Monica Sok Honorsthesis

ABSTRACT
The Effect of Age and Gender on the Prevalence and Severity of Hypertension in
Rural Western Kenya
Monica Sok
Director: Lisa Baker, MD PhD
In order to effectively lessen the drastically increasing morbidity of

hypertension in sub-Saharan Africa, new prevalences must be calculated and the
causes and aggravating factors of this disease must be more completely
understood in this region. The literature repeatedly states how prevalence tends to
be underestimated in this area of the world due to lack of blood pressure
measurements and the highlighted focus on infectious diseases. Many studies
have presented different, solid evidence on potential causes behind hypertension
including the common ones of BMI, age, diet, and level of physical activity. The
usual sample sizes have focused on older groups. The literature lacks concrete
data concerning the nature of high blood pressure in younger groups. This cross-
sectional study analyzes previously collected clinical data from a sample
population of 685 patients who attended a clinic in May 2010 in rural western
Kenya within the Nyando District. Data trends are inferred from blood pressure
measurements, anthropometric measurements, blood samples, and the patient’s
self-report of symptoms. After creating the cut-off age of 18 years, the study’s
sample size reduced to 321 patients. For those between the ages of 18 and 44, the
prevalence of high systolic blood pressure is 19.42%. For those equal to or over
the age of 45, the prevalence of high systolic blood pressure is 53.50%. The
prevalence of high blood pressure in males is 43.59% and the prevalence in
females is 35.48%. BMI is found to be a statistically significant predictor of the
severity and presence of high blood pressure only in the young group and the
female group. These results show that hypertension is a silent, severe,
proliferating problem in the rural area. The pathophysiology behind high blood
pressure is typical in the female group, but there is clearly a different
physiological process creating hypertension in those who are young and those
who are thin.
APPROVED BY DIRECTOR OF HONORS THESIS:
_____________________________________________________
Dr. Lisa Baker, Department of Biology
APPROVED BY THE HONORS PROGRAM:
______________________________________________
Dr. Andrew Wisely, Director
DATE:
THE EFFECT OF AGE AND GENDER ON THE PREVALENCE AND SEVERITY
OF HYPERTENSION IN RURAL WESTERN KENYA
A Thesis Submitted to the Faculty of

Baylor University
In Partial Fulfillment of the Requirements for the
Honors Program
By
Monica Sok
Waco, Texas
May 2012
TABLE OF CONTENTS
List of Figures, Tables, and Graphs iv

