SCHOOL OF PUBLIC HEALTH
COLLEGE OF HEALTH SCIENCES
UNIVERSITY OF GHANA

ROTAVIRUS DIARRHEA REINFECTION AND ITS OUTCOME ON
WEIGHT AND HOSPITALIZATION – DURATION AMONG
CHILDREN IN GHANA: A RANDOMIZED, DOUBLE – BLINDED,
PLACEBO – CONTROLLED TRIAL

CLEMENT TETTEH NARH

THIS THESIS/DISSERTATION IS SUBMITTED TO THE
UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF
THE REQUIREMENT FOR THE AWARD OF A MASTER OF
SCIENCE IN CLINICAL TRIALS DEGREE

JULY 2012
University of Ghana http://ugspace.ug.edu.gh

DECLARATION

I, Clement Tetteh Narh declare that except for the other people‟s investigations which
have duly been acknowledged in this dissertation. This work is the result of my own
original research carried out for the award of a Master of Science (Clinical Trials) degree.
This dissertation has not been presented elsewhere either in whole or in part for another
degree.

Author ……...…………………………………………
Clement Tetteh Narh
(MSc. Clinical Trial Resident – School of Public Health, Legon)

Supervisor ……...…………………………………………
Prof. George E. Armah
(Senior Faculty Member – School of Public Health and
Noguchi Memorial Institute for Medical Research)

University of Ghana http://ugspace.ug.edu.gh

DEDICATION

This dissertation is dedicated to my dear wife Naa Merley and the clinical trials baby,
Kea Badjo Narh.

University of Ghana http://ugspace.ug.edu.gh

iii

ACKNOWLEDGMENT
I foremost thank God for his mercies.
I wish to acknowledge with much gratitude the financial support received from the
Indepth Network through the Indepth Network Effectiveness and Safety Study (INESS)
Project, and the Dododwa Health Research Centre (DHRC) for making this dream a
reality.
My greatest appreciation and gratitude go to the staff and management of the School of
Public Health (SPH), University of Ghana, with particular reference to my Supervisor
and Mentor, Professor George E. Armah (Faculty – School of Public Health and Noguchi
Memorial Institute for Medical Research) under whose watchful eyes my dream has
come true.
Many thanks and appreciations also go to my director at the DHRC Dr. Margaret
Gyapong for her priceless support both in school and out of school.
My acknowledgment would be incomplete without paying a very special compliment to
the directors and staff of the following research centres, Navrongo Health Research
Centre (NHRC), Kintampo Health Research Centre (KHRC), and the Agogo Malaria
Vaccine Centre (MRC) for their tuition and support during our field residency.
To you all who may get the opportunity to read this thesis or may not, I extend my best
wishes to you for all your support as I count my blessings and name them one by one!
May the good Lord richly reward you!
University of Ghana http://ugspace.ug.edu.gh

ABSTRACT
Background
Rotavirus gastroenteritis is a major contributor to the overall burden of diarrhea disease
in Africa. Close to 40% of hospital admissions are as a result of diarrhea in children
below the ages of five years are traceable to rotavirus. There is incomplete immunity after
infection; however repeated infections tend to be less severe than the severity of the first
rotavirus infection.
Methods
The study estimated the effect of repeated episodes of the number of severe rotavirus
diarrhea on the weight of child at study entry, or hospitalization – duration using
proportions and regression models. Univariate and bivariate linear and logistic regression
models were used in testing for an association of the repeated episodes of the number of
severe rotavirus diarrhea and confounding variables on the weight of child at study entry,
or hospitalization – duration as outcomes.
Results
1098 children were randomly assigned to the Vaccine arm and 1102 children to the
Placebo arm. Of the three age groups studied, almost 45% of the children in both arms
were in the age group 7 – 9 weeks. A cumulative total of 142 and 151 severe diarrhea
episodes were recorded in the Vaccine and Placebo arms respectively. However there
were only 30 repeated episodes of gastroenteritis recorded. Of these were 16 cases in the
vaccine arm and 14 in the control arm. The occurrence of repeated gastroenteritis was
University of Ghana http://ugspace.ug.edu.gh

also related to the age at recruitment and was common in the younger age group.
Children in the age group 7 – 9 weeks suffered weight loss and hospitalization during the
first severe diarrhea episode (Vaccine 66/142 and Placebo 69/151). Fifty-four percent of
all children who had diarrhea were hospitalized and 46.0% were outpatients in the
Vaccine arm and in the Placebo arm inpatients were 59.7% and outpatients were 40.3%. .
The predominant G genotype of rotaviruses identified in diarrhea stools was G2 (Vaccine
46.5% and Placebo 45.0%).
Children randomized between 7 – 9 weeks and age group 10 – 12 weeks of age were
observed to have greater reduction in weight than the younger age groups (P-
Value=0.012 and P-value <0.001 respectively).
After adjusting for age group, number of severe diarrhea episode showed a statistical
evidence (P-value <0.001) of an association with average weight change in age group 7 –
9 weeks and 10 – 12 weeks in the Vaccine and Placebo arms. The regression coefficient
for age group 7 – 9 weeks changed from -0.4 (95% CI=-0.32, 0.24) to -0.08 (95% CI=-
0.31, 0.15) in the Placebo arm. Further indicating that vaccination at an earlier age
protects against severe diarrhea.
Conclusion
The risk of getting severe diarrhea episode is proportional to age, therefore if children are
vaccinated at an early age, the lesser the risks of an infection. There was no link between
G genotype and occurrence of repeat rotavirus diarrhea episodes. There was progression
in hospitalization with increase in the number severe rotavirus diarrhea episodes.
University of Ghana http://ugspace.ug.edu.gh

TABLE OF CONTENTS
Declaration ……. ………………………………………………………...………………..i
Dedication ……...………………………………………………………...……………….ii
Acknowledgment ...……………………………………………………...……………….iii
Abstract ...……...………………………………………………………...………………iv
Table of Contents ..…………………………………………………..……………...……vi
List of Figures .……………………….……………………….………..……...…………xi
List of Tables ......…………………………..……...………...…………..…..………..…xii
List of Acronyms .….……………………………..…………...……...…….……..……xiii
CHAPTER ONE ………………………………………………………………………….1
1.0 INTRODUCTION .………………………………………………………...…………1
1.1 Background ….…………………………………………………………….……..…...1
1.2 Rationale …………..……………………………….…………..…………………......3
1.3 Statement of the problem …………………………….……………………….…..…..4
1.4 Study Objectives………………………………..………………………………..……4
1.4.1 General Objective………………………...………………………….……..…….....4
1.4.2 Specific Objectives ...…………………………..………………………...…………4

University of Ghana http://ugspace.ug.edu.gh

vii

CHAPTER TWO ..………………………………………………………………………..5
2.0 LITERATURE REVIEW…………………………..…………………………..……5
2.1 Overview of Diarrhea Disease …………………………...…………………..……….5
2.2 Rotavirus Genotypes ………………………………………………………..…….…..8
2.3 Reinfection Rotavirus of Diarrhea …………………….………..…………………...11
CHAPTER THREE ……………………………………………………………………..13
3.0 METHODOLOGY…………………………………………………..………..……..13
3.1 Study Location ……………………………………………………...……………….13
3.2 Study Design …………………………………………………………...……………14
3.3 Sample Size and Power ………………………………………………...……………15
3.4 Research Questions ………………………………………………………………….15
3.5 Data Source and Variables …………………………………………………………..16
3.5.1 Outcome Variables ……….………………………………………………………..16
3.5.2 Exposure Variable ………..………………………………………………………..16
3.5.3 Confounding Variables ………..…………………………………………………..17
3.6 Data Collection Methods …………………………………………………...…...…..17
3.7 Laboratory Methods .…………….…………………………………………………..18
3.7.1 Rotavirus Antigen Detection Assay ………….………………………………..…..18
3.7.2 Rotavirus VP6 Assay ………….…………………………………………………..18
University of Ghana http://ugspace.ug.edu.gh

viii

3.7.3 Rotavirus VP7 Serotyping Assay ………….…………….…..…………...………..19
3.7.4 Rotavirus VP4 Serotyping Assay ………….……..………...……………….……..19
3.8 Outcome Measures ……………………………………………………………..……20
3.9. Statistical Analysis Plan …………………………………………………...………..20
3.9.1 Data Description and Extraction ……………………………………………..……20
3.9.2 Categorization of Variables ………………………………………...……………..21
3.9.3 Missing Data Management …………………………………………..……………21
3.9.4 Statistical Analysis …………………………………………………..…………….21
3.9.5 Definition for Measure of Association ……………………………………………22
3.9.6 Methods for Analysis …………………………………………………..………….22
3.9.7 Identifying Risk Factors …………………………………………………...………23
3.10 Analysis of Data ……………………………………………………….…………...23
3.11 Limitations ……………………………………………………………….………..24
3.12 Ethical Considerations ………………………………………………...…..……….24

