5295):2025-2031.
19. Kirn TJ, Jude BA, Taylor RK: A colonization factor links Vibrio cholerae
environmental survival and human infection. Nature 2005, 438(7069):863-866.

43
20. Faruque SM, Islam MJ, Ahmad QS, Faruque AS, Sack DA, Nair GB, Mekalanos JJ:
Self-limiting nature of seasonal cholera epidemics: Role of host-mediated
amplification of phage. Proceedings of the National Academy of Sciences of the
United States of America 2005, 102(17):6119-6124.
21. Morris JG, Jr.: Cholera and other types of vibriosis: a story of human pandemics
and oysters on the half shell. Clinical infectious diseases : an official publication of
the Infectious Diseases Society of America 2003, 37(2):272-280.
22. Longini IM, Jr., Yunus M, Zaman K, Siddique AK, Sack RB, Nizam A: Epidemic
and endemic cholera trends over a 33-year period in Bangladesh. J Infect Dis 2002,
186(2):246-251.
23. Ramamurthy T, Garg S, Sharma R, Bhattacharya SK, Nair GB, Shimada T, Takeda
T, Karasawa T, Kurazano H, Pal A et al: Emergence of novel strain of Vibrio
cholerae with epidemic potential in southern and eastern India. Lancet 1993,
341(8846):703-704.
24. Bik EM, Bunschoten AE, Gouw RD, Mooi FR: Genesis of the novel epidemic
Vibrio cholerae O139 strain: evidence for horizontal transfer of genes involved in
polysaccharide synthesis. The EMBO journal 1995, 14(2):209-216.
25. Ali M, Lopez AL, You YA, Kim YE, Sah B, Maskery B, Clemens J: The global
burden of cholera. Bulletin of the World Health Organization 2012, 90(3):209-
218A.
26. Deen JL, von Seidlein L, Sur D, Agtini M, Lucas ME, Lopez AL, Kim DR, Ali M,
Clemens JD: The high burden of cholera in children: comparison of incidence from
endemic areas in Asia and Africa. PLoS neglected tropical diseases 2008,
2(2):e173.
27. Khan AM, Hossain MS, Khan AI, Chisti MJ, Chowdhury F, Faruque AS, Salam
MA: Bacterial enteropathogens of neonates admitted to an urban diarrhoeal hospital
in Bangladesh. Journal of tropical pediatrics 2009, 55(2):122-124.
28. Koelle K, Rodo X, Pascual M, Yunus M, Mostafa G: Refractory periods and
climate forcing in cholera dynamics. Nature 2005, 436(7051):696-700.
29. Satcher D: Emerging infections: getting ahead of the curve. Emerg Infect Dis 1995,
1(1):1-6.
30. Periago MR: CholerainLatinAmerica: Lessons from history. In: PAHO. 2011.
31. PAHO: Cholera in the America-Situation summary. In: PAHO Epidemiological
Update. Washington; 2015.
32. Steinberg EB, Greene KD, Bopp CA, Cameron DN, Wells JG, Mintz ED: Cholera
in the United States, 1995-2000: trends at the end of the twentieth century. J Infect
Dis 2001, 184(6):799-802.
33. Cholera [http://www.who.int/gho/epidemic_diseases/cholera/en/]
34. Mengel MA, Delrieu I, Heyerdahl L, Gessner BD: Cholera outbreaks in Africa.
Current topics in microbiology and immunology 2014, 379:117-144.
35. von Seidlein L, Wang XY, Macuamule A, Mondlane C, Puri M, Hendriksen I, Deen
JL, Chaignat CL, Clemens JD, Ansaruzzaman M et al: Is HIV infection associated
with an increased risk for cholera? Findings from a case-control study in
Mozambique. Tropical medicine & international health : TM & IH 2008,
13(5):683-688.
36. Brennan RJ, Nandy R: Complex humanitarian emergencies: A major global health
challenge. Emergency Medicine 2001, 13(2):147-156.
37. Toole MJ: Mass population displacement. A global public health challenge.
Infectious disease clinics of North America 1995, 9(2):353-366.
38. Public health impact of Rwandan refugee crisis: what happened in Goma, Zaire, in
July, 1994? Goma Epidemiology Group. Lancet 1995, 345(8946):339-344.
39. Nelson EJ, Nelson DS, Salam MA, Sack DA: Antibiotics for both moderate and
severe cholera. The New England journal of medicine 2011, 364(1):5-7.
40. Holbourn KP, Shone CC, Acharya KR: A family of killer toxins. Exploring the
mechanism of ADP-ribosylating toxins. The FEBS journal 2006, 273(20):4579-
4593.