Acknowledgments vi
Dedication vii
Chapter One: Introduction 1
Chapter Two: Review of Literature 8
Chapter Three: Hypothesis 26
Chapter Four: Methods 31
Study Area and Population 31
Measurements 32
Study Design and Clinical Data 33
Analysis 36
IRB 38
Chapter Five: Results 39
Organization of Results 43
General Overview 43
Overall Data without Stratification 44
Data by Age Groups 48
Data by Gender 54
Data by Age Group and Gender 60
Creatinine 64
Data Results as Predicted by Hypotheses 68
Chapter Six: Discussion and Conclusions 72
Overview 72
Primary Hypertension 74
BMI vs. WHR 75
Curvilinear Relationship of Blood Pressure 77
Secondary Hypertension 78
Systematic Threats to Internal Validity 80
Random Threats to Internal Validity 80
Construct Validity 81
Generalization across Persons, Settings, and Time 82
Conclusion 82
Appendices 84
References 90
iii
LIST OF FIGURES, TABLES, and GRAPHS
FIGURES
III.1 Hypothesis Diagram 28
III.2 In-Depth Schematic Representation of the Old 29
III.3 In-Depth Schematic Representation of the Young 30
TABLES
IV.1 BMI WHO Severe Classification 34
IV.2 WHR Quintile Cut-Offs 34
V.1 Blood Pressure by Age Group 40
V.2 Blood Pressure by Gender 40
V.3 High Systolic Blood Pressure Prevalence 41
V.4 BMI Categorical Percentages by Geographical Area 42
V.5 High Blood Pressure Prevalence: Comparison of Studies 44
V.6 The Impact of Age, BMI, and Gender on Blood Pressure 44
V.7 The Impact of Hip Circumference on Blood Pressure 45
V.8 The Impact of Hip Circumference (Obese) on BP 46
V.9 Analysis of Variance Unstratified 47
V.10 Descriptive Averages by Age 48
V.11 The Association of Hypertension and BMI in Young 50
V.12 The Association of Hypertension and BMI in Old 51
V.13 Impact of BMI on BP Stratified by Age Group 52
V.14 Impact of Waist and Hip Circumferences on BP by Age Group 52
V.15 Impact of WHR on BP by Age Group 52
V.16 Analysis of Variance Stratified by Age Group 53
iv
V.17 Descriptive Statistics by Gender 54
V.18 The Association of Hypertension and BMI in Male 56
V.19 The Association of Hypertension and BMI in Female 57
V.20 The Impact of BMI and Age on Blood Pressure by Gender 58
V.21 Analysis of Variance Stratified by Gender 59
V.22 Descriptive Averages by Age-Gender Groups 60
V.23 The Impact of BMI on Blood Pressure by Age-Gender 61
V.24 The Impact of Waist and Hip Circumference by Age-Gender 62
V.25 Analysis of Variance Stratified by Age-Gender 63
V.26 Descriptive Statistics of Creatinine by Age-Gender 64
V.27 Abnormal Creatinine Descriptive Statistics by Age-Gender 65
GRAPHS
IV.1 US Creatinine Values 35
V.1 Comparison of BMI by Regions of the World 42
V.2 High Blood Pressure Prevalence: Comparison of Studies 43
V.3 By Age Group: Average Blood Pressure vs. BMI WHO 49
V.4 By Gender: Average Blood Pressure vs. BMI WHO 55
V.5 US Creatinine Values 64
V.6 Range of Blood Pressure 66
V.7 Range of BMI 67
v
ACKNOWLEDGEMENTS
I would like to thank Dr. Lisa Baker for being my thesis director and for her
constant, never failing support, strength, inspiration, and advice. I would also like to
thank Dr. Troy Abell for his valuable, cherished help with data analysis and for his never
failing support, inspiration, and guidance. I am also grateful to Pastor Habil Ogolla for
his lessons, his faith, his leadership, his goodness, and his courage.
vi
DEDICATION
This study is dedicated to my parents, my brother, and my friends who have never
stopped believing in me. This study is also dedicated to those living on the Nyakach
Plateau, my wonderful friends in Kenya, and to all those who suffer from non-
communicable diseases, especially hypertension
vii
CHAPTER ONE
Introduction
Hypertension: The Current World Status
If this word, ‘hypertension’, is mentioned to anyone in suburban America, the
person spoken to would nod and most likely shrug, minimizing it as another common
lifestyle problem prevalent in the states. The same reaction would probably be given if
the word ‘malaria’ were to come up in a conversation about the current health problems
in developing countries. There would be no surprise – just a look of sympathy.
Unfortunately, communicable diseases are not the only pathological processes
rising in numbers in developing countries. With the rise of urbanization and an increased
interest in health research in the developing countries, a new issue has come into light.
There is a new threat on the horizon of developing areas, quickly spreading, and that
threat is non-communicable diseases (NCD).
Cardiovascular diseases are spreading rampant throughout Africa. With
urbanization going on and different, more sedentary lifestyles being adapted, it does not
come as a surprise to see high numbers of cardiovascular diseases in the cities. In most
African cities now, a physician will find alarming rates of hypertension in patients. Even
in the rural areas, there have been surprising findings of high proportions of
cardiovascular diseases. It is this silent and undetected hypertension that is leading to
heart attacks and strokes, and recently, this is the issue making the headlines and
dominating current research.
In 2000, almost one billion people in the world were counted as having
hypertension. By the year of 2025, this number is expected to increase by 60% 3-5.
Centering in on sub-Saharan Africa, studies have shown that there were 75 million
hypertensives in the area in 2008. By 2025, the number will shoot up to 125.5 million
hypertensives. Along with this 2008 data, the numbers also show an overall prevalence of
20.7% of the urban population having hypertension and 13.7% of the rural population
having hypertension 73.
Although diseases such as malaria and HIV/AIDS are still very important to
consider and study in Africa, scientists, researchers, and physicians should not neglect the
potential negative effects non-communicable diseases could have upon the public.
The Pathophysiology
Hypertension is considered a serious problem because of the many complications
it could possibly lead to if not diagnosed, monitored, and treated. One of the many
serious complications hypertension could lead to is a cerebral hemorrhage or stroke.
Untreated hypertension can eventually lead to high blood pressures rupturing a cerebral
artery, thus bringing damage to the brain 15, 34.
2
There are many different types of hypertension. The most common one is called
primary or essential hypertension. Primary hypertension can arise in two ways. If there is
excessive vasoconstriction of the small arterioles, there can be high diastolic blood
pressure. High systolic blood pressure can come as an effect of high peripheral resistance;
to compensate for the resistance, the heart must pump harder. Because of the extra effort
the heart has to make and because of the added pressure on the vessels, there are many
effects on the heart, blood vessels, and kidneys 15, 34.
Left ventricular hypertrophy can result from the heart working so hard. The
muscles in the heart become enlarged. This does not cause immediate heart problems, but
eventually the heart can descend into cardiac failure or congestive heart failure. Many of
the symptoms include bloating, irregular pulse, fatigue, and swelling in the feet, ankles,
and abdomen. Besides left ventricular hypertrophy, a patient could also develop
hypertrophic cardiomyopathy, a similar condition. Another cardiac side-effect would be
ischemic heart disease. This is when a heart attack occurs because the heart is not getting
enough blood, and thus, oxygen. Symptoms involved include fatigue, irregular pulse,
dizziness, and chest pain 15, 34.
Besides heart problems, essential hypertension can lead to mild to severe vascular
effects. A person can develop arteriosclerosis – the high, continuous pressure leads to
arteriole injury. The injuries lead to degenerative changes and the lumen becomes
15,
narrowed. If the wall has reached a breaking point, it can turn necrotic and even burst
34
.
3
Lastly, essential hypertension has been known to cause renal effects. One huge
problem with hypertension is renal failure. The renal arterioles can narrow, thus
restricting blood flow to the kidneys which could lead to the glomeruli and renal tubules
dying 15, 34.
Although primary hypertension is the most common, there are other types of
hypertension as well. Conditions that can cause hypertension as a secondary effect
include: endocrine gland dysfunction, chronic kidney disease, sleep apnea, coarctation of
aorta, and preelampsia 15, 34.
Isolated systolic hypertension involves normal diastolic blood pressure while the
systolic blood pressure is slightly or highly elevated. This is a hypertension primarily
found in older people. With increasing age, the elasticity of the arteries decreases and can
no longer absorb the force the ventricles exert. Because there is no excessive
vasoconstriction, the diastolic pressure remains normal 15, 34.
One last type of hypertension is hepatic portal hypertension. One of the main
causes of portal hypertension is cirrhosis of the liver which causes scar tissue to block
portal veins. With the combination of the vein blockage and high pressure of the blood
coming through, there is usually leakage of fluid from the capillaries collecting in the
abdominal cavity and causing ascites. To compensate for the vein blockages, the body
will automatically create shunts or anastomoses, new pathways for the blood to flow 15, 34.
Other conditions that have to do with hypertension include preeclampsia and
metabolic syndrome. Preeclampsia results from high blood pressure during pregnancy
and can lead to seizures. Metabolic syndrome refers to a combination of hypertension,
4
insulin resistance, and hyperlipidemia. There are currently two standards for metabolic
syndrome. Based on the 2001 National Cholesterol Education Program Adult Treatment
Panel, some risk factors for this syndrome include abdominal obesity (waist
circumference >102 cm in men and >88 cm in women) and a blood pressure of +130/85
mmHg. WHO defines abdominal obesity by three measurements: a waist to hip ratio
greater than 0.9, a BMI of at least 30 kg/m2, and a waist measurement of over 37 inches.
Two other WHO classifications for metabolic syndrome include a cholesterol panel of at
least 150mg/dl and a blood pressure of +140/90 mmHg 15, 34.
The following list shows the typical risk factors for hypertension:
 Black
 Family history of hypertension, heart disease, or diabetes
 Over the age of 55
 Overweight
 Not physically active
 Drink excessively
 Smoke
 Diet high in saturated fats or salt
 Illicit drug usage
 NSAIDS or decongestant drugs
The problem with hypertension exists in the fact that it is usually a silent disease.
Hypertension does not usually alert the human with some textbook symptoms, but if
hypertension is severe or malignant enough, possible symptoms will be produced:
5
 Pounding in chest, neck, or ears
 Irregular heartbeat
 Blood in urine
 Difficulty breathing
 Chest pain
 Vision problems
 Fatigue or confusion
 Severe headache
Nyakach Plateau Clinic Observations
In the clinic conducted on the top of the Nyakach Plateau, a physician found
many cases of hypertension. The number noticed was alarming considering that the area
was a strictly rural, not urban, area like Nairobi where hypertension is starting to run
rampant.
It was also noted that there were all sorts of body types coming in with cases of
high blood pressure. The situation presented then did not lie solely with the obese and did
not follow the typical case of metabolic syndrome. There were many patients with high
blood pressure who were young and many who were thin. These alarming findings
opened many questions on the factors causing hypertension in the area. It was obvious
from just observing that typical hypertension, as similar to that occurring in America, was
not all that was happening in this area.
6
The high blood pressure mystery combined with the fact that most hypertension
goes untreated in this area raises red flags and many questions. This is a puzzle waiting to
be put together, but the pieces must be found first. Hypertension is a polygenic disease,
and the aggravating factors behind it must be discovered. The rise of hypertension in
developing countries is turning into a crisis.
Thus, this research investigates the possibility of finding a strange relationship
between the young people and hypertension, a relationship not commonly found before in
the rural area. This community-based research will hopefully contribute another piece of
the puzzle to understanding rural health in Kenya. A better understanding could lead
toward more effective interventions among this particular group of people, but it may
also add to the general scientific knowledge of this devastating disease.
7
CHAPTER TWO
Review of Literature
The Global Significance
When people hear underdeveloped or developing nations, many think of
infectious diseases as the foremost health problem. However, the truth is that although
infectious diseases remain one of the top problems and killers, non-communicable
diseases are starting to become an epidemic. Recent statistics show that
infectious/parasitic diseases caused 5.5 million deaths in Africa in 2005. In the same year,
non-communicable diseases (NCDs) caused over 2.4 million deaths. Non-communicable
diseases are gradually climbing to the forefront in developing nations’ health problems 52.
It has been predicted that by the year 2020, non-communicable diseases will be the
leading cause of mortality and morbidity 61. Already, the mortality rate from infectious
diseases has begun to decline by 3% while the mortality rate from chronic diseases has
increased by 17% 1. More specifically, in 2000, a quarter of the world’s adult population
had cases of hypertension. By the year of 2025, the rate of hypertension is predicted to
increase by 60% 3,4.
8
The Importance and Significance of Anthropometric Measurements
Particular anthropometric measurements indicate different things and can be more
or less important depending on what disease is involved and what group is being studied.
The top two standards of body fat used in the hypertension literature are body mass index
(BMI) and waist-to-hip ratio (WHR).
The WHR is an extremely important variable in hypertension. According to
Njelekela, with every 0.1 unit increase in WHR, there is a 70% increase in the odds of
hypertension. Weight also has a direct correlation. A person can have an increase of 3.0
mmHg in systolic blood pressure for every 10 kilogram increase in weight 9. A study by
Fezeu stated that waist circumference (WC) is a better indicator of cardiovascular risk
since it truly measures the adipose tissue than BMI or WHR does 17,22, 23. This finding
was further reinforced by a Nigerian study by Oldapo. Oladapo found that abdominal
obesity was a better CVD risk factor to measure than overall obesity. A person was found
to have higher risk for cardiovascular disease if a waist circumference was above 88cm in
women and above 102cm in men and his or her blood pressure was above 130/85 58.
The Gender Differences
Many studies compare men versus women and how the rates of hypertension
differ between the two genders. Hypertension in females is more likely to be detected
than in males because of women’s higher rate of hospital admission due to pregnancy.
While in the hospital, virtually all women have their blood pressures checked 2, 35.
9
Although there is a greater amount of in-hospital blood pressure data on women, outside
studies have found a trend of men having higher rates of hypertension than women.
According to a study by de Ramirez, men had an overall higher hypertension prevalence
of 24% compared to the women’s 20%. Ramirez also found that the gender trends only
related to systolic blood pressures (SBP) 61.
The trend of males having higher hypertension rates than females is confirmed by
other studies and further explained. In Njelekela’s study, it was found that women had
higher obesity rates (35% versus 13%) and a higher prevalence of abdominal obesity
(58% versus 11%) compared to men. These findings would seem to point to women
having more cardiovascular problems, but instead, Njelekela reinforced the finding of
females having a lower rate of hypertension than males. In this particular study, females,
compared to men, were found to have a 50% lower odds of having hypertension. Women
had the higher waist-to-hip ratio (WHR), BMI, waist circumference, and hip
circumference. Despite these being higher than men, women still had lower systolic and
diastolic blood pressures with all data being statistically significant. Despite the high
BMI and other high anthropometric measurements, women still had lower blood
pressures than men due to potential protective factors. One explanation could be that the
body fat in women is used in different ways by the body, such as for lactation, than the
body fat in men, resulting in extra fat not leaving harmful effects. Because of these
protective measures, women may need to reach a higher level of body fat than men
require to become hypertensive 19,24, 75.
One of the main reasons why women have a lower blood pressure than men,
however, could be because of the protective effect estrogen has against increasing blood
10
pressures. Many studies have demonstrated that before puberty, boys and girls have
similar values of systolic and diastolic blood pressures. Upon the onset of puberty, boys
start to have higher blood pressure values than girls, and only at a later age do women
surpass men in having higher blood pressure measurements. Studies like one done by
Boschitsch also looked into the role of progesterone and aldosterone in hypertension in
women. Although Boschitsch mainly looked at studies of USA, Canada, and Europe, the
physiological process found can most likely be applied across all women of all
ethnicities. Usually, angiotensin II, involved in the renin-angiotensin-aldosterone system,
raises blood pressure. Angiotensin II leads to the release of aldosterone which also causes
the cardiovascular system to retain salt and, in so doing, to retain water and increase
blood pressure. However, progesterone, a female hormone, has been found to have an
anti-aldosterone effect, and thus has the potential to lower blood pressure 10. Fisher also
found a sexual dimorphism in essential hypertension. As other studies have shown, only
upon a later age of around 50, during the time of menopause, do women start to show
higher blood pressures than men. This is due to the decrease at that time of the beneficial
and protective effect of estrogen. Before menopause and after puberty, women have been
shown to have lower peripheral resistance than men, and thus lower blood pressures. The
physiology behind this involves the menstrual cycle and fluctuating hormones. In
women, during their peak estrogen phases of the menstrual cycle (the follicular stage),
there happens to be a complementary release of nitric oxide, a vasodilator. Fisher also
noted that men have a higher concentration of endothelin which is a vasoconstrictor 24.
Furthermore, Hallberg studied gender specific associations with the vasoconstrictor
endothelin – a contributing vascular factor to hypertension. Hallberg found that in
11
females, the ovarian hormones can suppress endothelin production. In lab studies, it has
been found that testosterone, unlike estrogen, increases endothelin levels in the blood 30.
In another hypertension study, Hayes’s article supports all these findings by
demonstrating how the trend of women’s systolic blood pressure starts from being lower
than men’s during early adulthood but after the age of 60, women’s blood pressure
becomes as high or higher than men’s. The estrogen protective advantage only works
during pre-menopause 31. A study by Reyes shows the same blood pressure relationships
between males and females. Reyes re-emphasizes how estrogen depresses blood pressure
by releasing vasoactive substances and testosterone tends to hold onto sodium which
increases volume retention, and thus higher blood pressure 63.
Although most studies have displayed the relationship of men having higher rates
of hypertension than women, there have been some contradictory studies suggesting that
women have higher rates of hypertension than men. One of these studies is Muraguri’s
study among teenage secondary school students in Nairobi, Kenya. In the study, it was
found that girls from age 13-15 had higher systolic blood pressure than boys, but the
findings were not statistically significant. The second half of the study focusing on boys
aged 15-18 years had statistically significant higher systolic blood pressures than girls. In
the end, this follows the previously studied trends of pre-pubertal girls having equal or
higher rates of hypertension than boys, and girls having reached menarche having lower
blood pressures than boys. The same study has found that there are positive relationships
between weight, height, and BMI with systolic blood pressure for boys, but for girls,
positive relationships only exist between systolic blood pressure and weight and BMI, not
height 55.
12
The Age Group Differences
It is not an unknown fact that with older age comes a higher chance of developing
non-communicable diseases such as hypertension. The same trend occurs in developing
nations, and most research studying non-communicable diseases or cardiovascular risks
notes this trend 15, 61. According to Ramirez’s study, unlike how gender was only related
to SBP, age was found to be correlated with systolic blood pressure (SBP), diastolic
blood pressure (DBP), and mean arterial blood pressure (MABP) 15. In one study, Fezeu
showed that the group with the most chance of developing hypertension is the group age
55-74 years 17,22.
There are a great number of studies in the literature examining the relationship of
aging to high blood pressure among adults, but there are few studies looking into the
relationship between high blood pressure and age among children or adolescents. The
study done in the Ashanti region of Ghana, West Africa is one of the few studies looking
into blood pressure patterns among children. The researchers found similar urban versus
rural trends that have been repeatedly noticed in adults: blood pressure measurements are
usually higher in urban than rural boys. For girls in this study, however, there was no
difference in blood pressure noted between the rural, semi-urban, and urban areas. While
this is an interesting find, females have also been noted to have a more fluctuating
relationship between age and blood pressure due to puberty and the onset of hormones as
explained in the above section of hypertension and gender 4. Other studies done by
13
Amusa in the Tshannda Longitudinal Study showed the positive relationship between
blood pressure, BMI and body fat for children. In this study of school children, they
found that boys and girls had similar diastolic blood pressure and systolic blood pressure
values until after grade level 5. Afterwards, the girls showed higher measurements of
blood pressure than the boys. This is an interesting finding and seems to contradict the
findings made by Fisher who showed the protective effect of estrogen against high blood
pressures. It must be taken into account, however that this study focused on overweight
children which could negate the protective effect of estrogen 7, 24.
Another important cardiovascular study done on children was by Fourie who
looked into how increased cardiovascular reactivity in children could lead to later
development of cardiovascular disease. The pathophysiologied explanation given is that
“a sympathetic overreaction such as elevated B-adrenergic activation pattern in young
people over a long term leads to changes in the morphologic structure of blood vessels
and to chronic elevated blood pressure” 26.
The Problems in Africa: Potential Causes of Hypertension and other Non-Communicable