University of Ghana http://ugspace.ug.edu.gh

CHAPTER FOUR ...……………………………………………………………………..26
4.0 RESULTS ………………....……………………...…………..……………………..26
4.1 Introduction ..…………...…………………………………………………...……….26
4.2 Baseline Characteristics of Study Participants …...………………………...……….26
4.3 Distribution of Rotavirus Severe Diarrhea Episodes, Detected Genotypes and
Hospitalization ………………………………………………………………………….27
4.4 Unadjusted Linear Regression for Average Weight Change of Study Participants
…………………………………………………………………………………………....30
4.5 Unadjusted Logistic Regression for Hospital – Duration of Study Participants ……32
4.5 Adjusted Linear and Logistics Regression for Average Weight Change and Hospital –
Duration of Study Participants …………………………………………..........…34
CHAPTER FIVE ………………………………………………………………………..37
5.0 DISCUSSION ……………………………………………………………………….37
5.1 Introduction ………………………………………………………………………….37
5.2 Rotavirus Diarrhea Episode and Reinfection ………….………………...…………..37
5.3 Rotavirus G Genotypes ……………………………………………………………...38
5.4 Comparism of outcomes (average weight change of child in the study and
Hospitalization – Duration) ……………………………………………………...39
University of Ghana http://ugspace.ug.edu.gh

CHAPTER SIX ….………………………………………………………………………41
6.0 CONCLUSION AND RECOMMENDATION ……..………………………………41
6.1 Conclusion …………………………………………………………………………..41
6.2 Recommendation …………..………………………………………………………..42
3.14 References ………………………………………………………………...………..43

University of Ghana http://ugspace.ug.edu.gh

LIST OF FIGURES
Figure 1: Proportional distribution of diarrhea cases among children under five years of
age, by region, 2004 ………………………………………………………………………6
Figure 2: Map of Kassena – Nankana District (KND) Showing Location of Health
Facilities …………………………………………………………………………………12
Figure 3: Rotavirus Genotypes and Number of Diarrhea Episodes ……………………..28
Figure 4: Dot Plot of the Effect of Number of Rotavirus Diarrhea Reinfection on Weight
of Ghanaian Children in the Rotavirus Diarrhea Reinfection and Its Outcome on Weight
and Hospitalization – Duration at Study Entry …………………………………………30

University of Ghana http://ugspace.ug.edu.gh

xii

LIST OF TABLES
Table 1: Baseline Characteristics of Study Participants ………………………...………27
Table 2: Distribution of Rotavirus Severe Diarrhea Episodes, Detected Genotypes and
Hospitalization …………………………………………………………………………..29
Table 3: Unadjusted Linear Regression of Study Participants .…..…………….……….31
Table 4: Unadjusted Logistics Regression for Hospital – Duration of Study Participants
…………………………………………………………………………………………....33
Table 5: Adjusted Linear and Logistics Regression for Age and Sex of Study Participants
……………………………………………………………………………………………29

University of Ghana http://ugspace.ug.edu.gh

xiii

LIST OF ACRONYMS

CRF Case Report Form
EPI Expanded Programme on Immunization
EU European Union
KND Kassena – Nankana District
NMIMR Noguchi Memorial Institute for Medical Research
USA United States of America
WHO World Health Organization
95%CI 95% Confidence Interval
OR Odds Ratio
LR Likelihood Ratio

University of Ghana http://ugspace.ug.edu.gh

CHAPTER ONE
1.0 INTRODUCTION
1.1 Background
Diarrhea is a major cause of morbidity and mortality in young children, causing
approximately 3 million deaths per year. 1, 2 About 20% of these diarrhea episodes are
caused by rotavirus.3 Rotaviruses are double stranded RNA viruses and members of the
Reoviridae with a genome consisting of 11 segments of dsRNA.4 Rotaviruses are triple-
layered particles with the middle and outer layers comprising of viral proteins VP6, VP7
and VP4. They are classified into groups, subgroups according to the characteristics if the
VP6 antigen and genotypes by the VP7 and VP4.5, 6
Rotavirus infection is common worldwide but 80% of all mortality occurs in the
developing countries with Africa and Asia.7 In Ghana, rotavirus infection is responsible
for 39% to 58.9% of diarrheal cases in children. 5, 8, 9 Rotaviruses are transmitted through
the fecal-oral route, but some studies have reported evidence for possible respiratory
transmission.10 Rotavirus illness follows an incubation period of 24 to 48 hours followed
by a progression from asymptomatic to severe diarrhea.11 Vomiting, fever, mild to severe
dehydration, abdominal pain and respiratory disturbances are the clinical symptoms
which usually accompany the watery diarrhea.12
Studies that have been conducted to determine the distribution of rotavirus G and P types
circulating on the African continent indicated both diverse population of co – circulating
types and co – infection with more than one type. Steele in 2000 found out that Rotavirus
University of Ghana http://ugspace.ug.edu.gh

G1P[8] was the most common strain found. 14, 13 Furthermore, the detection of different
G and P types indicated that novel viruses may have originated through re – assortment in
the study area.
Due to the large effect of morbidity and mortality associated rotavirus infection in
children, the WHO recommends the inclusion of rotavirus vaccination of infants
alongside improved hygiene and sanitation, and oral rehydration therapy in Ghana‟s
immunization programmes.
Close to 40% of hospital admissions in children below the ages of five years are due to
diarrhea caused byrotavirus.15, 23 Although repeated infections tend to be less severe than
the first rotavirus infection.
In a recent prospective study by Armah et al in 2010 of children with group A Rotavirus
infections and reinfections, no evidence of reinfection by group A rotavirus were found.15
This finding was in contrast with a earlier studies that recorded variable levels of
rotavirus reinfection. 51-57

University of Ghana http://ugspace.ug.edu.gh

1.2 Rationale of the Study
The predominance of rotaviruses in causing morbidity and mortality in children is a
major issue for healthcare providers and parents. Although hand washing and improved
sanitation has helped reduce the incidence of diarrhea in children, notably those
attributable to bacteria and parasites diarrhea attributable to rotaviruses viruses has not
been reduced. Presently the only intervention available is by vaccination. As part of
global efforts to find a rotavirus vaccine, RotaTeq vaccine been licensed and is being
used in the United States of America (USA). It was tested in Ghanaian children to
determine its Efficacy, Safety, and Immunogenicity between 2006 and 2009.15 Whilst
the study looked at vaccine safety, efficacy and immunogenicity; data was also collected
on the circulating strain types, number of re-infections and co-infections in addition to
other information.
This study was a secondary analysis of a trial data to determine the number of diarrhea
reinfections attributable to rotavirus and the presence or absence to strain types. The
study will find out the effect of the number of rotavirus diarrhea reinfection on growth of
child and the hospital - duration (as a proxy severity) for diarrhea. Finally, the study
estimated and determined confounding variables associated with any of these outcome
variables, thus average weight change and hospital – duration of the children as they
entered into the trial.

University of Ghana http://ugspace.ug.edu.gh

1.3 Statement of the problem
Considering the impact of rotaviruses in causing morbidity and mortality in children, it
was necessary to determining if there is any relation between the number of diarrhea
infection and the risk factors such as rotavirus genotypes, age and sex of the children with
the average weight change of the children and severity of infection.

1.4 Study Objectives
1.4.1 General Objective
To estimate rotavirus diarrhea reinfection and determine its outcome on weight of child at
study entry and hospitalization during the first 106 weeks of life - duration among
children in the Kassena Nankana District of Ghana.
1.4.2 Specific Objectives
1. To estimate rotavirus diarrhea reinfection among children in Ghana.
2. To determine rotavirus genotypes that causes severe diarrhea reinfection requiring
hospitalizations among children in the Kassena Nankana District of Ghana.
3. Determine and Compare outcomes (average weight change of child in the study and
Hospital – Duration) of Rotavirus Diarrhea reinfection among children in Ghana.

University of Ghana http://ugspace.ug.edu.gh

CHAPTER TWO
2.0 LITERATURE REVIEW
2.1 Overview of Diarrhea Disease
Diarrhea is defined as the passage of more than 3 looser than normal stools over a 24
hour period.12 The stools could be either watery or in addition contain some blood or
mucus. The diarrhea disease could be either acute or chronic: An acute diarrhea usually
has a duration of 1 or 2 days whilst, chronic diarrhea is usually greater than 14 days in
duration. The symptoms of chronic diarrhea may be continual or they may be recurring.
Any duration of diarrhea episode may cause dehydration, which will result in loss of
essential electrolytes needed to function properly. People of all ages are susceptible to the
diarrhea disease. 16
An estimated 2.5 billion cases of diarrhea occur among children under five years of age
each year, and estimates suggest that overall incidence comparatively remained stable
over the past two decades. 17, 18 Africa and South Asia account for more than half of these
cases (Figure 1), where bouts of diarrhea are more likely to result in death or other severe
outcomes. The variation of incidence of diarrhea diseases is greatly accounted for by the
seasons and a child‟s age. The most vulnerable are youngest children: Incidence is
highest in the first two years of life and as a child grows older it declines.

University of Ghana http://ugspace.ug.edu.gh

Figure 1: Proportional distribution of diarrhea cases among children under five
years of age, by region, 2004 [Source: Based on World Health Organization, Global Burden of
Disease estimates, 2004 update. The proportional distribution for UNICEF regions was calculated by
applying the WHO cause of death estimates to the most recent estimates for the total number of under-five
deaths (2007)].