44
41. Chatterjee A, Mahalanabis D, Jalan KN, Khatua SP: Plasma specific gravity and
haematocrit values as indices of the degree of dehydration in infantile diarrhoea.
The Indian journal of medical research 1979, 70:229-232.
42. Saha D, Karim MM, Khan WA, Ahmed S, Salam MA, Bennish ML: Single-dose
azithromycin for the treatment of cholera in adults. The New England journal of
medicine 2006, 354(23):2452-2462.
43. Cohen MB, Giannella RA, Losonsky GA, Lang DR, Parker S, Hawkins JA,
Gunther C, Schiff GA: Validation and characterization of a human volunteer
challenge model for cholera by using frozen bacteria of the new Vibrio cholerae
epidemic serotype, O139. Infection and immunity 1999, 67(12):6346-6349.
44. Weil AA, Khan AI, Chowdhury F, Larocque RC, Faruque AS, Ryan ET,
Calderwood SB, Qadri F, Harris JB: Clinical outcomes in household contacts of
patients with cholera in Bangladesh. Clinical infectious diseases : an official
publication of the Infectious Diseases Society of America 2009, 49(10):1473-1479.
45. Mahalanabis D, Wallace CK, Kallen RJ, Mondal A, Pierce NF: Water and
electrolyte losses due to cholera in infants and small children: a recovery balance
study. Pediatrics 1970, 45(3):374-385.
46. South Sudan Centre for Census SaE: Statistical Yearbook for South Sudan 2010.
In.: SSCCSE; 2010.
47. Trombley M: Strategy for Integrating a Gendered Response in Haiti’s Cholera
Epidemic. In. Haiti: UNICEF; 2010.
48. Andre FE, Booy R, Bock HL, Clemens J, Datta SK, John TJ, Lee BW, Lolekha S,
Peltola H, Ruff TA et al: Vaccination greatly reduces disease, disability, death and
inequity worldwide. Bulletin of the World Health Organization 2008, 86(2):140-
146.
49. Wierzba TF, Kar SK, Mogasale VV, Kerketta AS, You YA, Baral P, Khuntia HK,
Ali M, Kim YH, Rath SB et al: Effectiveness of an oral cholera vaccine campaign
to prevent clinically-significant cholera in Odisha State, India. Vaccine 2015,
33(21):2463-2469.
50. Ruiz-Palacios GM, Perez-Schael I, Velazquez FR, Abate H, Breuer T, Clemens SC,
Cheuvart B, Espinoza F, Gillard P, Innis BL et al: Safety and efficacy of an
attenuated vaccine against severe rotavirus gastroenteritis. The New England
journal of medicine 2006, 354(1):11-22.
51. Preziosi MP, Halloran ME: Effects of pertussis vaccination on disease: vaccine
efficacy in reducing clinical severity. Clinical infectious diseases : an official
publication of the Infectious Diseases Society of America 2003, 37(6):772-779.
52. Vazquez M, LaRussa PS, Gershon AA, Niccolai LM, Muehlenbein CE, Steinberg
SP, Shapiro ED: Effectiveness over time of varicella vaccine. JAMA : the journal of
the American Medical Association 2004, 291(7):851-855.
53. Mangtani P, Abubakar I, Ariti C, Beynon R, Pimpin L, Fine PE, Rodrigues LC,
Smith PG, Lipman M, Whiting PF et al: Protection by BCG vaccine against
tuberculosis: a systematic review of randomized controlled trials. Clinical
infectious diseases : an official publication of the Infectious Diseases Society of
America 2014, 58(4):470-480.
54. Prevention and control of cholera outbreaks: WHO policy and recommendations
[http://www.who.int/cholera/technical/prevention/control/en/]
55. WorldHealthOrganisation: Situation Report: Health Bulletin #36 South Sudan. In.
South Sudan; 2015.
56. Philipson H, Ekman I, Forslund HB, Swedberg K, Schaufelberger M: Salt and fluid
restriction is effective in patients with chronic heart failure. European journal of
heart failure 2013, 15(11):1304-1310.
57. Siddique AK: Failure of treatment centres to prevent cholera deaths in Goma.
Lancet 1995, 346(8971):379.
58. Siddique AK: Cholera epidemic among Rwandan refugees: experience of ICDDR,B
in Goma, Zaire. Glimpse (Dhaka, Bangladesh) 1994, 16(5):3-4.

45
59. Faruque AS, Eusof A, Rahman AS, Zaman K: Study of makeshift hospital during
cholera outbreak. Bangladesh Medical Research Council bulletin 1984, 10(2):45-
52.
60. Siddique AK, Mutsuddy P, Islam Q, Majumder Y, Akram K, Zaman K: Makeshift
treatment centre during a cholera epidemic in Bangladesh. Tropical doctor 1990,
20(2):83-85.

46