Diseases
(1) Rural to Urban
Non-communicable diseases are becoming a problem in developing countries
presumably due to increasing urbanization. Western lifestyles are being adopted, which
14
has precipitated a shift in physical activity and diets 56. The urbanization leads to
sedentary lifestyles, obesity, unhealthy habits and diets. There are reported rates of non-
communicable diseases increasing all over the world. In Tanzania, there have been
reports of increasing hypertension rates in both urban and rural areas. Furthermore, an
increase in the incidence of diabetes mellitus (DM) has also been reported in urban
locations. Most notably, DM is becoming an increasing problem in Uganda and South
Africa where the prevalence rates are between 8% and 13%. Hyperlipidemia is also
becoming common with women and those over the age of thirty-five 57.
Before urbanization, research reports in Kenya showed that there was a low
incidence of hypertension in rural areas 35. Urbanization and its effects, however, are
becoming a major topic of study because of the proliferation of urban living – 40% of
Africans currently live in urban areas, and by 2030, half of the Africans will live in urban
areas 47. Thus non-communicable diseases have been appearing in both urban and rural
areas of developing countries. Yet they are more pronounced in urban areas, particularly
hypertension. According to Agyemang, the age-adjusted systolic and diastolic blood
pressure levels were revealed as lower in rural men and women than urban men and
women 3-5, 47. A study by Fezeu found that with increasing waist circumference, there was
a greater increase in diastolic blood pressure in urban areas compared to rural areas.
Urban and rural areas of developing countries generally consist of people with the same
genetic make-up. What could contribute to the urban area’s higher rates of cardiovascular
risk factors are the environmental differences – these include diet and physical activity
levels 2, 19, 22, 24. In urban areas, the jobs people have are less physical than the duties
required by rural life. The foods people eat in those areas are more “dense” than what is
15
consumed in rural areas. A combination of all these factors usually leads to the people in
urban areas being more overweight or obese than those in the rural area. Kenya actually
has one of the worst rates of obesity in SSA. In the study by Ziraba, an interesting trend
was found. Instead of an increase in obesity in the wealthier women, it was noted that
obesity increased in women of lower socio-economic class. This striking finding
emphasizes an important lesson: although higher rates of non-communicable diseases are
currently being found in the urban areas more than the rural areas, with time, the rates of
non-communicable diseases will proliferate in the non-wealthy and non-urban areas 79.
Most articles in the literature have thoroughly documented the increasing trend of
non-communicable diseases like hypertension in urban areas of developing nations. What
is lacking in the literature is an explanation of the rise of hypertension in non-urban
settings. A most recent statistic notes 2.3% to 41.1% prevalence of hypertension in SSA
17
. A better understanding is especially urgent because studies have noted that
cardiovascular diseases occur earlier among sub-Saharan African populations than any
other populations 51-53.
Comparing the urban and rural populations, Agyemang reiterated the common
knowledge of the direct relationship between age and blood pressure. Despite the
environment or social circumstances a person lives in, whether it be urban or rural, both
systolic and diastolic blood pressure generally increase with age 3-5.
16
(2) Genetics
Hypertension has been defined in genetic terms as a polygenic disease, one that
results from the environment and the interaction with several genes. The variety of
factors playing into hypertension is the reason why hypertension in one person can be
vastly different from the hypertension in another person. Because 40% of high blood
pressure can be explained genetically, there have been myriad recent studies focusing on
the genes causing high blood pressure 14. Going further into the breakdown, 20-60% of
hypertension can be explained by genetics while 0-16% can be explained by dietary
factors 64. According to Crook, one of the strongest linkages to hypertension could arise
from the angiotensinogen gene 14. Another study by Ehret looked at 200,000 individuals
of European descent and found 16 important loci related to the regulation of blood
pressure. Ehret also set forth “strong genetic risk scores” for systolic blood pressure in
individuals of African background 17.
Although many studies have been done on the genetics behind hypertension, there
have been even more studies done on the environmental factors behind hypertension. In
Africa especially, many studies have been focused on seeing repeated patterns of
increasing hypertension in urban areas as compared to rural areas. Consequently, the
conclusion has been that environmental circumstances are more closely linked to
hypertension than to genetics 9,11.
A study by Seedat has found that there are potential differences in the
pathogenesis behind hypertension in whites and blacks. Upon further investigation into
the blacks of Sub-Saharan Africa (SSA), it has been noted that there are significant
17
biochemical differences in their bodies versus other races. Blacks seem to have lower
levels of low density lipoprotein (LDL) and higher levels of high density lipoprotein
(HDL) 78. HDLs are considered the “good cholesterol” while LDLs are considered to be
the “bad cholesterol”. Furthermore, blacks are more “salt-sensitive”, while whites are
more “salt-resistant”. Salt sensitivity refers to the physiological reaction of “renal
vascular resistance” and “glomerular pressure” during high sodium intake. Because of
these factors, black patients with hypertension are also sometimes subject to renal failure
78
. To make this case stronger, Tiffin’s study looked into the genes involved in salt
sensitivity in indigenous South Africans and found that the salt sensitivity leading to
hypertension is more prevalent in people of African origin 73.
(3) Enzymes and Inflammation
Some studies have found connections between certain enzymes and high blood
pressure. According to Brewster’s study, there could be a potential linkage between
increased levels of creatine kinase and higher chances of developing hypertension.
Creatine kinase is present in vascular and cardiac muscle tissue. Its function is to help the
cell under high stress or conditions of high demands by increasing salt retention in the
renal tubules 11.
In another article, Kuklinska showed that hypertension patients had higher CRP
(C-reactive protein) concentrations. CRP is a marker of inflammation, and recently it has
been used as a marker for cardiovascular activity 38. Studies have shown that perhaps
CRP is a measure of arterial stiffness. Recent studies have shown that chronic arterial
18
wall inflammation leads to endothelial dysfunction and thus, a higher chance of
developing hypertension. Along with CRP, HTN patients also have high levels of BNP,
interleukin-6, and tumor necrosis factors 38.
(4) Malnutrition and Low Birth Weight
Recently, there has been a fair amount of literature investigating the relationship
between birth size and weight and later incidence of hypertension. Omolola addressed the
potential mechanism behind this relationship. It could be that low birth weight is a result
of poor maternal nutrition and maternal iron deficiency anemia, and thus the baby is born
with reduced vascular elasticity, which could lead to later on adult hypertension. In
Omolola’s study, low birth weight and later life hypertension was a result of parasitaemia
(malaria) during pregnancy and not parasitaemia at delivery. Thus, it has been
demonstrated that a low birth weight could lead to later cases of hypertension in life 8.
Woelk also investigated this issue and found an inverse relationship in England between
systolic and diastolic blood pressures and birth weight. Although Woelk found these
relationships in England, the study notes that this relationship can be applied to Africa
due to the country’s major problem of undernutrition 77. Furthermore, stunting, a
potential result of malnutrition, has been associated with high blood pressure. One study
investigated the cardiovascular changes in stunted children. It found that vascular
compliance (needed to maintain lower blood pressure) was lower in stunted children.
With stunting being one of the most common nutritional problems in Africa, this is a
huge concern since these early physiological changes can lead to later non-communicable
19
diseases. Although this study’s findings reinforced previous studies, it found in the end,
that after adjusting for BMI and heart rate, the systolic blood pressure and diastolic blood
pressure were no different between stunted and non-stunted children. The lack of
difference could be because as a child, the complications of reduced arterial compliance
have yet to take its toll on the body 64.
(5) Secondary Hypertension
It has been found that hypertension could also lead to chronic kidney disease or
chronic kidney disease could lead to high blood pressures. Interestingly, while
cardiovascular diseases are increasing at a rapid rate in underdeveloped nations, kidney
diseases are also increasing. In Sub-Saharan Africa, hypertension (HTN) has been linked
to the development of chronic kidney disease such as chronic glomerulonephritis.
Chronic kidney disease has been mainly affecting young people age twenty to fifty years.
Usually, this renal disease affects middle aged to elderly patients in developed nations 56.
Kidney function can be measured by assessing glomerular filtration rate (GFR). A crude
estimate of GFR is serum creatinine which is a breakdown product of muscle tissue that
should be almost totally cleared from the blood by the kidneys. A creatinine level of more
than 1 standard deviation (SD) above the mean for normal, healthy adults is considered
abnormal, and the progression of kidney disease can be monitored by following serum
creatinine levels over time.
Other Contributing Factors to Hypertension
20
Diet and social factors are two of the main predictors of high blood pressure.
Unfortunately, the realms of diet and/or social factors playing a role with hypertension
are very large and complex and will not be investigated thoroughly in this study. It is
important, however, to mention their significance.
(1) Diet
Hypertension has been labeled the most common cardiovascular risk on the
planet, and it is also the most preventable if humans understood how to manage it. There
are many factors that go into the development of hypertension 9. Diet has a significant
impact on a person’s likelihood of developing hypertension. With more high-fat foods
and meat, a person has a greater chance of developing cardiovascular problems.
According to the study by de Ramirez, who looked into hypertension of the rural areas of
Malawi, Rwanda, and Tanzania, those who had a high diet of meat and foods with high-
fat content had twice the odds of developing hypertension compared to those who lacked
a diet of high-fat food and meats. It was further noted that those with a diet mainly of
vegetables and fruits had a 54% smaller chance of developing hypertension17. Freedman,
who studied the relationship between salt and blood pressure, found that a daily salt
reduction of 100mmol leads to systolic pressure reduction from 1 to 6 mmHg.
Diet has thus been demonstrated to contribute to a person’s likelihood of
developing hypertension. Diet, however, by the THUSA BANA study, has been further
broken down into specific components. The study’s results only indicated that a
deficiency of folic acid and biotin are potential dietary risk factors. Rooyen, one of the
21
study’s authors, hypothesized that perhaps the lack of folic acid and biotin could lead to a
reduction of arterial compliance, which in turn causes higher blood pressures. Rooyen
addressed how magnesium, potassium, and calcium are known for their anti-hypertensive
abilities. These three dietary factors were found in limited amounts in hypertensive
patients, but yet they were also found in deficiencies in normotensive patients. Rooyen
then stated that the causes of hypertension must be a mix of dietary, environmental, and
genetic factors 64.
(2) Social Factors
Social factors also come into play when dealing with non-communicable diseases
like hypertension. Effective intervention cannot be done unless the cultural values
attached to particular risk factors, such as obesity, are understood. Many women may
view their weight as normal while outsiders may view it as an obesity problem. Others do
not want to lose any weight 9. In some societies, obesity or having a great amount of
weight is looked upon as a symbol of fortune or high status 14. Furthermore, along with
the positive relationship between blood pressure and age, BMI, and elevated glucose,
Maher has shown higher rates of hypertension along those with no education or
secondary education. Another study reinforced the inverse relationship between level of
socio-economic status and blood pressure. Longo-Mbenza found in their study that
children who had higher rates of blood pressure were usually from the lower socio-
economic class 43.
22
The Gap
As observed in many prevalence studies, non-communicable diseases are on the
rise in the developing world. One of these increasing conditions is hypertension. There is
a great importance in being able to treat hypertension. The importance comes from many
factors. One, although there are many prevalence numbers in the literature, many of them
are underestimations. Much of the hypertension in developing countries may have gone
undetected or at least unnoticed, and by the time the condition has been recognized by a
physician, the patient has already reached an advanced and severe stage of cardiovascular
disease or malignant hypertension. With this severe stage comes an increase in heart
attack and stroke for the patient as well as a “loss of man hours, diminished work
productivity, social burden, and increased health expenditure” for the economy and
family 58.
Studies have suggested a list of potential interventions for alleviating the effect
that non-communicable diseases are having on developing countries. Researchers suggest
increased physical activity and healthier diet programs and maintenance 9. These
measures can be implemented in an urban area quickly and effectively with the right
resources and financial aid, but in a very underdeveloped, remote, rural area, the
interventions are almost impossible. Thus, most of the studies that focus on new drug
trials being conducted in major hospitals of developing nations’ capitals have little to do
with the rural problem.
Because of the difficulties that come with studies in a rural, impoverished area,
many studies do not venture into those areas to calculate prevalence rates of non-
23
communicable diseases, specifically hypertension. Therefore, in developing countries
like Kenya, the prevalence rates of hypertension may have been underestimated.
Despite the large number of studies on the various issues related to hypertension,
there are still a great number of questions to be explored. Hypertension in rural areas can
result from a modifiable or a non-modifiable cause. It is important to narrow the causes
down and attempt to identify the primary cause behind hypertension in a particular
community. If the cause is modifiable (diet/physical activity/malnutrition), then the
problem can be and should be addressed. If the cause is non-modifiable (genetics), then
the problem will be harder to fix and will continue to plague society, but it is important,
however, to implement treatment programs and interventions.
Furthermore, the literature of hypertension only contains studies reinforcing the
patterns of how men are more hypertensive than women and how blood pressure
increases with BMI and age. Few studies have addressed where young adults fit the
picture of hypertension. This could be due to the fact that research and health
policymakers generally have focused on the trend of increasing age with alarmingly
increasing blood pressure. There is less attention given to atypical non-obese young
adults who are hypertensive. This point of view is the wrong way to approach the
problem of hypertension.
If the young people are walking around with untreated hypertension, then they
become ticking time bombs. It is important, then, to study the risk factors and hidden
hypertension in young adults. Only by doing so could there be a chance of preventing
24
further cardiovascular complications during old age. The goal should be to find and to
eliminate the problem while people are young and the bodies are still actively changing.
Thus, what the literature lacks are studies looking into how hypertension affects
rural villages in developing countries, and more specifically, how hypertension affects
younger people in rural areas. The literature needs a study to carefully describe the rising
phenomenon of hypertension in rural areas. This is essential in order to reduce the
proliferating rates of hypertension.
25
CHAPTER THREE
Hypothesis
With the general objective of investigating how age and gender affect the
prevalence of hypertension (i.e. high blood pressure) in rural western Kenya, this study
proposes to test, via a cross-sectional study, on a sample size of 321 patients from the
“Straw to Bread” 2010 clinic, these hypotheses:
Research Questions
Research Question One: “How does the age (i.e. two categories of young and old) of
patients affect the relationship between blood pressure and BMI?”
Hypothesis 1A: Among old patients, there is a positive relationship between BMI
and blood pressure.
Hypothesis 1B: Among young patients, there is no relationship between BMI and
blood pressure.
Research Question Two: “Is the relationship between blood pressure and BMI stronger in
men or women?”
26
Hypothesis 2A: Controlling for age, the positive relationship between BMI and
blood pressure is stronger in male than in female patients.
Research Question Three: “Among old and young patients, will body fat distribution (i.e.
truncal vs. peripheral fat) have a positive relationship with blood pressure?”
Hypothesis 3A: Among both old male and female patients, there is a positive
relationship between truncal fat and blood pressure.
Hypothesis 3B: Among both young male and female patients, there is no
relationship between truncal fat and blood pressure.
Hypothesis 3C: Among both old and young patients of both genders, there is no
relationship between peripheral fat and blood pressure.
27
Schematic Representation
Predictor Variables: Outcome Variables:
Age
High Systolic Blood Pressure
Gender
BMI
Truncal Fat
Peripheral Fat
Creatinine
Figure III.1
28
In-Depth Schematic Representation
For Older Patients
in Correlation with Blood
Pressure
Figure III.2
29
In-Depth Schematic Representation
For Younger Patients
in Correlation with Blood
Pressure
Figure III.3
30
CHAPTER FOUR
Materials and Methods
Study Area and Population
An analysis of cross-sectional clinical data was performed on previously collected
data from the May 2010 Bethlehem Home clinic located in the rural area on the Upper
Nyakach Division within the Nyando District in western Kenya. This location is
approximately 36 km southeast of Kisumu, Kenya and 12 km southeast of Lake Victoria.
Each year since 2009, a temporary clinic has been set up by American medical
professionals and students as part of a long-term collaborative medical/public health
project. Patients are seen at no cost, and the doctors see 750 – 850 patients during the
annual clinic. The data analyzed in this study came from this group of patients.
The area’s local population mostly consists of people from the Luo ethnic group.
Those who went to the clinic were mostly individuals who lived within walking distance.
The medical clinic saw approximately 685 patients. This analysis was restricted to these
321 patients, who were 18 years of age and older. This age cut-off reflects the CDC 2005
31
- 2008 standard of looking at hypertension in America (CDC) 60. The age of those
included ranged from 18 years to 100 years.
Measurements
All measurements were taken before the patient was seen by the physician.
Weight in kilograms was measured with a spring scale purchased locally in Kisumu,
Kenya. Height in centimeters was measured by using a long strip of measuring tape duct-
taped to a wall and having patients stand with their backs against a wall. Tricep skin-fold
measurements in millimeters were measured using Lange calipers (Beta Technology
Inc.). Lange calipers measure subcutaneous tissue by using its pivoted tips and rounded
corners to pinch the skin external to the muscle. The calipers can measure up to 60 mm of
subcutaneous fat. Waist and hip circumferences were measured by a measuring tape in
centimeters.
Creatinine values were measured by gathering a single venous blood sample from
each patient and then using the i-STAT handheld device, manufactured by Abbott
(Princeton, New Jersey). Controls were performed daily on all machines. The Lange
calipers and i-STAT were all used appropriately by trained technicians and volunteers.
Blood pressure was measured using a sphygmomanometer with a standard adult
cuff. Most measurements were taken on the right arm, but a second measurement was
sometimes taken on the left arm or again on the right arm. Second measurements were
taken into consideration if the first measurement seemed inaccurate (too high or too low).
32
With two measurements of blood pressures for certain patients, only the lowest blood
pressure was considered for this study. Highest blood pressures were not considered due
to possible random instrumental errors. While taking manual measurements via the
sphygmomanometer, the cuff was inflated to a pressure higher than the systolic blood
pressure. Then the cuff was deflated and upon the first ‘whooshing’ sound, the systolic
blood pressure was recorded. Once the noise disappeared, the diastolic blood pressure
was recorded. Blood pressures were only taken on adolescents and adults.
Study Design and Clinical Data
This cross-sectional study used previously gathered clinical data from May 2010.
All physical examinations and assessments were performed by Dr. Lisa Baker, a
physician trained and board certified in the United States of America. Anthropometric
measurements and lab data were taken by trained technicians and volunteers.
High blood pressure was defined as a blood pressure of 140/90 mmHg or above.
Only systolic blood pressure was analyzed in this study.
Patients were stratified into categories of young and old, according to the CDC
2005 - 2008 categorization 60. The young group consisted of patients between the ages of
18 and 44 years. The older group consisted of patients above the age of 44 years.
Body Mass Index (BMI) was calculated according to the formula: mass
(kg)/(height m)2. The WHO uses six categories of BMI to focus on underweight
individuals: severe thinness, moderate thinness, mild thinness, normal, pre-obese, and
obese. This is named BMI WHO Severe Classification (Table IV.1). 2
33
BMI WHO Severe Classification
Classification Principal cut-off points (kg/m2)
Severe thinness <16.00
Moderate thinness 16.00 – 16.99
Mild thinness 17.00 – 18.49
Normal 18.50 – 24.99
Pre-Obese 25.00 – 29.99
Obese ≥30.00
Table IV.1 28
Waist-to-hip ratio (WHR) was calculated according to the formula: waist
circumferencecm/hip circumferencecm. The ratios were then categorized into quintiles
(Table IV.2)
WHR Quintile Cut-Offs