Diarrhea related deaths are the second leading cause of mortality in children younger than
five years in most developing countries, with an estimated 1.3 million deaths occurring
worldwide annually of which 800 000 deaths occur in Africa. 15, 19, 20
Bacterial infections, viral infections, parasites, functional bowl disorders, intestinal
diseases, food intolerances and sensitivities, and reaction to medicines are the most
common causes of diarrhea.21 There are several types of bacteria which can infect us
through contaminated food or water. The most common of these include Campylobacter,
Salmonella, Shigella, and Escherichia coli (E. coli). There are many viruses responsible
for causing diarrhea; these viruses include rotavirus, norovirus, cytomegalovirus, herpes
simplex virus, and viral hepatitis. The rotavirus infection is the most common cause of
acute diarrhea in children and the very elderly.22
435
Million
480
Million
783 Million
696 Million
East Asia & Pacific Rest of the World South Asia Africa
University of Ghana http://ugspace.ug.edu.gh

Rotavirus gastroenteritis is a major contributor to the overall burden of diarrhea disease
in Africa. Close to 40% of hospital admissions as a result of diarrhea in children below
the ages of five years are traceable to rotavirus.15, 23 Six of the seven countries with the
highest mortality from rotavirus (>500 deaths per 100000 livebirths) are in sub-Saharan
Africa, where nearly 240 000 rotavirus-related deaths out of the global yearly estimate of
527 000 occur.15, 24, 25
The most common cause of acute childhood diarrhea in both developed and developing
countries are rotaviruses.26 In 2001, the rotavirus-induced diarrhea episodes that occurred
globally was around 125 million, resulting in 500,000 to 600,000 deaths.27, 28
Diarrhea may be accompanied by the following symptoms; cramp, abdominal pain,
nausea, or loss of bowel control which sometimes require an urgent need to use the
bathroom. However, some infections that cause diarrhea can also cause a fever and chills
or bloody stools. Dehydration in children, older adults, and people with weakened
immune systems is particularly dangerous. Due to this danger, dehydration must be
treated promptly to avoid serious health problems, such as organ damage, shock, or
coma.21 Rotavirus-positive episodes of diarrhea are tended to be more acute, to be
associated with vomiting, caused greater dehydration, and were more likely to require
hospitalization.5
Some common dehydration in signs infants and young children below five years include
dry mouth and tongue, no tears when crying, no wet diapers for 3 hours or more, sunken
eyes, cheeks, or soft spot in the skull, high fever, and listlessness or irritability. These
University of Ghana http://ugspace.ug.edu.gh

signs dehydration makes rotavirus diarrhea reinfections a serious problem for patients
and parents/guardians alike.
In terms of diarrhea prevention, most of the efforts are concentrated on rotavirus since it
is a major cause of morbidity and mortality in children. In the USA, two oral vaccines
have been approved by the USA Food and Drug Administration to protect children from
rotavirus infections. The two vaccines are rotavirus vaccine which is live, oral,
pentavalent (RotaTeq) and rotavirus vaccine which is also live and oral (Rotarix).
RotaTeq is given to infants in three doses at 2, 4, and 6 months of age. Rotarix is given in
two doses. The first dose is given when infants are 6 weeks old, and the second is given
at least 4 weeks later but before infants are 24 weeks old.15
Due to the established efficacy of these vaccines in the USA, European Union (EU), and
other countries, the WHO intends to include RotaTeq in the EPI programmes in most
African countries. Therefore, the WHO Expert Committee on Biological Standardization
has recommended “new” rotavirus efficacy should be demonstrated in various
geographical regions especially developing countries before a wider implementation
takes place. 29

University of Ghana http://ugspace.ug.edu.gh

2.2 Rotavirus Genotypes
The human group A rotaviruses are the major etiological agents accountable for acute
diarrhea in children under the age of 5 years globally. The high rates of morbidity
throughout the world and mortality in developing countries, accounting for more than
500,000 infant deaths per year, at least before vaccine introduction is associated with the
infection.24 The classification of rotavirus strains into G-types and P-types is based on the
genetic and antigenic diversity of the two outer capsid VP7 and VP4 proteins
respectively.30 The two proteins are independently able to elicit neutralizing antibodies
and induce protective immunity. Decades ago, rotavirus molecular genotyping has
demonstrated the diversity of the VP7 and VP4 strains going around the world.31 In
recent times, there has been a proposed 11 genome segments sequence-based advanced
form of the original genotyping concept.32 Among the group A rotaviruses, there are at
least 23 different G genotypes (15 G serotypes) and 31 different P genotypes (15 P
serotypes) have been identified.31 - 40 The incidence variation and distribution of rotavirus
G-types and P-types between geographical areas during a rotavirus season and from one
season to the next is a common feature in rotavirus epidemiology.5, 31 The G-types and P-
types combinations, G1P[8], G2P[4], G3P[8], G4P[8], and G9P[8], are the five most
commonly associated with human infections globally.30,31 In addition, the existence of
different genotypes is not limited to genes encoding structural proteins.41,46
The two vaccines currently in use by US, EU, etc., are Rotarix1 (GlaxoSmithKline,
Rixensart, Belgium), containing a single human G1P[8] rotavirus strain, and RotaTeq1
(Merck & Co., Whitehouse Station, NJ), containing 5 human-bovine re – assortant
rotavirus strains expressing five human serotypes (G1, G2, G3, G4, and P[8]) have
University of Ghana http://ugspace.ug.edu.gh

currently been licensed whiles licensing trials in other countries have been completed.
These vaccines have been reported to stimulate significant protection against severe
diarrhea caused by homotypic and heterotypic rotavirus strains43-45 and reduce childhood
deaths.46
The epidemiological surveys of the spread of rotavirus types are critical in developing
countries to determine the protective efficacy of rotavirus vaccines against multiple
serotypes and to detect the eventual emergence of antigenically different strains.15 A
monitoring of rotavirus infection conducted in the past two decades in Venezuela showed
the clinical importance of rotavirus disease in children, and rotavirus was also recognized
as the major cause of death due to the diarrhea.47,48 A clinical trial conducted in Latin
America to determine efficacy, safety and immunogenicity of the Rotarix1 vaccine in
2001–2002, showed that G1 strains predominated, but multiple rotavirus genotypes also
circulate in Venezuela.49

University of Ghana http://ugspace.ug.edu.gh

2.3 Reinfection of Rotavirus Diarrhea
Re-infection is common and the first infection is most severe and subsequent ones less
severe.11 The most common cause of severe diarrhea among children is rotavirus. Until
the introduction of rotavirus vaccines in the United Sates in 2006, approximately 55,000
U.S. children are hospitalized each year as a result of rotavirus.49 Globally, rotavirus is
estimated to cause 527,000 deaths in children annually.50 The incubation period for
rotavirus disease is approximately 2 days. Rotavirus disease is characterized by vomiting
and watery diarrhea for 3 to 8 days, and fever and frequent occurrence of abdominal pain.
There is incomplete immunity after infection; however repeated infections tend to be less
severe than the first rotavirus infection. Children who are vaccinated and unvaccinated
may develop rotavirus disease more than once due to many different types of rotavirus
and because neither vaccine nor natural infection provide full immunity (protection) from
future infections.
The characteristic of a rotavirus is a wheel-like appearance when viewed by electron
microscopy (the name rotavirus is derived from the Latin rota, meaning "wheel"). The
rotaviruses are non-enveloped and double-shelled viruses.29
The epidemiology of rotavirus shows that the rotavirus uses the fecal-oral route as
primary mode of transmission. Since the virus is stable in the environment, transmission
can occur through ingestion of contaminated water or food and contact with contaminated
surfaces or objects. The USA and other countries with a temperate climate, the disease
has a seasonal pattern, with annual epidemics occurring from December through June.
The highest rates of rotavirus illness occur among infants and young children.
University of Ghana http://ugspace.ug.edu.gh

The diagnosis of rotaviruses may be made by rapid antigen detection of rotavirus in stool
specimens. The strains of the rotavirus may be further characterized by enzyme
immunoassay or reverse transcriptase polymerase chain reaction, but this testing is not
commonly performed.

University of Ghana http://ugspace.ug.edu.gh

CHAPTER THREE
3.0 METHODOLOGY
3.1 Study Location
The original study was conducted in the Kassena – Nankana District (KND) in northern
Ghana. The KND lies within the Guinea Savannah woodland of northern Ghana and are
among the districts with the worst social economic status in the country. The district
covers a land area of 1675 square kilometers and shares borders with Burkina Faso in the
Upper East region of Ghana. The KND has a population of 140000 people living in
approximately 13000 dispersed compounds. The KND is one of the most arid districts in
northern Ghana with a long dry season punctuated with only three months of rains and
average temperatures ranging between 20 and 40 degrees Celsius. Lack of nutritious meal
is common in the KND worsening the morbidity and mortality of infectious diseases.
Poverty and economic challenges has been retarding efforts to improve health conditions
in the district. There is the existence of evidence that the Upper East region of Ghana is
among the poorest regions in Ghana, as a result, the government of Ghana has designated
the KND as a special research zone for assessing health problems in the area and
practical means of assessing them. The KND has low levels of education and literacy.59

University of Ghana http://ugspace.ug.edu.gh

Figure 2: Map of Kassena – Nankana District (KND) Showing Location of Health
Facilities [Source: Owusu-Agyei et al. (2007). Assessing malaria control in the Kassena – Nankana
district of northern Ghana through repeated surveys using the RBM tools]

3.2 Study Design
The study was a randomized, double-blinded, placebo-controlled trial to be conducted in
Ghana to estimate rotavirus diarrhea reinfection and key factors associated with the
reinfection among children in Ghana.