Category WHR Range
Category 1 ≤ 0.8172
Category 2 0.8173 - 0.8498
Category 3 0.8499 - 0.8777
Category 4 0.8778 - 0.9228
Category 5 ≥ 0.9229
Table IV.2
34
Creatinine values were compared to American men and women from 1988 to
1994 (Jones 1998) (Graph IV.1).
Graph IV.1 16
Summary of Variables
The variables used in the analysis were the following:
 Study ID
 Gender
 Age
 Systolic Blood Pressure
 Weight (kg)
 Height (cm)
 Waist Circumference (cm)
 Hip Circumference (cm)
35
 Tricep skinfolds (mm)
 Creatinine.
The variables created from the above initial variables were the following:
 High systolic blood pressure (vs. normal or low)
 BMI WHO 6 categories
 BMI (continuous)
 WHR
 Age Group (young vs. old).
Statistical Analysis
All clinical data were coded and double-entered (blinded) into Microsoft Excel
(Redmond, Washington) and then imported into SAS 9.2 (Carey, North Carolina). All
variables were checked for errors and corrected when necessary. The statistical analysis
was on three levels. The initial step was a descriptive analysis of the major variables
crucial to this study: gender, age, systolic blood pressure, weight, height, waist
circumference, hip circumference, and tricep skinfold. The bivariate, or, second, level of
analysis relied on contingency table analysis, regression analysis, and ANOVA. The
third tier of analysis was multivariate in approach; using regression and ANOVA, it was
possible to estimate the effects of one predictor variable on an outcome variable while
adjusting for the impact of additional variables. Alpha was set at 0.05.
36
Descriptive Statistics
Descriptive univariate data included averages, medians, standard deviations, and
ranges. Average systolic blood pressures, BMIs, tricep skinfold measurements, WHRs,
waist circumferences, hip circumferences, and creatinine values are presented for each
category within the variables for age groups, gender, and age-gender groups. Average
systolic blood pressures, BMsI, tricep skinfold measurements, WHRs, waist
circumferences, hip circumferences, and creatinine values were stratified by age and
gender.
Analytic Statistics
Regression analysis was performed to adjust for potential confounding factors.
Regression analysis was also performed to evaluate the relationship between potential
determinants. Blood pressure, as a continuous variable, was regressed on BMI, WHR,
tricep skinfolds, truncal measurements (waist circumference and hip circumference), and
creatinine. Adjusted r2 values showed the relationship between the single determinant to
the variable of interest: blood pressure. Alpha was set at 0.05, and P values are reported.
P values of less than 0.05 were considered to indicate statistical significance.
Analysis of variance (ANOVA) was also performed to evaluate the relationship
between potential determinants. Blood pressure, as a dichotomous variable (normal
versus high), is evaluated with BMI WHO six categories, age as a dichotomous variable,
and WHR quintiles. R2 values show the relationship between the single determinant to
37
the variable of interest. Alpha was set at 0.05, and P values were reported. P values of
less than 0.05 were considered to indicate statistical significance.
IRB
This study was approved by the Baylor University Institutional Review Board.
The data abstracted from the clinical record used no unique personal identifiers. All
subjects remained anonymous.
38
CHAPTER FIVE
Results
Organization of Results
The results of this study are organized into six sections beginning with a general
overview. Descriptive results are reported first followed by analytic results. Only systolic
blood pressure measurements and analysis were taken into consideration in this analysis.
Diastolic blood pressure measurements were not used in this analysis. Finally, when the
term high blood pressure or hypertension (HTN) is used, it refers to high systolic blood
pressure unless otherwise specified.
General Overview
A total of 685 patients attended the clinic between May 15 and May 29, 2010. Out
of 685 patients, only 321 patients are included in this analysis because of the study’s cut-
off age of 18 years or older. Out of the study’s sample size of 321 patients, the average
39
patient age was 47.6 years (SD=18.2) with a range of 18 years to 100 years. Patients were
grouped according to age into two groups: the young group included patients 18 – 44
years and the old group included patients 45 years or more. There were a total of 111
patients found with high blood pressure. High blood pressure was defined as any blood
pressure measurement equal to or greater than 140 mmHg. Out of the young group, 27
patients had high blood pressure and out of the old group, 84 patients had high blood
pressure (Table V.1). When patients were grouped by gender, there were 34 males and 77
females with high blood pressure (Table V.2). Thus, in our study’s sample, there are
more old patients than young patients and more females than males.
Total Patients 685

Study's Sample Size 321
Age Group (18 - 44 years) - 148 (45+ years)- 173
Blood Pressure 112 27 9 73 84 16
Total High Blood Pressure 111 Normal Blood Pressure

Total Normal Blood Pressure 185 High Blood Pressure
Total Missing 25 Missing
Table V.1
Total Patients 685

Study's Sample Size 321
Gender Males - 78 Females - 217 26
Blood Pressure 44 34 140 77
Total High Blood Pressure 111 Normal Blood Pressure

Total Normal Blood Pressure 184 High Blood Pressure
Total Missing 26 Missing
Table V.2
40
The prevalence of elevated blood pressure (HTN) stratified by age in the
population was higher in the old age group (53.50%) compared to the young age group
(19.42%) (Table V.3). When the prevalence of HTN was stratified by gender, there was a
higher prevalence found in the males (43.59%) compared to the females (35.48%) (Table
V.3). The overall prevalence of high blood pressure in this study’s sample size was
34.58% (Table V.3).
High Systolic Blood Pressure Prevalence

Young 19.42%
Old 53.50%
Males 43.59%
Females 35.48%
Overall 34.58%
Table V.3
All BMI values were categorized into six groups based on WHO’s BMI
classification scheme: severe thinness, moderate thinness, mild thinness, normal or
moderate BMI, pre-obese, and obese (Table IV.2). WHR (Waist to hip ratio) was divided
into quintiles (Table IV.3).
Comparing the Nyakach Plateau region of this study to the United States shows
the United States to have ten times the prevalence of obesity. The United States has
approximately three times the prevalence of persons being pre-obese.
The Kenyan urban geographical area has BMI categorical percentages similar to
the Nairobi Province, the capital of Kenya, and thus a very populated, urban area. When
compared to the Nyakach Plateau, the Kenyan urban data shows double the pre-obesity
rates of the Nyakach Plateau while the Nyakach Plateau presents almost five times the
rate of underweight people as the Kenyan urban data.
41
The Nyakach Plateau BMI percentages are relatively comparable to the Nyanza
Province and the Kenyan rural areas. The area of this study has the highest prevalence of
underweight people (Table V.4).
High Blood Pressure Prevalence: Comparison of Studies

Geographical Region Year of Study Sample Size Age Criteria Prevalence by Gender Prevalence by Age Overall Prevalence
Males Females Young (18 - 44 years) Old (45+ years)
Nyakach Plateau * 2010 321 18+ 43.59% 35.48% 19.42% 53.50% 34.58%
Sub-Saharan Africa 2008 300+ 15+ 16.80% 15.70% 13.70%
Cameroon 2003 1762/1398 24-74 44.00% 34.10%
United States 2005 - 2008 10,488 18+ 30.60% 28.70% 10.50% 40.60% and 70.30% 29.90%
*Data from this study
Not available
Table V.4 2,22,28
Comparison of BMI by Regions of the World

80.00%
70.00%
Nyakach Plateau *
60.00%
Nyanza Province
50.00% Kenya Rural
40.00% Kenya Urban
Nairobi Province
30.00%
United States (Both Genders)
20.00%
United States (Female)
10.00% United States (Male)
0.00%
% Underweight % Normal % Pre Obese % Obese
Graph V.1 28
When compared to other studies in Sub-Saharan Africa, Cameroon, and the
United States, the Nyakach Plateau’s high blood pressure prevalence numbers by gender
are most similar to Cameroon’s estimates. The prevalence of HTN is 40.60% for the age
group 45-64 years and 70.30% for the age group 65+ years. When comparing the
prevalence stratified by age, the Nyakach Plateau has a higher prevalence of high blood
pressure in the young than does the United States. Out of all the regions, the highest
42
overall prevalence of high blood pressure lies in the region of this study on the Nyakach
Plateau (Table V.4).
High Blood Pressure Prevalence:

Comparison of Studies
50.00%
High Blood Pressure Prevalence
45.00%
40.00%
35.00%
30.00% Nyakach Plateau *
25.00%
20.00% Sub-Saharan Africa
15.00% Cameroon
10.00%
United States
5.00%
0.00%
Males Females
Gender
Graph V.2 2,22,28
43
Overall Data without Stratification
Averages by Variable Uncategorized

Variable N Mean (SD)
Systolic Blood Pressure (mmHg) 303 137 (21.2)
BMI (kg/m2) 301 21.23 (3.9)
Age (years) 321 47.6 (18.2)
WHR 288 0.87 (0.13)
Table V.5
Table V.5 summarizes the average systolic blood pressure, BMI (body mass
index), age, and waist to hip ratio (WHR) for the overall study’s sample size. The average
systolic blood pressure was 137 mmHg (SD=21.2). The average BMI was 21.23 kg/m2
(SD=3.9) and ranged from 13.98 to 34.84 kg/m2. The average BMI was in the normal
range according to WHO standards. The range of BMI was between the severe thinness
and obese class I categories. The average WHR was 0.87 (SD=0.13).
Analytic Statistics
The Impact of Age, BMI, and Gender on Systolic Blood Pressure (mmHg)
Systolic Blood Pressure Regressed On:
Regression Incremental Cumulative
Variable Coefficient S.E. t P value R2 R2
Intercept 94.37 7.45 12.67 0.0001
Age (years) 0.47 0.06 7.37 0.0001 0.1495 0.1495
2
BMI (kg/m ) 0.96 0.29 3.34 0.0010 0.0293 0.1788
Gender* -2.98 2.69 -1.11 0.2691 0.0005 0.1793
*0=Male, 1=Female
Table V.6
44
Regressing systolic blood pressure, as a continuous variable, on the potential
determinants created the regression coefficients in Table V.6-8. Out of all the potential
determinants, age explained the most variance (14.95%) and was a significant predictor
of blood pressure (p=0.0001). When systolic blood pressure was regressed on age as a
dichotomous variable, the explained variance was less than for age as a continuous
variable. Thus, age was a better predictor of systolic blood pressure when analyzed
continuously than when broken down into two categories of young and old. BMI was a
significant indicator of blood pressure but explained less of the variance than age
(p=0.0010). When BMI was combined with age, they explained together a total of
17.88% of the variance: when the two variables were combined with gender, the
explained variance increased minimally to 17.93%, but gender was not significant
(p=0.2691) while BMI and age remained significant (p=0.0010 and p=0.0001). Thus,
gender is not a significant predictor of blood pressure (Table V.6).
The Impact of Hip Circumference on Systolic Blood Pressure (mmHg)

Variable Coefficient S.E. t P value R2 R2
Intercept 130.87 4.65 28.15 0.0001
Hip
Circumference 0.070 0.05 1.26 0.2072 0.0020 0.0020
(cm)
Table V.7
45
The Impact of Hip Circumference (Obese) on Systolic Blood Pressure (mmHg)
2 2
Variable Coefficient S.E. t P value R R
Intercept 117.51 23.74 4.95 0.0011
Hip
Circumference
(cm) on BMI 0.42 0.24 1.77 0.1145 0.1919 0.1919
WHO Severe
Category: Obese
Table V.8
Hip circumference explained little of the variance in systolic blood pressure (R2 =
0.0020) (p=0.2072) (Table V.7). However, when systolic blood pressure was regressed
on hip circumference among those in the obese BMI WHO category, the potential
determinant was able to explain 19.19% of the variance in blood pressure (p=0.1145).
The insignificant p value could be due to the small sample size of 10 patients in the BMI
WHO obese category (Table V.8). The potential determinants of waist circumference,
WHR, and tricep skinfolds were not significant predictors of blood pressure. When blood
pressure was regressed on waist circumference and hip circumference combined, the p
value remained insignificant. The same outcome of insignificant p value was
demonstrated when BMI and waist circumference were regressed together with systolic
blood pressure.
46
Analysis of Variance
AVERAGES Total R-
Variable P
Values in red are average systolic blood pressure measurements in mmHg Square
Severe Thinness Mod Thinness Mild Thinness Normal Pre-Obese Obese
BMI Classification 133 136 138 134 138 158 0.0266 0.043574
by WHO N=9 N=18 N=44 N=169 N=39 N=10
SD=23.5 SD=32.0 SD=20.6 SD=19.9 SD=18.6 SD=25.8
Category 1 Category 2 Category 3 Category 4 Category 5
WHR (All 5 132 135 135 138 140 0.2348* 0.01951
Quintiles) N=57 N=57 N=58 N=56 N=58
SD=18.1 SD=22.8 SD=16.0 SD=19.1 SD=27.5
* Not statistically significant
Table V.9
The potential impact of BMI on systolic blood pressure was explored using
analysis of variance (ANOVA). BMI was analyzed using six WHO groups (Table V.9).
The average systolic blood pressure was 158 mmHg for obese persons and 133 mmHg
for the severely thin (p = 0.0266, r2 = 0.0436). Waist-hip ratio (WHR), in quintiles, was
not as good a predictor of systolic blood pressure as BMI (Table V.9).
Systolic blood pressure was regressed on BMI as a continuous variable (r2 =
0.0293, p = 0.0010); BMI as a categorical variable explains more variance in systolic
blood pressure than BMI as a continuous variable. It is clear than the BMI categories of
the WHO have clinical relevance.
47
Data by Age Groups
Averages by Variable and Category

Variable N Mean (SD) Median Minimum Maximum
Young (18 - 44 years)
Systolic Blood Pressure (mmHg) 142 128 (15.4) 125 90 190
BMI (kg/m2) 141 21.63 (3.7) 20.47 16.29 34.84
Age (years) 148 31 (7.3) 30.5 18 44
WHR 138 0.87 (0.14) 0.86 0.36 1.54
Waist Circumference (cm) 138 69.72 (20) 74.4 26 114.5
Hip Circumference (cm) 138 81.15 (23.1) 88.5 25.7 118.6
Tricep Skinfold (mm) 139 15.3 (8.4) 14 3 41
Old (45+ years)
BMI (kg/m2) 160 20.88 (4.1) 20.01 13.98 34.77
Age (years) 173 62 (11.5) 60 45 100
WHR 160 0.89 (0.08) 0.88 0.67 1.48
Waist Circumference (cm) 160 74.04 (20.5) 77.6 26 114.5
Hip Circumference (cm) 160 83.9 (22.4) 88.55 27.5 119
Table V.10
Table V.10 stratifies the sample by young and old age groups. Descriptively
those patients who are older have a higher average systolic blood pressure than those
patients who are younger. The highest blood pressure measurement of 226 mmHg was
found in the old age group. The average BMI in the younger group was 21.63 kg/m2
which falls in the BMI WHO normal category, and the average BMI in the older group
was 20.88 kg/m2 which also falls in the BMI WHO normal category. The highest average
BMI value was higher in the young than the old patients (Table V.10).
48
By Age Group
Average Blood Pressure vs. BMI WHO
Classification
180
169
Average Blood Pressure (mmHg)
160
140 146 141 143 145 146
133 134 130
120 126
114
100
80 Old
60 Young
40
20
0
1 2 3 4 5 6
BMI WHO Classification
Graph V.3
Graph V.3 shows the relationship stratified by age group between average blood
pressures and the six categories of BMI by WHO. The old age group has higher average
blood pressures across all six BMI categories than the young age group. Those patients in
the old age group have high blood pressure averages, and those patients in the young
group have normal to high blood pressure averages (Graph V.3).
49
Analytic Statistics
The Association of Hypertension and BMI Classification in Younger Patients

Systolic Hypertension
Percentage of those with
Yes No Total hypertension in each BMI
group
0 0 0
Severe Thinness 0.00
0.00 0.00 0.00
1 5 6
Moderate Thinness 3.7
16.67 83.33 4.32
7 12 19
Mild Thinness 25.93
36.84 63.16 13.67
12 79 91
Normal 44.44
13.19 86.81 65.47
3 15 18
Pre-Obese 11.11
16.67 83.33 12.95
4 1 5
Obese 14.81
80.00 20.00 3.60
27 112 139
Total 100
19.42 80.58 100
χ2 = 17.18; p=0.0014
Table V.11
50
The Association of Hypertension and BMI Classification in Older Patients
group
4 5 9
44.44 55.56 5.73
7 7 14
50.00 50.00 8.92
15 12 27
Mild Thinness 17.86
55.56 44.44 17.20
42 39 81
Normal 50.00
51.85 48.15 51.59
11 10 21
Pre-Obese 13.10
52.38 47.62 13.38
5 0 5
Obese 5.95
100.00 0.00 3.18
84 73 157
Total 100
53.50 46.50 100
χ2 = 4.86; p=0.4337
Table V.12
The association between BMI WHO categories and systolic hypertension (HTN)
was explored using contingency table analysis. Among the younger age group (Table
V.11), 19.42 % of the sample was hypertensive. Among the obese 80% were
hypertensive and among the severely thin, 0.0 % were hypertensive (p = 0.0014). Among
the older adults (Table V.12) the association between BMI and HTN was not statistically
significant; however, all 5 of the obese older adults were hypertensive.
51
The Impact of BMI on Systolic Blood Pressure (mmHg) Stratified by Age Group
Variable S.E. t P value
Coefficient R2 R2
Intercept 112.31 7.68 14.63 0.0001
2
BMI (kg/m ) 0.73 0.35 2.08 0.0390 0.0238 0.0238
Intercept 125.84 9.49 13.25 0.0001

BMI (kg/m2) 0.86 0.45 1.93 0.0552 0.0172 0.0172
Young
Old
Table V.13
The Impact of Waist and Hip Circumferences on Systolic Blood Pressure (mmHg) Stratified by Age
Group
Coefficient R2 R2
Intercept 119.84 4.90 24.46 0.0001
Waist Circumference (cm) 0.09 0.07 1.41 0.1613 0.0072 0.0072
Hip Circumference (cm) 0.18 0.12 1.53 0.1275 0.0098 0.0170
Intercept 143.41 7.00 20.51 0.0001

Waist Circumference (cm) 0.06 0.09 0.71 0.4807 -0.0032 -0.0032
Hip Circumference (cm) -0.57 0.24 -2.35 0.0201 0.0284 0.0252
Young
Old
Table V.14
The Impact of WHR on Systolic Blood Pressure (mmHg) Stratified by Age Group
Coefficient R2 R2
Intercept 139.20 8.28 16.82 0.0001
WHR -12.60 9.39 -1.34 0.1819 0.0058 0.0058
Intercept 97.58 19.29 5.06 0.0001

WHR 52.26 21.70 2.41 0.0172 0.0299 0.0299
Young
Old
Table V.15
52
Regressing systolic blood pressure, as a continuous variable, stratified by age
group, on the potential determinants produced the regression coefficients presented in
Table V.13-15. BMI was found to explain 2.38% of the variance in blood pressure in the
young and 1.72% of variance in the old, and is a significant predictor of blood pressure in
both age groups (p=0.0390 and p=0.0552) (Table V.13). When waist circumference and
hip circumference were analyzed together, they were shown to be significant predictors
of blood pressure only in the old (p=0.0201) with hip circumference explaining most of
the variance (Table V.14). Lastly, WHR only was found to be a significant predictor with
an R2 value of 0.0299 in the old age group (p=0.0172) (Table V.15). Tricep skinfold
measurements were found not to be statistically significant predictors of blood pressure.
Thus, for the young age group, BMI was the only significant predictor of blood
pressure. For the old age group, BMI and WHR were significant predictors of blood
pressure.
Analysis of Variance Stratified by Age Group

AVERAGES Total R-
Variable P
--- 114 134 126 130 146 0.002 0.117835
N=0 N=6 N=19 N=91 N=18 N=5
BMI Classification SD=---- SD=18.2 SD=19.8 SD=13.2 SD=14.1 SD=18.6
by WHO Severe Thinness Mod Thinness Mild Thinness Normal Pre-Obese Obese
133 146 142 143 145 169 0.1288* 0.054474
N=9 N=14 N=27 N=81 N=21 N=5
SD=23.5 SD=29.4 SD=22.7 SD=22.3 SD=19.4 SD=28.5