University of Ghana http://ugspace.ug.edu.gh

3.3 Sample Size and Power
The number of rotavirus diarrhea reinfections was estimated with its associated 95%
confidence interval. The type I error (α) was controlled at α = 0.025 (one-sided). Based
on the assumption that the true reinfection was 75% and the captured attack rate was
1.5% with an unevaluable rate of 20% of subjects, a sample size of 5468 (1RotaTeq: 1
placebo) provided approximately 93% power to declare rotavirus diarrhea reinfection
significant. This assumption was based on the Efficacy, Safety, and Immunogenicity of
RotaTeq among Infants in Asia and Africa study in Ghana.29

3.4 Research Questions
1. How many children have had rotavirus diarrhea reinfection after the first dose of
vaccination?
2. What is the predominant rotavirus genotype that causes diarrhea reinfection among
children in Ghana?
3. What is the effect of the number of severe rotavirus diarrhea reinfection on the weight
of child at study entry?
4. What is the effect of the number of repeated severe rotavirus diarrhea episodes on
hospitalization – duration?
5. Could the following confounding variables:
i. Rotavirus genotype response diarrhea reinfection
ii. Child‟s age at first dose of vaccination
University of Ghana http://ugspace.ug.edu.gh

iii. Sex of child and
iv. Trial Arm of child has an association with the effect of severe rotavirus diarrhea
reinfection on its outcomes?

3.5 Data Source and Variables
The data set that was used for this analysis contained information on 2227 children who
participated in the Efficacy, Safety, and Immunogenicity of RotaTeq among Infants in
Asia and Ghana study in Ghana.
The data set was collected from the principal investigator in Ghana for this study. The
whole data set was available for maximum statistical output since the sample size was
large enough to do an epidemiological analysis. The following variables were analyzed in
the study:
3.5.1 Outcome Variables
i. Weight of child at study entry and
ii. Hospitalization – Duration

3.5.2 Exposure Variable
i. Number of Rotavirus Diarrhea Reinfection after first dose of vaccination

University of Ghana http://ugspace.ug.edu.gh

3.5.3 Confounding Variables
i. Rotavirus genotype response diarrhea reinfection
ii. Child‟s age at first dose of vaccination
iii. Sex of child
iv. Trial Arm of child.

3.6 Data Collection Methods
A Case Report Forms (CRFs) were used in the primary data collection. The study was
based at the designated study health facilities. Trained field staff assigned to the
designated health facilities screened the children and mothers/guardians to make sure that
the inclusion and exclusion criteria of the study are met.
For this study the inclusion criteria was children with one or more gastroenteritis
(particular focus on diarrhea) after first dose of vaccination.29

University of Ghana http://ugspace.ug.edu.gh

3.7 Laboratory Methods
3.7.1 Rotavirus Antigen Detection Assay
The purpose of the Rotavirus Antigen Detection Assay was to detect rotavirus in stool
before and after vaccination with the rotavirus content vaccine. The results for the assay
were reported as positive (+) or negative (-) for the presence of rotavirus.
The assay is valid if a Plate Control is found to be positive by both cut – off rules for
positive and negative samples (positive ODs must be >.31 and must be at least 1.63 fold
than negative ODs) and the maximum/minimum ratio of the duplicate OD readings for
the control should not exceed 1.67 (or 1.23). 29

3.7.2 Rotavirus VP6 Assay
The purpose of the rotavirus VP6 assay was to determine the VP6 genotype of any
rotavirus present in stool samples. Samples from patients with acute gastroenteritis were
first evaluated for the presence of rotavirus antigen in the stool samples and all antigen –
positive samples were then evaluated to distinguish between wild type rotavirus strains
from vaccine – virus strains. The results for the assay were reported as the species with
the closest identity to the resulting sequence. 29

University of Ghana http://ugspace.ug.edu.gh

3.7.3 Rotavirus VP7 Serotyping Assay
The purpose of the rotavirus VP7 serotyping assay was to determine the VP7 (G)
genotype of any rotavirus present in stool samples. The rotavirus VP7 serotyping assay
was used in the original study to determine the rotavirus VP7 genotypes (G – types) in
biological samples. The results for the assay are reported as G genotypes, for example,
G1, G2, G3, G4, G5, G8, G9, G10, or G12.29

3.7.4 Rotavirus VP4 Serotyping Assay
The purpose of the rotavirus VP4 serotyping assay was to determine the VP4 (P)
genotype of any rotavirus present in stool samples. The rotavirus VP4 serotyping assay
was used to determine the rotavirus VP4 genotypes (P – types) in biological samples. The
results for the assay were reported as G genotypes, for example P1A [8], P1B [4], P2A
[6], P2B [6], P7 [5], etc. Precision for both VP7 and VP4 was defined as a measure of the
assay‟s ability to reproduce a test sample result across independent runs of the procedure
(inter – assay) and between analysts. The assay validity criteria was based on the assay
controls and evaluated at several steps in the assay. 29

University of Ghana http://ugspace.ug.edu.gh

3.8 Outcome Measures
The secondary analysis of the dataset focused on the estimation and determination of the
effect of the number of rotavirus diarrhea reinfection on weight of child at study entry
and hospitalization – duration among the children after the first dose of vaccination. The
reinfection among the children was estimated as a child having diarrhea episode after first
dose of vaccination.
The confounding factors associated with severe rotavirus diarrhea reinfection were
considered in the subgroup analysis.

3.9 Statistical Analysis Plan
3.9.1 Data Description and Extraction
The baseline data set was used to explore the children‟s baseline characteristics and
variations using simple description methods of analysis such as summary statistics and
frequency distributions.
All the variables were identified and described. Data quality checks were performed to
screen for any outliers and missing values.

University of Ghana http://ugspace.ug.edu.gh

3.9.2 Categorization of Variables
Continuous variables such as age were categorized into three (3) groups for interpretation
of results; age 4-6 weeks, age 7-9 weeks and age 10-12 weeks. The number of rotavirus
diarrhea reinfection was categorized into one reinfection, two reinfections, and three
reinfections. Sex and trial arm of the child both had a binary outcome; sex being male or
female and trial arm being treatment or placebo arm. The rotavirus genotype responsible
for the diarrhea reinfection was grouped according the G genotypes.55
Children were considered to have reinfection of rotavirus diarrhea if they had any episode
of diarrhea after first dose of vaccination. Two (2) outcomes (weight of child at study
entry and hospitalization – duration) were the only outcome variables that were
determined.

3.9.3 Missing Data Management
Children with missing baseline data were excluded from the analysis.

3.9.4 Statistical Analysis
The baseline data that was analyzed included age (weeks), sex, weight, and height.
Severe rotavirus diarrhea reinfection was described as any diarrhea episode after first
dose of vaccination against rotavirus diarrhea.

University of Ghana http://ugspace.ug.edu.gh

3.9.5 Definition for Measure of Association
Linear regression was used to measure the effect of number of repeated severe rotavirus
diarrhea episodes on weight of child at study entry whiles confounding for rotavirus
genotype responsible for diarrhea reinfection, child‟s age at first dose of vaccination, sex
of child, and trial Arm of child.
Logistic regression was also used to measure the effect of the number of rotavirus
diarrhea reinfection on hospitalization – duration whiles confounding for rotavirus
genotype responsible for diarrhea reinfection, child‟s age at first dose of vaccination, sex
of child, and trial Arm of child.
The Logistic regression is the appropriate statistical measure of the effect because it
allows more flexibility in examining the effects of the exposure variable which is
categorical.

3.9.6 Methods for Analysis
The statistical software package Stata version 10 was used throughout the data cleaning
and the analysis.

University of Ghana http://ugspace.ug.edu.gh

3.9.7 Identifying Risk Factors
A univariate linear and logistic regression model was fitted to measure the effect each of
the exposure variables on either outcome variables. The exposure variable or any
confounder having P-value ≤0.05 was included in the regression model.

3.10 Analysis of Data
The Stata version 10 software was used during the data analysis. Simple proportions and
confidence intervals were used to estimate rotavirus diarrhea reinfection rate among
children in Ghana. Again, simple proportions were used to determine the dominant
rotavirus genotype that causes diarrhea reinfection among children in northern Ghana
using the genotype data.
In order to determine the confounding variables (rotavirus genotype responsible for
diarrhea reinfection, child‟s age at first dose of vaccination, sex of child, and trial arm of
child) associated with rotavirus diarrhea reinfection among children in Ghana, the Linear
and Logistic Univariate Regression Analysis was performed to test for an association between
each factor and average weight change or hospitalization – duration among the children in
northern Ghana.

University of Ghana http://ugspace.ug.edu.gh

3.11 Limitations
Apart from normal limitations of secondary data used to answer more research questions
other than its primary and secondary endpoints. The data answered all the research
questions required in this study except for feeding status of the children that was not
available as a risk factor average weight change.

3.12 Ethical Considerations
The Ghana Health Service Ethical Approval: Ethical clearance was obtained from the
Ghana Health Service Ethical Review Committee of the Research and Development
Division of the Ghana Health Service (GHS-ERC: 62/03/12).
Approval from Principal Investigator: Verbal approval was sought from the Principal
Investigator of the Efficacy, Safety, and Immunogenicity of RotaTeq among Infants in
Asia and Africa study in Ghana for the use of the data for a secondary analysis in this
study.
Description of Subjects Involved in The Study: There wasn‟t any vulnerable participant
involved in this study, nor were there any risk to anyone in the earlier study. Information
provided on participants involved in the secondary analysis was kept confidential.
Privacy and Confidentiality: was exercised in handling the participant‟s information
provided.
University of Ghana http://ugspace.ug.edu.gh

Voluntary Consent: Participation in the study was voluntary and the participants were
free to refuse to answer any question(s) and could opt out of the study at any time of the
study.
Conflict of Interest: I hereby declare that there was no conflict of interest apart from the
academic and public health relevance of this study. I also declare that this study was self
– sponsored and an amount of about US$1035.00 was spent in carrying out this study.
Contact Information: If there are any question(s) about this study, please direct them to
Mr. Clement T. Narh (Principal Investigator) School of Public Health, College of Health
Sciences University of Ghana, Legon.
Mobile: 0265540346/0243104190 Email: iteetee@gmail.com
Or
Academic Supervisor: Prof. George E Armah
Lecturer, School of Public Health (SPH)
Noguchi Memorial Institute for Medical Research (NMIMR)
Tel.: 0208246513 Email: garmah@noguchi,mimcom.org
Contact of Ethics Review Committee Administrator
Ghana Health Service Research Ethics Committee
Ghana Health Service
Post Office GP 184
Accra, Ghana
Phone: 233 302 681109
Fax: 233 302 226739
Email: Hannh.Frimpong@ghsmail.org

University of Ghana http://ugspace.ug.edu.gh

CHAPTER FOUR
4.0 RESULTS
4.1 Introduction
The results chapter presents the baseline characteristics of the trial participants and the
strains of rotavirus genotypes, unadjusted linear and logistic regressions, adjusted linear
and logistic regressions and a box plot.

4.2 Baseline Characteristics of Study Participants
The RotaTeq Ghana study dataset contained information on 2227 children of which 27
were excluded because they were older than 12 weeks of age. 1098 children were
randomly assigned to the Vaccine arm and 1102 children to the Placebo arm.
Of the three age groups of children in this study, almost 45% of the children in both arms
were in the age group 7 – 9 weeks. The proportion of females to males in each arm was
very similar.

University of Ghana http://ugspace.ug.edu.gh

Table 1: Baseline Characteristics of Study Participants

Characteristics
Vaccine Arm Placebo Arm
Number (%) Number (%)

Number of Infants Randomly Assigned

1098 (49.9%)

1102 (50.1%)

Age at Randomization (weeks)

4-6 297 (27.1%) 314 (28.5%)
7-9 511 (46.5%) 493 (44.7%)
10-12 290 (26.4%) 295 (26.8%)

All 1098 (100%) 1102 (100%)

Sex

Female 556 (50.6%) 557 (50.5%)
Male 542 (49.4%) 545 (49.5%)

All 1098 (100%) 1102 (100%)

Mean Weight at Enrollment 4.6 (1.0) * 4.6 (0.9)*

Mean Height at Enrollment

58.2 (3.6)*

57.9 (3.4)*

* Data are mean (SD) unless otherwise stated

Other baseline characteristics such as weight of the children at randomization, and height
at trial entry were similar between the vaccine and placebo arms of the study as shown in
Table 1.

University of Ghana http://ugspace.ug.edu.gh

4.3 Distribution of Rotavirus Severe Diarrhea Episodes, Detected Genotypes and
Hospitalization
There were a cumulative total of 142 diarrhea infections in the Vaccine arm and 151 in
the Placebo arm. Children in the age group 7 – 9 weeks suffered weight loss and
hospitalization during the first severe diarrhea episode (66/142 in the vaccine arm and
69/151 in the placebo arm). The same age group of children was mostly hospitalized as
shown in Table 2. During the study, the maximum number of times a child had repeated
episodes of diarrhea was three times in the Vaccine and two times in the Placebo arms
respectively. Fifty-four percent (54%) of all children who had diarrhea were hospitalized
and 46.0% were outpatients in the vaccine arm whiles inpatients were 59.7% and
outpatients were 40.3% in the Placebo arm respectively.
The genotypes of rotaviruses identified in diarrhea stools in the Vaccine arm were G1
(40.1%), G2 (46.5%), G3 (11.3%), and G8 (2.1%), whilst in the Placebo arm the
genotypes identified were G1 (31.8%), G2 (45.0%), G3 (13.2%), and G8 (9.9%).There
was no statistical evidence to show that the rotavirus genotype that causes that first
diarrhea infection is different from the subsequent infections in Figure 2.

Figure 3: Rotavirus Genotypes and Number of Diarrhea Episodes
0
20
40
60
80
100
120
G1 G2 G3 G8
1st Episode
2nd Episode
3rd Episode
Total
University of Ghana http://ugspace.ug.edu.gh

Table 2: Distribution of Rotavirus Severe Diarrhea Episodes, Detected Genotypes
and Hospitalization

Characteristics

Vaccine Arm

Placebo Arm

Age group (Weeks) Age group (Weeks)
4 – 6 7 – 9 10 – 12 4 – 6 7 – 9 10 – 12

Number of Severe Diarrhea
Episodes

Number (%)

Number (%)
0 253
(85.2%)
434
(84.9%)
253
(87.2%)
262
(83.4%)
418
(84.8%)
257
(87.1%)
1 42
(14.2%)
66
(12.9%)
34
(11.7%)
46
(14.7%)
69
(14.0%)
36
(12.2%)
2 1
(0.3%)
9
(1.8%)
3
(1.0%)
6
(1.9%)
5
(1.0%)
2
(0.7%)
3 1
(0.3%)
2
(0.4)
- - 1
(0.2)
-
All 297
(100%)
511
(100%)
290
(100%)
314
(100%)
493
(100%)
295
(100%)

Hospitalization

Inpatient 6
(2.0%)
6
(1.2%)
4
(1.4%)
5
(1.6%)
10
(2.0%)
4
(1.4%)
Outpatient 291
(98.0%)
505
(98.8%)
286
(98.6)
309
(98.4%)
483
(98.0%)
291
(98.6%)
All 297
(100%)
511
(100%)
290
(100%)
314
(100%)
493
(100%)
295
(100%)

Detected Genotypes

G1 57 (40.1%)
66 (46.5%)
16 (11.3%)
3 (2.1%)

142 (100%)
48 (31.8%)
68 (45.0%)
20 (13.2%)
15 (9.9%)

151 (100%)
G2
G3
G8

All

University of Ghana http://ugspace.ug.edu.gh

4.4 Unadjusted Linear Regression for Average Weight Change of Study
Participants
Table 3 shows the univariate (unadjusted) linear regression of the average weight change
(outcome), number of rotavirus infections and other baseline characteristics for the
Vaccine and Placebo arms respectively. Of all the baseline variables included in the
univariate analysis, there is strong evidence (P-value <0.001) of an associated reduction
in the average weight for multiple rotavirus infections in both the vaccine and placebo
arm of the trial. Children randomized at between 7 – 9 weeks and age group 10 – 12
weeks of age were observed to have a greater reduction in weight than the other age
groups (P-Value=0.012 and P-value <0.001 respectively). However, in the placebo there
was no statistical evidence of reduced average weight in the age group 10 – 12 weeks
group. There was no association between the sex of the children and the infecting
genotypes as shown in Table 3.