127 129 132 128 124 0.3941* 0.030277
N=32 N=32 N=28 N=20 N=25
WHR (All 5 SD=15.2 SD=16.5 SD=15.0 SD=15.3 SD=15.1
Quintiles) Category 1 Category 2 Category 3 Category 4 Category 5
139 142 137 144 153 0.0576 0.06113
N=25 N=25 N=30 N=36 N=33
SD=19.8 SD=27.5 SD=16.9 SD=18.8 SD=28.3
Young
Old
Table V.16
53
The potential impact of BMI WHO classification and WHR, adjusted by age, on
systolic blood pressure were evaluated using ANOVA (Table V.16). Among the younger
adults, the average systolic blood pressure was 146 mmHg in the obese and 114 mmHg in
the moderately thin (there were no young adults in the severely thin WHO category) (r2 =
0.1178, p = 0.002). Similar trends were found among the older adults, but the differences
in blood pressure between the BMI groups did not reach statistical significance. WHR
was associated with blood pressure among the older adults (r2 = 0.0611, p = 0.0576), but
not among the younger adults.
Data by Gender

Males
Systolic Blood Pressure (mmHg) 79 141 (21) 136 108 226
BMI (kg/m2) 82 19.81 (3.0) 18.96 15.51 31.29
Age (years) 87 55 (19) 59 19 86
WHR 80 0.88 (0.2) 0.89 0.008 1.19
Waist Circumference (cm) 81 72.99 (19) 77 28.3 114.5
Hip Circumference (cm) 81 82.75 (20) 87 31 115.8
Tricep Skinfold (mm) 81 8.81 (6.9) 5.8 2 30
Females
BMI (kg/m2) 218 21.76 (4.1) 20.70 13.98 34.84
Age (years) 232 45 (17) 43 18 100
WHR 207 0.87 (0.1) 0.86 0.44 1.54
Waist Circumference (cm) 216 71.67 (21) 76.4 26 114.5
Hip Circumference (cm) 216 82.5 (24) 89 25.7 119
Table V.17
54
Table V.17 stratifies by males and females. Descriptively those patients who are
males have a higher average systolic blood pressure than those patients who are females.
The highest blood pressure measurement of 226 mmHg was found in the males. The
average BMI in the males was 19.81 kg/m2 which falls in the BMI WHO normal
category, and the average BMI in the females was 21.76 kg/m2 which falls in the BMI
WHO normal category. Descriptively the highest average BMI value was higher in the
females than the males (Table V.17).
By Gender
Average Blood Pressure vs. BMI WHO
Classification
180
160 159
151
Average Blood Pressure (mmHg)
145 142
140 139 140
137 139
133 132 130
120
114
100
80 Males
Females
60
40
20
0
1 2 3 4 5 6
BMI WHO Categories
Graph V.4
Graph V.4 shows the relationship stratified by gender between average blood
pressures and the six categories of BMI by WHO. Descriptively males had higher
55
average blood pressures in the BMI WHO groups of moderate thinness, mild thinness,
and normal. Females, descriptively, had higher average blood pressures in the BMI WHO
groups of severe thinness, pre-obese, and obese (Graph V.4).
Analytic Statistics
The Association of Hypertension and BMI Classification in Male Patients

group
0 2 2
0.00 100.00 2.56
3 3 6
50.00 50.00 7.69
12 13 25
Mild Thinness 35.29
48.00 52.00 32.05
17 24 41
Normal 50.00
41.46 58.54 52.56
0 15 2
Pre-Obese 0.00
0.00 83.33 2.56
2 0 2
Obese 5.88
100.00 00.00 2.56
34 44 78
Total 100
43.59 56.41 100
χ2 = 6.05; p=0.3011
Table V.18
56
The Association of Hypertension and BMI Classification in Female Patients
group
4 3 7
57.14 42.86 3.23
5 9 14
35.71 64.29 6.45
10 11 21
Mild Thinness 12.99
47.62 52.38 9.68
37 93 130
Normal 48.05
28.46 71.54 59.91
14 23 37
Pre-Obese 18.18
37.84 62.16 17.05
7 1 8
Obese 9.09
87.50 12.50 3.69
77 140 217
Total 100
35.48 64.52 100
χ2 = 15.13; p=0.0098
Table V.19
Among males, BMI and HTN were not significantly associated (p = 0.3011)
(Table V.18). Among females, the relationship between BMI and HTN was statistically
significant (p =0.0098), but the pattern is difficult to interpret (Table V.19). The highest
proportion (87.5%) of HTN was among the obese, but the next highest proportion (57.14)
was among the severely thin group. The lowest proportion (28.46) of HTN was among
the normal BMI group.
57
The Impact of BMI and Age on Systolic Blood Pressure (mmHg) Stratified by
Gender
Coefficient R2 R2
Intercept 117.60 18.10 6.50 0.0001
2
BMI (kg/m ) 0.44 0.81 0.54 0.5894 -0.0093 -0.0093
Age (years) 0.21 0.12 1.69 0.0949 0.0242 0.0149
Intercept 83.42 7.84 10.65 0.0001

BMI (kg/m2) 0.88 0.35 2.53 0.0122 0.0243 0.0243
Age (years) 0.60 0.07 8.11 0.0001 0.2258 0.2501
Male
Female
Table V.20
Regressing systolic blood pressure, as a continuous variable, on the potential
determinants, stratified by gender, produced the regression coefficients presented in
Table V.20. BMI was found to be significant only in the females explaining 2.43% of the
variance in blood pressure (p=0.0122). Age was found to be significant only in the
females explaining 22.58% of the variance in blood pressure (p=0.0001). Thus, when
systolic blood pressure was regressed on BMI and age combined, the p value was
significant only among the females with an r2 value of 0.2501 (p=0.0122 and p=0.0001)
(Table V.20). The other potential determinants of systolic blood pressure - waist
circumference, hip circumference, WHR, and tricep skinfolds - were found not to be
statistically significant predictors (data not shown). The regression analyses did not
produce any significant predictors of blood pressure among the males.
58
Analysis of Variance Stratified by Gender
Cumulative
AVERAGES
Variable P (Total) R-
Values in red are average systolic blood pressure measurements in mmHg
Square
114 145 140 142 130 150 0.4836* 0.059049
N=2 N=6 N=25 N=41 N=2 N=2
BMI Classification SD=8.5 SD=43.1 SD=18.9 SD=19.1 SD=7.1 SD=7.8
139 133 137 132 139 159 0.0095 0.069055
N=7 N=14 N=21 N=132 N=37 N=8
SD=23.8 SD=23.9 SD=22.7 SD=19.7 SD=18.9 SD=28.8
Young Old 0.4631*
134 144 0.104179
N=24 N=54
BMI Classification 0.0627*
SD=17.4 SD=22.2
by WHO and Age
(dichotmous) Young Old 0.0027
127 144 0.231853
N=114 N=103
<.0001
SD=14.9 SD=23.4
Males
Females
Table V.21
An analysis of variance among the males revealed BMI not to be related to
systolic blood pressure (Table V.21). Among the females, the relationship was
statistically significant (p = 0.0095), with BMI explaining 6.9% of the variance in
systolic blood pressure; the pivotal difference between the BMI groups was the fact that
the obese women had the highest average systolic blood pressure (159 mmHg). When
the dichotomous age variable was added to the ANOVA model, the two variables of BMI
and age explained 23% of the variance in systolic blood pressure among the women and
10.4% of the variance among the men.
59
Data by Age Group and Gender

Young Males (18-44 years)
BMI (kg/m2) 25 20.1 (2.8) 19.77 16.29 25.94
Age (years) 26 30 (6.8) 30.5 19 43
WHR 25 0.86 (0.16) 0.87 0.36 1.19
Waist Circumference (cm) 24 67.71 (18.7) 73.5 28.3 87.2
Hip Circumference (cm) 24 80.48 (19.9) 85 33 107.5
Old Males (45+ years)
BMI (kg/m2) 57 19.68 (3.1) 18.83 15.51 31.29
Age (years) 61 66 (10.7) 65 47 86
WHR 55 0.89 (0.14) 0.90 0.78 1.10
Waist Circumference (cm) 57 75.2 (18.9) 78 29.3 114.5
Hip Circumference (cm) 57 83.7 (20.4) 88 31 115.8
Young Females (18-44 years)
BMI (kg/m2) 115 21.95 (3.8) 20.66 16.49 34.84
Age (years) 120 31 (7.3) 32 18 44
WHR 111 0.87 (0.14) 0.85 0.44 1.54
Waist Circumference (cm) 113 70.1 (20.4) 75 26 114.5
Hip Circumference (cm) 113 81.15 (23.8) 89 25.7 118.6
Tricep Skinfold (mm) 114 16.24 (8) 15 3 41
Old Females (45+ years)
BMI (kg/m2) 103 21.55 (4.4) 20.77 13.98 34.77
Age (years) 112 59 (11.4) 58 45 100
WHR 96 0.87 (0.09) 0.87 0.67 1.48
Waist Circumference (cm) 103 73.4 (21.4) 76.9 26 113.5
Hip Circumference (cm) 103 84 (23.5) 89 27.5 119
Table V.22
60
Table V.22 stratified by both age groups and gender, creating four groups: young
males, old males, young females, and old females. Descriptively the old male patients
and the old female patients have the highest average systolic blood pressure (144 mmHg).
All average BMI measurements fall in the BMI WHO normal category. The highest
average BMI measurement was in the young female group (21.95 kg/m2) (Table V.22).
Analytic Statistics
The Impact of BMI on Systolic Blood Pressure (mmHg) Stratified by Age-

Gender Groups
Coefficient R2 R2
Intercept 138.56 27.29 5.08 0.0001
2
BMI (kg/m ) -0.22 1.35 -0.17 0.8701 -0.0442 -0.0442
Intercept 128.71 19.38 6.64 0.0001

BMI (kg/m2) 0.77 0.98 0.79 0.4348 -0.0072 -0.0072
Intercept 104.12 7.85 13.26 0.0001

2
BMI (kg/m ) 1.03 0.35 2.93 0.0041 0.063 0.063
Intercept 123.56 11.55 10.69 0.0001

BMI (kg/m2) 0.94 0.53 1.79 0.076 0.0212 0.0212
Young Males
Old Males
Young Females
Old Females
Table V.23
61
The Impact of Waist and Hip Circumferences on Systolic Blood Pressure (mmHg) Stratified by
Age-Gender Groups
Coefficient R2 R2
Intercept 117.74 15.35 7.67 0.0001
Waist Circumference (cm) 0.28 0.19 1.47 0.1560 0.0479 0.0479
Hip Circumference (cm) -0.09 0.27 -0.34 0.7362 -0.0398 0.0081
Intercept 137.81 12.71 10.85 0.0001

Hip Circumference (cm) -0.39 0.47 -0.84 0.4073 -0.0059 -0.0154
Intercept 120.00 5.03 23.85 0.0001

Hip Circumference (cm) 0.29 0.13 2.19 0.0309 0.033 0.0329
Intercept 145.89 8.49 17.18 0.0001

Hip Circumference (cm) -0.65 0.29 -2.23 0.0279 0.0383 0.0299
Young Males
Old Males
Young Females
Old Females
Table V.24
Regressing systolic blood pressure, as a continuous variable, stratified by age
group and gender, on the potential determinants produced the regression coefficients
presented in Table V.23 and Table V.24. BMI was found to be a significant predictor of
blood pressure only in the young females with an r2 of 0.063 (p=0.0041) (Table V.23).
When waist circumference and hip circumference were analyzed together, hip
circumference became significant only in the old females (p=0.0279) with hip
circumference explaining most of the variance (Table V.24). The other potential
determinants - waist circumference, hip circumference, WHR, and tricep skinfolds - were
found not to be statistically significant predictors of blood pressure (data not shown).
62
Analysis of Variance Stratified by Gender and Age Group
AVERAGES Total R-
Variable P
--- 121 148 132 125 --- 0.1130* 0.252969
N=0 N=3 N=6 N=14 N=1 N=0
SD=---- SD=18.1 SD=23.1 SD=12.0 SD=--- SD=---
114 168 137 148 135 151
0.0697* 0.186419
N=2 N=3 N=19 N=27 N=1 N=2
BMI Classification SD=8.5 SD=51.9 SD=17.3 SD=19.9 SD=--- SD=7.8
--- 106 128 125 131 146
0.0021 0.142071
N=0 N=3 N=13 N=76 N=17 N=5
SD=---- SD=17.8 SD=15.3 SD=13.3 SD=14.5 SD=18.6
139 140 152 141 146 182 0.0680* 0.098994
N=7 N=11 N=8 N=54 N=20 N=3
SD=23.8 SD=20.1 SD=25.9 SD=23.4 SD=19.7 SD=31.8