University of Ghana http://ugspace.ug.edu.gh

Table 3: Unadjusted Linear Regression of Study Participants

Average Weight
Vaccine Arm Placebo Arm
Regression
Coefficient (95% CI)
P-Value Regression
Coefficient (95% CI)
P-Value
Number of Severe
Diarrhea Episodes

1 -2.42 (-2.70, -2.15) <0.001 -2.33 (-2.59, -2.07) <0.001
2 -3.47 (-4.31, -2.64) <0.001 -3.28 (-4.13, -2.64) <0.001
3 -3.39 (-5.12, -1.66) <0.001 -3.45 (-6.37, -0.52) 0.021
Constant 4.80 (4.69, 4.91) <0.001 4.75 (4.64, 4.86) <0.001

Age at Randomization
(weeks)

7-9 -0.37 (-0.67, -0.08) 0.012 -0.4 (-0.32, 0.24) 0.80
10-12 -0.54 (-0.86, -0.22) <0.001 -0.47 (-0.78, -0.16) <0.001
Constant 4.66 (4.43, 4.89) <0.001 4.44 (4.22, 4.66) <0.001

Sex

Male 0.07 (-0.17, 0.31) 0.58 0.07 (-0.16, 0.31) 0.54
Constant 4.31 (4.14, 4.48) <0.001 4.26 (4.09, 4.43) 0.001

Detected Genotypes

G2 -0.56 (-1.83, 0.71) 0.38 -0.42 (-1.29, 0.45) 0.34
G3 0.28 (-1.89, 2.44) 0.80 -0.68 (-1.92, 0.56) 0.28
G8 0.11 (-3.91, 4.12) 0.96 -0.15 (-1.50, 1.19) 0.82
Constant 4.39 (3.45, 5.34) <0.001 4.09 (3.43, 5.43) <0.001

University of Ghana http://ugspace.ug.edu.gh

4.4 Unadjusted Logistic Regression for Hospitalization of Study Participants
There is strong evidence of an association between number of diarrhea infection and the
length of hospitalization among the children in both the Vaccine and Placebo arms (LR:
P-value <0.001). The odds (OR=11.16) of having at least one diarrheal infection is
similar among children in both the Vaccine and Placebo arms is 11 times that of those
who were not admitted with diarrhea infection. Similarly, the odds of having two
infections is about 15 times the odds of hospitalization of children without diarrhea when
Vaccine was administered as compared to the 28 times odds of hospitalization of children
in the Placebo arm. It appears that the odds of the length of hospitalization increases with
the number of infections. With regards to age, there is no statistical evidence of
association between ages and hospitalization in both arms since none of the 95%
Confidence Intervals (95% CIs) contains the null value (1 of independence). Sex of the
children in the Vaccine arm shows statistical evidence (P-value=0.03) of an association
with hospitalization but not in the Placebo arm (P-value=0.23). The odds of
hospitalization – duration among males is about 3 times that among females in the
Vaccine arm. The infecting rotavirus genotypes have no association with hospitalization
– duration irrespective of the trial arm in Table 4.

University of Ghana http://ugspace.ug.edu.gh

Table 4: Unadjusted Logistics Regression for Hospitalization – Duration of Study
Participants

Hospitalization–
Duration
Vaccine Arm Placebo Arm
Odds Ratio (95% CI) LR: P-Value Odds Ratio (95% CI) LR: P-Value
Number of Severe
Diarrhea Episodes

1 11.16 (3.60, 34.63) <0.001 11.00 (3.94, 30.75) <0.001
2 15.58 (1.69, 143.66) <0.001 28.21 (5.12, 155.52) <0.001
3 374 (29.02, 4820.05) <0.001 - -

Age at
Randomization
(weeks)

7-9 0.58 (0.18, 1.80) 0.64 1.28 (0.43, 3.78) 0.76
10-12 0.68 (0.19, 2.43) 0.64 0.85 (0.23, 3.19) 0.76

Sex

Male 3.12 (1.00, 9.75) 0.03 1.77 (0.69, 4.53) 0.23

Detected Genotypes

G2 0.33 (0.07, 1.60) 0.36 1.89 (0.34, 10.58) 0.80
G3 0.50 (0.05, 5.51) 0.36 2.78 (0.34, 22.75) 0.80
G8 - - 1.79 (0.14, 22.70) 0.80

University of Ghana http://ugspace.ug.edu.gh

4.6 Adjusted Linear and Logistics Regression for Age and Sex of Study Participants
After adjusting for age group in a bivariate linear regression analysis in Table 5, number
of infection shows statistical evidence (P-value <0.001) of having an association with
average weight change and age group 7 – 9 weeks and age group 10 – 12 weeks also
showed similar statistical evidence of an association in the Vaccine arm and in the
Placebo arm, age group 7 – 9 weeks did not again show any statistical evidence of a
reduction in the average weight change, however, the regression coefficient for age group
7 – 9 weeks changed from -0.4 (95% CI;=-0.32, 0.24) to -0.08 (95% CI=-0.31, 0.15) in
the Placebo arm.
In the adjusted logistic regression, sex of the children in both the Vaccine and Placebo
arms of the study were not having any statistical evidence of an association with hospital
– duration of the children who had rotavirus diarrhea after first dose of vaccination (P-
value=0.068 and P-value=0.305) respectively.

University of Ghana http://ugspace.ug.edu.gh

Table 5: Adjusted Linear and Logistics Regression for Age and Sex of Study
Participants

Average Weight
Linear Regression
Vaccine Arm Placebo Arm
Regression
Coefficient (95% CI)
P-Value Regression
Coefficient (95% CI)
P-Value
Number of Severe
Diarrhea Episodes

1 -2.44 (-2.72, -2.17) <0.001 -2.35 (-2.61, -2.09) <0.001
2 -3.43 (-4.25, -2.60) <0.001 -3.36 (-4.20, -2.64) <0.001
3 -3.49 (-5.20, -1.78) <0.001 -3.57 (-6.46, -0.68) 0.016

7-9 (Age in weeks) -0.33 (-0.57, -0.09) 0.007 -0.08 (-0.31, 0.15) 0.500
10-12 (Age in weeks) -0.62 (-0.89, -0.35) <0.001 -0.59 (-0.85, -0.32) <0.001

Constant 5.12 (4.92, 5.32) <0.001 4.95 (4.76, 5.14) <0.001

Hospitalization–
Duration
Logistic Regression
Odds Ratio (95% CI) P-Value Odds Ratio (95% CI) P-Value
Number of infection
1 11.03 (3.54, 34.33) <0.001 11.41 (4.07,32.00) <0.001
2 15.10 (1.61, 141.55) 0.017 24.95 (4.46, 139.71) <0.001
3 370.12 (26.47, 5174.43) <0.001 - -
Male 3.03 (0.92, 10.01) 0.068 1.68 (0.62, 4.50) <0.305

University of Ghana http://ugspace.ug.edu.gh

The Box plot in Figure 1 shows the weight of the children at the trial entry (weight0),
weight at first infection (weight1), weight at second infection (weight2), weight at third
infection (weight3), and the average weight change of all the weights at each infection
(average weight change). The Box plot also shows the range and distribution of the
weights.

Figure 4: Dot Plot of the Effect of Number of Rotavirus Diarrhea Reinfection on
Weight of Ghanaian Children in the Rotavirus Diarrhea Reinfection and Its
Outcome on Weight and Hospitalization – Duration at Study Entry

-5
0
5
10
Weight0 Weight1 Weight2 Weight3 Average Weight Change
Weight Category

Weight
Range
(kg)
University of Ghana http://ugspace.ug.edu.gh

CHAPTER FIVE
5.0 DISCUSSION
5.1 Introduction
The study was a randomized, double-blinded, placebo-controlled trial to investigate
severe rotavirus diarrhea reinfection and its outcome on weight and hospitalization –
duration among 2200 Ghanaian children during the first 106 weeks of life - duration
among children in the Kassena Nankana District of Ghana. The discussion will cover the
following section rotavirus diarrhea reinfection among children in northern Ghana,
rotavirus genotypes that causes severe diarrhea reinfection requiring hospitalizations
among children in northern Ghana and Comparism of outcomes (average weight change
of child in the study and Hospitalization – Duration) in each trial arm.

5.2 Rotavirus Severe Diarrhea Episode and Reinfection
Of the 1098 children in the Vaccine arm of the trial, about eighty – five percent of the
children were protected from any severe diarrhea episode during their first two years of
life. This protection was similar (85%) among the children in the Placebo arm as well.
Out of the remaining fifteen percent of the children who had between one and three
episodes of severe diarrhea, thirteen percent had one episode in the Vaccine and fourteen
percent in the Placebo. This means that the first episode of severe diarrhea was more in
the children who were vaccinated with the placebo. The distribution of two episodes of
University of Ghana http://ugspace.ug.edu.gh

severe diarrhea were similar among the children in both the Vaccine and Placebo arms
with the third infection in the Vaccine arm being three times that of the Placebo.
Hospitalization – duration of the children who had any number of diarrhea infections was
categorized into a binary outcome as an Inpatient or Outpatient. Although, in both arms
the most of the children who an infection were treated as inpatients; More children in the
Placebo arm (56%) require hospitalization for their infections compared to that of the
Vaccine arm (54%).

5.3 Rotavirus G Genotypes
The rotavirus G genotypes that were present in the stools of children who had diarrhea
infection were G1 (40.1%), G2 (46.5%), G3 (11.3%), and G8 (2.1%) in the Vaccine arm
and G1 (31.8%), G2 (45.0%), G3 (13.2%), and G8 (9.9%) in the Placebo arm. In both
arms of the trial the G2 genotypes of the rotavirus caused more infections, followed by
the G1 genotype. In terms of hospitalization – duration and the strains of genotypes in the
G2 of the rotavirus dominated the others in both arms (Vaccine – 51.4% and Placebo
44.6%) for outpatients. In the Vaccine arm most of the hospitalized children were as a
result of the G1 strain (60.0%), whiles G2 (50.0%) again was responsible for
hospitalization in the Placebo arm. The study showed no evidence of different G
genotypes causing any reinfection.

University of Ghana http://ugspace.ug.edu.gh

5.4 Comparism of outcomes (average weight change of child in the study and
Hospitalization – Duration)
The predominance of weight loss for the first diarrhea infection is consist with most
studies on diarrhea dehydration especially Ruuska and Vesikari in 199158 and Velazquez
et al in 199611. There is 2.44 average weight loss in the children at their first infection
compared to about 3.49 weight loss on the second and third infections in both arms. The
children in the age group 7 – 9 weeks old lost 0.33 (P-value=0.007) weight compared to
the children in the age group 4 – 6 weeks which is a significant weight reduction in the
Vaccine arm and in the Placebo arm the children in the age group 7 – 9 weeks old lost
0.08 (P-value=0.50) weight compared to the children in the age group 4 – 6 weeks which
was not significant. About 0.62 (P-value<0.001) reduction in the weight of the children
was recorded in the age group 10 – 12 weeks compared to the age group 4 – 6 weeks in
both Vaccine and Placebo arms which are significant reductions in weight after adjusting
for age. It was also clear in the Box plot that much weight was lost on the second and
third infection compared those who had one infection.
The strong evidence (P-value <0.001) of an association that any number of diarrhea have
with hospitalization – duration of the children means that irrespective of the number of
infections that a child gets they will be a hospitalized. This hospitalization – duration
increases with an increase in the number of diarrhea infections. Though, the number of
infection increases hospital – duration, the hospital – duration for a child vaccinated with
Placebo (OR=24.95) is almost twice that of the child vaccinated with RotaTeq vaccine
(OR=15.10) after adjusting for the age of these children which did not show any
University of Ghana http://ugspace.ug.edu.gh

statistical evidence of increasing hospitalization – duration in the bivariate analysis in
either arm. These findings of hospitalization as a result of diarrhea episodes were earlier
reported by Binka et al 20035 and WHO 200823.

University of Ghana http://ugspace.ug.edu.gh

CHAPTER SIX
6.0 CONCLUSION AND RECOMMENDATION
6.1 Conclusion
This study was a randomized, double-blinded, placebo-controlled trial investigating the
outcome of rotavirus diarrhea reinfection in terms of average weight change and
hospitalization–duration of northern Ghanaian children during the first 106 weeks of life.
The findings of this study would help in the management of rotavirus diarrhea
dehydration and hospitalization of patients.
This study has shown that the risk of getting diarrhea infection is proportional to the age
of the children, therefore if children are vaccinated at an early age, the lesser the risks of
an infection. There was no link for a different genotype responsible for rotavirus diarrhea
reinfections; hence the prevention of the first infection would serve as prevention for
subsequent infections. The study also showed that there is progression in hospitalization
with increase in the number rotavirus severe diarrhea episodes.
These results have shown significant evidence of weight loss and an increase
hospitalization at any severe rotavirus diarrhea infection. Using this new knowledge on
the Vaccine, we are now aware with our own data on weight loss and increase in hospital
admissions as a result of severe rotavirus diarrhea incidents.

University of Ghana http://ugspace.ug.edu.gh

6.2 Recommendation
This study makes the following recommendations:
1. Children should be provided with the Vaccine early in life since severe rotavirus
diarrhea infection and hospitalization – duration varies with the age and the
number of severe diarrhea episodes which is prevented in children vaccinated at
early age.
2. This study suggests further studies designed with enough sample of repeated
episodes of diarrhea. Since this study had fewer cases of repeated episodes of
diarrhea, which is not powered enough.

University of Ghana http://ugspace.ug.edu.gh

REFERENCES
1. Bern C, Martines J, de Zoysa I and Glass RI (1992). The magnitude of the global
problem of diarrhoeal disease: a ten-year update. Bulletin of the World Health
Organization 70, 705–714.
2. Murray CJ and Lopez AD (1997). Alternative projections of mortality and
disability by cause 1990–2020: global burden of disease study. Lancet 349, 1498–
1504.
3. de Zoysa I and Feachem RG (1985). Interventions for the control of diarrhoeal
diseases among young children: rotavirus and cholera immunization. Bulletin of
the World Health Organization 63, 569–583.
4. Estes MK (2001). Rotaviruses and their replication. In: Fields Virology. (eds MK
David, PM Howley, DE Griffin, RA Lamb, MA Martin, B Roizman, SE Straus
and DM Knipe) Lippincott Williams & Wilkins, Philadelphia, pp. 1747–1785.
5. Binka FN, Anto FK, Oduro AR, Awini EA, Nazzar AK, Armah GE, Asmah RH,
Hall AJ, Cutts F, Alexander N, et al.: Incidence and risk factors of paediatric
rotavirus diarrhoea in northern Ghana. Trop Med Int Health 2003, 8(9):840-846.
6. Kapikian AZ, Yasutaka H, Chanock RM. Rotaviruses. In: Knipe DM, Rowley
PM, Griffin DE, Martin MA, Lamb RA, Roizaman B, editors. Fields virology. 4th
ed. Philadelphia: Williams & Wilkins; 2001. p.1787-833.
7. Parashar UD, Alexander JP, Glass RI. Prevention of rotavirus gastroenteritis
among infants and children. Recommendations of the Advisory Committee on
Immunization Practices (ACIP). MMWR Recomm Rep 2006; 55:1–13.
8. Armah, GE, Steele AD, Binka FN, Esona MD and Asmah RH, et al.,
2003. Changing patterns of rotavirus genotypes in Ghana: Emergence of human
rotavirus G9 as a major cause of diarrhea in children. J. Clin. Microbiol., 41:
2317-2322.
9. Reither K, Ignatius R, Weitzel T, Seidu-Korkor A, Anyidoho L, Saad E, Djie-
Maletz A, Ziniel P, Amoo-Sakyi F, Danikuu F, Danour S, Otchwemah, R N,
Schreier E, Bienzle, U, Stark K, Mockenhaupt F P (2007). Acute childhood
diarrhoea in northern Ghana: epidemiological, clinical and microbiological
characteristics. BMC Infect. Dis. 7: 104.
University of Ghana http://ugspace.ug.edu.gh

10. Prince GA, Hemming VG, Horswood RL, Baron PA, Chanock RM. Effectiveness
of topically administered neutralizing antibodies in experimental immunotherapy
of respiratory syncytial virus infection in cotton rats. J Virol. 1987 Jun;
61(6):1851-4.
11. Velázquez FR, Matson DO, Calva JJ, et al. Rotavirus infection in infants as
protection against subsequent infections. N Engl J Med. 1996; 335:1027.
12. Casadei BM, Ciabatti AM, Ignesti C, Livatino L. Incidence of rotavirus in
pediatric patients hospitalized with gastroenteric symptoms. Quad Sclavo Diagn.
1987 Sep; 23(3):246-50.
13. Steele AD (2000). Distribution of rotavirus VP7 and VP4 types circulating across
Africa in 1996 – 1999. Seventh International Symposium on Double-Stranded
RNA Viruses, Aruba.
14. Ramig RF (1997). Genetics of the rotaviruses. Annual Review of Microbiology
51, 225–255.
15. Armah GE, Sow SO, Breiman RF, Dallas MJ, Tapia MD, Feikin DR, et al.:
Efficacy of pentavalent rotavirus vaccine against severe rotavirus gastroenteritis
in infants in developing countries in sub-Saharan Africa: a randomised, double-
blind, placebo-controlled trial. Lancet 2010; 376: 606–14
16. DuPont HL, Practice Parameters Committee of the American College of
Gastroenterology. Guidelines on acute infectious diarrhea in adults. The American
Journal of Gastroenterology. 1997; 92(11):1962–1975.
17. Boschi-Pinto, C., et al., „The Global Burden of Childhood Diarrhoea‟, in: Ehiri,
J.E., M. Meremikwu (editors), International Maternal and Child Health, 2009 (in
press).
18. UNICEF/WHO, Diarrhoea: Why children are still dying and what can be done,
2009 (Accessed 26th January 2012)
19. Boschi-Pinto C, Velebit L, Shibuya K. Estimating child mortality due to diarrhoea
in developing countries. Bull World Health Organ 2008; 86: 710–17.
20. Black RE, Cousens S, Johnson HL, et al. Global, regional, and national causes of
child mortality in 2008: a systematic analysis. Lancet 2010; 375: 1969–87.
21. National Digestive Diseases Information Clearinghouse (NDDIC)/ National
Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)/ National
Institutes of Health (NIH): Diarrhea:
University of Ghana http://ugspace.ug.edu.gh