122 160 139 142 131 0.0387 0.382817
N=7 N=2 N=5 N=3 N=8
SD=11.7 SD=42.4 SD=14.5 SD=15.7 SD=7.6
148 138 137 141 150
0.576* 0.058495
N=5 N=6 N=10 N=12 N=19
WHR (All 5 SD=23.7 SD=21.4 SD=18.8 SD=18.8 SD=27.2
Quintiles) Category 1 Category 2 Category 3 Category 4 Category 5
128 127 131 126 121
0.3261* 0.042452
N=24 N=30 N=23 N=17 N=17
SD=16.3 SD=12.7 SD=15.1 SD=14.3 SD=16.8
136 144 137 146 157 0.0695* 0.089173
N=20 N=19 N=20 N=24 N=14
SD=18.6 SD=29.5 SD=16.3 SD=18.9 SD=30.3
*Not statistically significant
Young Males
Old Males
Young Females
Old Females
Table V.25
An analysis of variance of systolic blood pressure among the four age-gender
groups revealed BMI to be associated with blood pressure only among the young females
(r2 = 0.14, p = 0.0021) (Table V.25). WHR was associated with systolic blood pressure
only among the young men (r2 = 0.38, p = 0.0387).
63
Creatinine

Creatinine 23 0.99 (0.17) 1 0.7 1.3
Creatinine 46 1.07 (0.36) 1 0.6 3
Creatinine 92 0.76 (0.16) 0.7 0.5 1.4
Creatinine 84 0.79 (0.18) 0.8 0.5 1.9
Table V.26
Table V.26 stratified creatinine by both age groups and gender, creating four
groups: young males, old males, young females, and old females. Descriptively the
highest average creatinine value was in the old male group and the lowest average
creatinine value was in the young female group. The highest creatinine value out of the
entire sample size (3.00 mg/dL) also fell into the old male group (Table V.26).
Graph V.5 16
64
The sample’s creatinine values were compared to the creatinine values of
American men and women from 1988 to 1994 (Jones 1998) (Graph V.5).
Abnormal Creatinine Descriptive Statistics by Age and Gender
Number of
Creatinine Range Abnormal Blood Pressure BMI Range
2
(mg/dL) Creatinine Values Age Range (years) Range (mmHg) (kg/m )
All were 1.3 3 19 - 35 132 - 156 18.01 - 21.78
1.3 - 3.0 8 48 - 82 122 - 185 17.26 - 30.9617
1.1 - 1.4 4 21 - 38 110 - 126 17.15 - 20.52
1.1 - 1.9 4 45 - 78 118 - 152 16.51 - 28.58
Table V.27
When systolic blood pressure, as a continuous variable, was regressed on
creatinine, creatinine explained 1.34% of the variance in blood pressure (p=0.0382).
According to US standards presented in Graph V.5, the normal levels of
creatinine in the blood are approximately 0.5 – 1.1 mg/dL in females. Out of the 176
women in the study sample with measured creatinine values, 8 of them were considered
abnormal or above 1.1 mg/dL, and, thus, there was a 4.55% prevalence of abnormal
creatinine in women. Six women had creatinine values of 1.1 mg/dL (+1 SD); 2 women
were >2SD above the normal creatinine value. There was an equal amount of young
women and old women with abnormal creatinine values. Among the young women with
abnormal creatinine, their blood pressure measurements were considered mostly normal.
Among the old women with abnormal creatinine, their blood pressure measurements
65
were mostly high and considered hypertensive. Finally, there was a prevalence of 2.27%
of abnormal creatinine in both young and old females (Table V.27).
According to US standards (Graph V.5), the normal levels of creatinine in the
blood are approximately 0.6 – 1.2 mg/dL in males. Out of the 69 males with measured
creatinine values, 11 of them were considered abnormal or above 1.2 mg/dL; and, thus,
there was a 15.94% prevalence of abnormal creatinine in men. Seven men had creatinine
values that were +1SD above normal; four men were >2SD above the normal creatinine
value. There was a greater amount of older men with abnormal creatinine values than
younger men. Among the young men with abnormal creatinine, their blood pressure
measurements were high. Finally, there was a prevalence of 4.35% of abnormal
creatinine in young males and 11.59% of abnormal creatinine in old males (Table V.27).
Range of Blood Pressure

Percentage numbers are abnormal creatinine prevalence numbers
Old Females 2.27%
Young Females 2.27%
Old Males 11.59%
Young Males 4.35%
0 50 100 150 200 250 300 350 400