http://digestive.niddk.nih.gov/ddiseases/pubs/diarrhea/#what (accessed on 12th
January 2012)
22. Ramaswamy K, Jacobson K. Infectious diarrhea in children. Gastroenterology
Clinics of North America. 2001; 30(3):611–624.
23. WHO. Global networks for surveillance of rotavirus gastroenteritis, 2001–2008.
Wkly Epidemiol Rec 2008; 83: 421–425.
24. Parashar UD, Burton A, Lanata C, et al. Global mortality associated with
rotavirus disease among children in 2004. J Infect Dis 2009; 200: S9–15.
25. WHO. Rotavirus vaccines. Wkly Epidemiol Rec 2007; 82: 285–95.
26. Linhares AC and Bresee JS. Rotavirus vaccines and vaccination in Latin America.
Pan Am J Public Health. 2000; 8:305-31.
27. Kapikian AZ, Yasutaka H, Chanock RM. Rotaviruses. In: Knipe DM, Rowley
PM, Griffin DE, Martin MA, Lamb RA, Roizaman B, editors. Fields virology. 4th
ed. Philadelphia: Williams & Wilkins; 2001. p.1787-833.
28. Miller MA, McCann L. Policy analysis of the use of hepatitis B, Haemophilus
influenzae type b-, Streptoccus pneumoniae-conjugate and rotavirus vaccines in
national immunization schedules. Health Econom. 2001; 9:19-35.
29. Protocol: Efficacy, Safety, and Immunogenicity of RotaTeq Among Infants in
Asia and Africa. V260, Protocol 015 – 04 Issue Date: 17 – Dec – 2008.
30. Estes MK, Kapikian AZ (2007). Rotaviruses. In: Knipe DM, Howley PM, Griffin
DE, Lamb RA, Martin MA, Roizman B, Straus SE, editors. Fields virology. Vol.
2, 5th edition. Philadelphia: Kluwer Health/Lippincott, Williams and Wilkins, pp
1917–1974.
31. Santos N and Hoshino Y (2005). Global distribution of rotavirus
serotypes/genotypes and its implication for the development and implementation
of an effective rotavirus vaccine. Rev Med Virol 15:29–56.
32. Matthijnssens J, Ciarlet M, Rahman M, Attoui H, Ba´nyai K, Estes MK, ….., Van
Ranst M (2008). Recommendations for the classification of group A rotaviruses
using all 11 genomic RNA segments. Arch Virol 153:1621–1629.
33. Rao CD, Gowda K, Reddy BS (2000). Sequence analysis of VP4 and VP7 genes
of nontypeable strains identifies a new pair of outer capsid proteins representing
novel P and G genotypes in bovine rotaviruses. Virology 276:104–113.
University of Ghana http://ugspace.ug.edu.gh

34. Rahman M, Matthijnssens J, Nahar S, Podder G, Sack DA, Azim T, Van Ranst M
(2005). Characterization of a novel P[25], G11 group A rotavirus. J Clin
Microbiol 43:3208–3212.
35. Martella V, Ciarlet M, Banyai K, Lorusso E, Cavalli A, Corrente M, …..,
Buonavoglia C (2006). Identification of a novel VP4 genotype carried by a
serotype G5 porcine rotavirus strain. Virology 346:301–311.
36. Khamrin P, Maneekarn N, Peerakome S, Chan-it W, Yagyu F, Okitsu S, Ushijima
H (2007). Novel porcine rotavirus of genotype P[27] shares new phylogenetic
lineage with G2 porcine rotavirus strain. Virology 361:243–252.
37. Abe M, Ito N, Morikawa S, Takasu M, Murase T, Kawashima T, Kawai Y,
Kohara J, Sugiyama M (2009). Molecular epidemiology of rotaviruses among
healthy calves in Japan: Isolation of a novel bovine rotavirus bearing new P and G
genotypes. Virus Res 144:250–257.
38. Greenberg HB and Estes MK (2009). Rotaviruses: From pathogenesis to
vaccination. Gastroenterology 136:1939–1951.
39. Solberg OD, Hasing ME, Trueba G, Eisenberg JN (2009). Characterization of
novel VP7, VP4, and VP6 genotypes of a previously untypeable group A
rotavirus. Virology 385:58–67.
40. Ursu K, Kisfali P, Rigo D, Ivanics E, Erdelyi K, Dan A, Melegh B, Martella V,
Ba´nyai K (2009). Molecular analysis of the VP7 gene of pheasant rotaviruses
identifies a new genotype, designated G23. Arch Virol 154:1365–1369.
41. Ciarlet M, Liprandi F, Conner ME, Estes MK (2000). Species specificity and
interspecies relatedness of NSP4 genetic groups by comparative NSP4 sequence
analyses of animal rotaviruses. Arch Virol 145:371–383.
42. Mori Y, Borgan MA, Ito N, Sugiyama M, Minamoto N (2002). Sequential
analysis of nonstructural protein NSP4s derived from Group A avian rotaviruses.
Virus Res 89:145–151.
43. Salinas B, Pe´rez-Schael I, Linhares AC, Ruiz Palacios GM, Guerrero ML,
Yarza´bal JP, et al. (2005). Evaluation of safety, immunogenicity and efficacy of
an attenuated rotavirus vaccine, RIX4414: A randomized, placebo-controlled trial
in Latin American infants. Pediatr Infect Dis J 24:807–816.
44. Ruiz-Palacios GM, Pe´rez-Schael I, Vela´zquez FR, Abate H, Breuer T, Clemens
SC, et al. Human Rotavirus Vaccine Study Group (2006). Safety and efficacy of
University of Ghana http://ugspace.ug.edu.gh

an attenuated vaccine against severe rotavirus gastroenteritis. N Engl J Med 354:
11–22.
45. Vesikari T, Matson DO, Dennehy P, Van Damme P, Santosham M, Rodriguez Z,
et al. Rotavirus Efficacy and Safety Trial (REST) Study Team (2006). Safety and
efficacy of a pentavalent humanbovine (WC3) reassortant rotavirus vaccine. N
Engl J Med 354: 23–33.
46. Richardson V, Hernadez-Pichardo J, Quintanar-Solares M, Esparza-Aguilar M,
Johnson B, Gomez-Altamirano CM, Parashar U, Patel R (2010). Effect of
rotavirus vaccination on death from childhood diarrhea in Mexico. N Engl J Med
362:300–305.
47. Pe´rez-Schael I, Salinas B, Gonza´lez R, Salas H, Ludert JE, Escalona M, Alcala´
A, Rosas MA, Matera´n M. 2007. Rotavirus mortality confirmed by etiologic
identification in Venezuelan children with diarrhea. Pediatr Infect Dis J 26:393–
397.
48. Salinas B, Gonza´lez G, Gonza´lez R, Escalona M, Matera´n M, Pe´rez-Schael I
(2004). Epidemiologic and clinical characteristics of rotavirus disease during five
years of surveillance in Venezuela. Pediatr Infect Dis J 23:S161–S167.
49. Parashar UD, Alexander JP, Glass RI. Prevention of rotavirus gastroenteritis
among infants and children. Recommendations of the Advisory Committee on
Immunization Practices (ACIP). MMWR Recomm Rep 2006; 55:1–13.
50. Parashar UD, Burton A, Lanata C, Boschi-Pinto C, Shibuya K, Steele D, et al.
Glass Global Mortality Associated with Rotavirus Disease among Children in
2004J Infect Dis. (2009) 200(Supplement 1): S9-S15 doi:10.1086/605025.
51. Linhares AC, Gabbay YB, Freitas RB, Travassos da Rosa ES, Mascarenhas JDP,
Loureiro ECB. Longitudinal study of rotavirus infections among children from
Belém, Brazil. Epidemiol Infect. 1989; 102:129-45.
52. Wyatt RG, Yolken RH, Urrutia JJ, Mata L, Greenberg HB, Chanock RM, et al.
Diarrhea associated with rotavirus in rural Guatemala: a longitudinal study of 24
infants and young children. Am J Trop Med Hyg. 1979; 28(2):325-8.
53. Bishop RF, Barnes GL, Cipriani E, Lund JS. Clinical Immunity after neonatal
rotavirus infection. N Engl J Med. 1983; 309:72-6.
University of Ghana http://ugspace.ug.edu.gh

54. Mata L, Simhon A, Urrutia JJ, Kronmal RA, Fernández R, García B.
Epidemiology of rotavirus in cohort of 45 Guatemalan Mayan Indian children
observed from birth to the age of three years. J Infect Dis. 1983; 148:452-61.
55. Flores J, Pérez-Schael I, González M, Garcia D, Perez M, Daoud N, ….., Cunto
W, Protection against severe rotavirus diarrhoea by rhesus rotavirus vaccine in
Venezuelan infants. Lancet. 1987; 1:882-4.
56. O'ryan MO, Matson DO, Estes MK, Bartlett AV, Pickering LK. Molecular
epidemiology of rotavirus in children attending day care centers in Houston. J
Infect Dis. 1990; 162:810-6.
57. Moulton LH, Staat MA, Santosham M, Ward RL. The protective effectiveness of
natural rotavirus infection in an American Indian population. J Infect Dis. 1998;
178:1562-6.
58. Ruuska T, Vesikari T. A prospective study of acute diarrhoea in Finnish children
from birth to 2½ years of age. Acta Paediatr Scand. 1991; 80:500-7.
59. Binka FN, Ngom P, Phillips JF, Adazu K and MacLeod BB (1999). Assessing
population dynamics in a rural African society: the Navrongo Demographic
Surveillance System. Journal of Biosocial Sciences 31, 375–391.

University of Ghana http://ugspace.ug.edu.gh