Blood Pressure (mmHg)
Graph V.6
66
Range of BMI
Percentage numbers are abnormal creatinine prevalence numbers
Old Females 2.27%
Young Females 2.27%
Old Males 11.59%
Young Males 4.35%
0 10 20 30 40 50 60
BMI (kg/m2)
Graph V.7
Graph V.6 presents a graph showing the range of blood pressures stratified by age
group and gender with accompanying prevalences of abnormal creatinine. The data
shows descriptively that the highest blood pressures were found in the old male group,
and the highest abnormal creatinine prevalence was in the same group (Graph V.6).
Graph V.7 presents a final graph showing the range of BMI stratified by age group and
gender with accompanying prevalences of abnormal creatinine. The data shows
descriptively that the highest BMI’s are found in the young females group and some of
the lowest BMI’s are found in the old female group and the old male group. Among the
young male patients, BMI ranges from moderate thinness to pre-obese. Among the old
male patients, BMI ranges from severe thinness to obese. Among the young female
patients, BMI range from moderate thinness to obese. Among the old female patients,
people range from severe thinness to obese (Graph V.7).
67
Data Results as Predicted by Hypotheses
Research Question One: “How does the age (i.e. two categories of young and old) of
patients affect the relationship between high blood pressure and BMI?”
Hypothesis 1A: Among old patients, there is a positive relationship between BMI and
blood pressure.
Table V.13 displays the results relevant to this hypothesis. According to the
regression model, there was a positive relationship between BMI and blood pressure
among old patients, but the relationship is not strong. BMI was shown to be a significant
predictor of blood pressure in the old group with a small r2 of 0.0172 (p=0.0552).
Therefore the data supports this hypothesis.
Hypothesis 1B: Among young patients, there is no relationship between BMI and blood
pressure.
Table V.13 displays the results relevant to this hypothesis. According to the
regression analysis, there was a positive relationship between BMI and blood pressure
among young patients. BMI was shown to be a significant predictor of blood pressure in
the young group with an r2 of 0.0238 (p=0.0390). Therefore this hypothesis is rejected.
Upon further investigation, it was found that BMI as a categorical variable was an
even better predictor of blood pressure than BMI as a continuous variable. When BMI
was categorized using WHO criteria, it gained in explanatory power in the young group
(r2=0.117835; p=0.0020) (Table V.16).
68
Research Question Two: “Is the relationship between blood pressure and BMI stronger in
men or women?”
Hypothesis 2A: Controlling for age, the positive relationship between BMI and blood
pressure is stronger in male than in female patients.
Tables V.20 display the results relevant to this hypothesis. Table V.20 show that
BMI was a significant predictor of blood pressure only in females (p=0.0122). BMI was
not statistically significant for the males (p=0.5894).
Upon closer inspection in Table V.21, BMI as a categorical variable explained
more variance than BMI as a continuous variable for females. BMI as a categorical
variable remains statistically significant for females with an r2 of 0.069055 (p=0.0095).
BMI as a categorical variable was not statistically significant for males (p=0.4836).
Table V.19 shows that BMI was a significant predictor of high blood pressure
among female patients (p=0.0098). BMI was not significant among males associated with
HTN (p=0.3011) (Table V.18). The highest prevalence of high blood pressure in males
was in the mild thinness category, suggesting a negative relationship between blood
pressure and BMI in males.
Therefore, the relationship between BMI and blood pressure seems to be stronger
in female patients than male: a potential negative relationship exists between BMI and
blood pressure in males while one does exist in females. Thus, we reject this hypothesis.
69
Research Question Three: “Among old and young patients, will body fat distribution (i.e.
truncal vs. peripheral fat) have a positive relationship with blood pressure?”
Hypothesis 3A: Among both old male and female patients, there is a positive relationship
between truncal fat and blood pressure.
Truncal fat in this study was measured by waist circumference, hip
circumference, and WHR (waist to hip ratio). Table V.24 displays the results relevant to
this hypothesis. Waist circumference as a single variable and hip circumference as a
single variable were found not to be statistically significant predictors of blood pressure
for both old male and old female patients. However, when waist circumference and hip
circumference were analyzed together, hip circumference became significant only in the
old female patients (p=0.0279) with hip circumference explaining most of the variance.
WHR as a continuous variable was not a significant predictor of blood pressure;
when WHR was categorized as quintiles, it remained insignificant (Table V.25).
Thus, the data showed no positive relationship between truncal fat and blood
pressure in old male and female patients. Thus, we reject the hypothesis.
Hypothesis 3B: Among both young male and female patients, there is no relationship
between truncal fat and blood pressure.
Table V.29 displays the results relevant to this hypothesis. Waist circumference as
a single variable and hip circumference as a single variable were found not to be
70
statistically significant predictors of blood pressure for both young male and female
patients.
WHR as a continuous variable was not a significant predictor of blood pressure,
but when WHR was categorized as quintiles, it became significant in young male patients
with an r2=0.382817. This high variance, however, could be due to the small sample size
of young male patients (Table V.25).
Hypothesis 3B is partially supported and partially rejected.
Hypothesis 3C: Among both old and young patients of both genders, there is no
relationship between peripheral fat and blood pressure.
Peripheral fat in this study was measured by tricep skinfold measurements. Tricep
skinfold was found not to be a statistically significant predictor of blood pressure for both
old and young patients of both genders. Therefore, there was no relationship between
peripheral fat and blood pressure. Thus, this hypothesis was supported by the data.
71
CHAPTER SIX
Discussion and Conclusions
Overview
The purpose of this study was to investigate the prevalence of high blood pressure
and associated factors in a rural area of western Kenya with the long-term goal of finding
a way to lessen the increasing morbidity and mortality rates of hypertension, an
increasingly serious non-communicable disease in developing countries. Despite the
world’s international focus on infectious/parasitic diseases, there is a steep rise of non-
communicable diseases in developing areas. In 2005, infectious diseases caused 5.5
million deaths in Africa. In the same year, non-communicable diseases (NCDs) caused
over 2.4 million deaths. While the mortality rate of infectious diseases is still higher, the
mortality rate of non-communicable diseases is not too far behind 58. Thus, there is a need
for more research in the area of NCD, especially with regard to hypertension.
Much of the literature concentrates on hypertension in urban areas in Africa. Even
in these studies, however, many articles draw attention to the lack of studies done in rural
areas. Thus, the hypertension prevalence numbers generated in studies may be
underestimations of the population. In a 2008 study done in Sub-Saharan Africa, the
72
prevalence in the rural area was reported as 13.70%. In our 2010 study done in the
Nyakach Plateau in rural western Kenya, the prevalence was found to be 34.58%. The
prevalence number in this region more than doubles the prevalence stated in the Sub-
Saharan African study. It is startling to see that this prevalence even exceeds the
prevalence of hypertension in the United States. The CDC estimated the overall
prevalence of hypertension in United States to be 29.90%.
When stratifying by gender, the prevalence of hypertension in 43.59% of males
and 35.48% of females is comparable to Cameroon’s 44.00% in males and 34.10% in
females, lending face validity to the current study. The numbers from the Cameroon
study were generated from urban and rural areas. The gender stratified prevalence
numbers are higher than the United States (30.60% in males and 28.70% in females).
According to Table V.8, the prevalence of hypertension in the United States is 40.60%
among those aged 45 – 64 and 70.30% among those aged 65+. This study found that the
age stratified prevalence for younger adults is 19.42% - almost double the prevalence rate
of the United States (10.50%). These prevalence numbers of the Nyakach Plateau are
startling, and they surpass the rural hypertension rates found in other relevant studies
except for Cameroon. Because hypertension is a multifactorial disease, it is crucial to
investigate further what factors are associated with alarming rates of high blood pressure
in our rural area.
73
Primary Hypertension
The most common cause of high blood pressure in the United States and in many
cases around the world is primary, or essential, hypertension. Excessive vasoconstriction
is frequently associated with and exacerbated by obesity, a component of metabolic
syndrome. However, in a part of rural Kenya where the majority of people are
malnourished and thin, the data has shown that many cases of hypertension are occurring
without the pattern that we have to come to think of as typical. Hypertension associated
with increased weight seems to be occurring in the females more than the males of this
study. While BMI has a weak, albeit positive relationship with blood pressure in males,
BMI shares a stronger positive relationship with blood pressure in females. BMI was also
proven to be a statistically significant predictor of the severity of high blood pressure for
the females. Furthermore, the average BMI in females (21.76 kg/m2) is higher than the
average BMI in males (19.81 kg/m2). This indicates that the female patients in our study
have higher body fat content than the male patients. Upon further investigation and
stratification by age group and gender, it was found that BMI was a statistically
significant predictor of the severity of blood pressure specifically in younger females.
While this finding seems surprising, the descriptive statistics show the young female
patients having the highest BMI in the study sample compared to old females, old males,
and young males.
When BMI was broken down categorically, it revealed more into the reason why
hypertension may be the explanation of high blood pressures in females. BMI was broken
down into six categories using WHO BMI cut-off standards. Excluding the category of
74
normal BMI, the high percentage of women (18.18%) with high blood pressure is in the
pre-obese BMI category (Table V.19). Graph V.4 shows a linear relationship between
BMI and average blood pressure for females. As the BMI category jumps from normal to
pre-obese and to obese, average blood pressure spikes in the obese category (151 mmHg)
(Graph V.4). The data conclusively shows that women follow the typical primary
hypertension pattern. With increasing BMI, women tend to have increasing blood
pressures.
Age was found to be a better predictor of blood pressure than BMI in females.
Thus, when BMI and age were combined, together the variables were able to explain
25% of the variance in blood pressure. The data shows that women follow the typical
primary hypertension pattern. With increasing BMI and increasing age, women tend to
have increasing blood pressures.
BMI vs. WHR
There were two measures of body fat content used in this study. BMI was still
considered a statistically significant predictor of blood pressure for the old patients, but
BMI was found to be a better predictor of blood pressure for the young patients.
Furthermore, BMI was found to be a better predictor of blood pressure than WHR for the
young patients. WHR, however, was found to be a better predictor of blood pressure than
BMI for old patients. This finding brings in the discussion of apple-shaped bodies versus
pear-shaped bodies.
75
Apple-shaped bodies are at more risk of developing cardiovascular diseases and
hypertension than pear-shaped bodies. This is due to the fact that apple-shaped bodies
have more abdominal adiposity or central obesity and a larger waist circumference than
pear-shaped bodies which carry fat in their hips and thighs. WHR rather than BMI is a
better measure of central obesity. This explains why WHR is a better predictor of blood
pressure for the old group because there are more old people than younger people with
central obesity. BMI is not the best measure for abdominal adiposity and may not be a
good measure of hypertension risk in old people. BMI remains a better predictor of blood
pressure in the young since most young people do not take on the apple-shaped bodies.
Lastly, when BMI was operationalized categorically, it became a better
statistically significant predictor of the severity of blood pressure than continuous BMI.
Thus, this shows the BMI WHO categories are clinically relevant.
One interesting finding in our study shows WHR being a very significant
predictor of blood pressure in young males. WHR in young males was able to explain
around 38% of the variance in blood pressure (Table V.25). This could be due to random
error from a small sample size or because of another physiological reason. WHR could
act as a better predictor of the severity of blood pressure than BMI in young males
primarily because BMI is not the best predictor of cardiovascular or hypertensive risk due
to the fact that muscle mass is never taken into consideration. Because muscle weighs
more than fat and young males tend to have more muscle than other age-gender groups,
WHR may eclipse BMI as a better predictor of blood pressure.
76
Lastly, although much of the literature states how waist circumference is an
excellent indicator of cardiovascular risk, this study has found that waist circumference
was not a statistically significant predictor of blood pressure for any age group or gender.
Curvilinear Relationship of Blood Pressure
Another interesting finding in these data was a curvilinear blood pressure versus
BMI relationship in both young and old. In the old age group, there is an increase in
average blood pressure in the moderate thinness BMI category and in the pre-obese and
obese categories versus the average blood pressure in the normal BMI category. In the
young age group, there is an increase in average blood pressure in the mild thinness BMI
category and in the pre-obese and obese categories. The curvilinear relationship trends
can be viewed in Graph V.3. When comparing prevalence, there were greater percentages
of people with high blood pressure in the mild thinness category (25.93% and 17.86%)
compared to the percentages of people with high blood pressure in the pre-obese and
obese categories (Table V.11-12).
What we find in our study sample is a curvilinear relationship. Instead of having
high blood pressures concentrated in those with high body fat content, high blood
pressures are being concentrated among those who are thin and those who are fat;
however, blood pressure remains normal for those in the normal BMI category. This
trend is happening in both the young and the old. Those who have high blood pressure
and are in the pre-obese and obese categories follow the typical blood pressure pattern of
higher BMI and higher age equals higher blood pressure. What is curious are those who
77
have high blood pressure and are in the thin BMI categories. There are a number of
reasons why this relationship could happen. There could be a different pathophysiological
process going on that involves genetics. Another explanation could be that the thinner
people are more likely to have secondary hypertension not associated with the typical
weight pattern. Studies have also shown that perhaps low birth weight or childhood
malnutrition could lead to a later incidence of hypertension. If the mother has poor
nutrition and the baby is born with a low birth weight, there could be some physiological
differences in the infant. Infants could develop less vascular elasticity, leading to adult
hypertension. Those children who are malnourished and stunted could have less vascular
compliance and thus, later cases of hypertension. This is a reasonable explanation for
hypertension in thin people in this sample given the frequency of drought, famine, and
poverty. However, it is difficult to test this hypothesis in this population since very few
babies are born in a hospital and are weighed.
Secondary Hypertension
In our analysis, we have found that there are few variables that could be
significant predictors of blood pressure for males. BMI, age, waist circumference, and
WHR were found not to be statistically significant predictors of blood pressure for males.
Looking at the prevalence, there is a high percentage of males with high blood pressure in
the mild thinness category. In Graph V.4, it shows a peak of average blood pressure for
males in the moderate thinness category and a drop in average blood pressure in the pre-
obese category. Thus, there are more males with high blood pressure who are thin rather
than fat.
78
This is a very interesting and odd finding and could be explained in terms of
secondary hypertension, where another disease is causing hypertension. There are
multiple reasons for secondary hypertension, including primary renal disease, Cushing’s
disease, thyroid disease, hyperaldosteronism, pheochromocytoma, obstructive sleep
apnea, and coarctation of the aorta. The only one of these diseases that this study was
able to explore was renal disease using serum creatinine values. Through the results
indicate that renal disease may indeed be associated with hypertension in this sample, it is
impossible to know if the renal compromise was the cause or the effect of the
hypertension. When analyzing creatinine values, it was found that there are a higher
percentage of males rather than females with abnormal creatinine values. Abnormal
creatinine values are an indicator of renal problems, specifically a decrease in glomerular
filtration rate. Upon closer inspection, the results showed that those male patients with
abnormal creatinine values have mostly high blood pressure measurements. Their BMI’s
revealed that about half of the males with abnormal creatinine values are considered
mildly thin. There is also a higher number of older males with abnormal creatinine values
than young males, and as mentioned before, there are more old patients who are thinner
than young patients. All of these findings together point toward the potential of a kidney
disease causing high blood pressures among the thin males. A potential kidney disease
causing high blood pressures could be chronic glomerulonephritis which the literature has
indicated may be causing high blood pressure in Africa.
People who are malnourished may have lower creatinine values than expected for
their age and gender. Because this study is looking at creatinine values which are higher
79
than the normal and because this study has revealed a high percentage of underweight
patients, the prevalence numbers of abnormal creatinine could be underestimated.
One other cause of secondary hypertension is the use of oral contraceptives.
Though the use of oral contraceptives in this population was not measured, informal
reports suggest that it is very low. Nevertheless, oral contraceptive use would be a useful
variable to include in future studies.
Systematic Threats to Internal Validity
Although BMI and age explained much of the variance in blood pressure in
females and WHR explained much of the variance in blood pressure in young males,
there is still much of blood pressure that cannot be explained in any of the groups.
Therefore, there must be other factors contributing to blood pressure of which we are
unaware or were unable to measure in this study. Furthermore, this study was done on a
clinic sample, and therefore, the prevalence of hypertension could be overestimated for
the general population.
Random Threats to Internal Validity
The study found small to moderately large p-values in the regression and analysis
of variance models. Although some p-values may be moderately large, they were less
than alpha. In some regression and analysis of variance models, p values larger than alpha
were found. Therefore, this study could have been significantly affected by random errors
such as small sample size when variables were stratified into smaller groups.
80
Furthermore, the sample has potential bias in that it represents only those able to travel to
the clinic, only those in close proximity to the clinic, only those who were aware of the
clinic, only those able to be seen in the 14 days of clinic, and only those willing to endure
the 8 or more hours waiting in line for the clinic.
Construct Validity
A non-automated sphygmomanometer was used to determine high blood pressure.
Automated readings of blood pressure are more accurate, and the manual technique was a
source of potential systematic bias. BMI and WHR are accepted measurements of body
fat, and the WHO BMI categories used are widely accepted. Weight could have been
more precisely measured, but whatever error was present was at least consistent between
patients since only one scale was used.
The greatest threat to construct validity in this study is the method used to
determine if a patient has hypertension. Because blood pressure was only taken at a
single point in time, this method of determining high blood pressure as a chronic disease
is only suggestive. And though care was taken to train those who performed the blood
pressure measurements, some degree of inter-observer and intra-observer variability is
possible. Nevertheless, this method offers the best approximation of high blood pressure
in a clinic where most patients cannot be seen multiple times due to time constraints and
travel. Follow-up measurements of blood pressure would provide a better standard in
whether patients truly had chronic high blood pressure or whether the abnormal
measurement was due to chance, an error in measurement, or a transient response to
81
illness or stress. It would be expected that some of the elevation in blood pressure would
be due to acute illness that brought the patient to the clinic. However, almost none of the
adults were severely ill as indicated by temperature > 101°F or a clinical evaluation by
the physician of severe illness.
Generalization across Persons, Settings, and Time
By identifying only patients with blood pressure greater than 140/90 mmHg, the
researchers attempted to minimize falsely labeling as positive those who may truly be
normal or pre-hypertensive. Some features of this study can be generalized to similar,
rural populations in sub-Saharan Africa. To the degree that this group of Luo people is
genetically unique, however, the study may only describe the prevalence among this
tribe.
Conclusion
This study reveals many different concerning points about hypertension in
developing countries. It has shown that perhaps there is a greater prevalence of high
blood pressure in rural areas than many assume. This could be due to the difficulty of
venturing into a rural area and taking measurements. The other frustration that could lead
to researchers not spending enough time studying rural areas could be the difficulty of
maintaining a treatment program to help the people.
This study has shown how females follow the typical blood pressure correlation
with higher BMI and age. This study has also shown that there are a great many more
82
younger patients with this silent disease than previously thought. It is unclear how much
high blood pressure is resulting from a non-modifiable factor (genetics) compared to the
contribution of modifiable factors (diet/lifestyle/childhood nutrition), especially in those
patients who are young, those patients who are male, and those patients who are thin. The
degree of hypertension in these groups suggests that there may be a particularly strong
genetic component in this ethnic group. Addressing modifiable risk factors and
identifying treatable causes of secondary hypertension, especially among young patients,
would reduce the problem of future hypertension during old age.
In conclusion, this community-based research which has grown out of troubling
clinical observations has indeed confirmed surprising findings of high blood pressure in
those patients who are thin and those patients who are young. Results have shown BMI
and age to be significant predictors of blood pressure in the females and not in males, and
have also shown how females fit the model of metabolic syndrome with obesity. BMI
was proven to be a better measurement of high blood pressure in the young and WHR
was proven to be a better measurement of high blood pressure in the old. The study has
also demonstrated the possibility of kidney disease causing hypertension in the groups
where BMI and age cannot explain high blood pressure. Hypertension can lead to many
life-threatening or fatal cardiovascular diseases such as heart attack and stroke, and the
rates are increasing at an alarming rate. It is essential to the future of developing countries
that we investigate and try to find a solution to high blood pressure before it succeeds in
taking its devastating toll on millions of people and countless communities.
83
APPENDICES
84
APPENDIX A
Bethlehem Home
The “Straw to Bread” Project
Bethlehem Home is located on the Nyakach Plateau. It was started by Pastor
Habil Ogola as a ministry to care for sixty orphans and destitute elders in 2001. Through
the partnership of Pastor Habil Ogola, four other local people, the “Straw to Bread”
project (started by Dr. Lisa Baker, MD PhD), and Baylor University students, this project
has grown beyond the task of providing food and healthcare to the people. It is not just a
“band-aid” mission trip and project; it is a growing relationship and friendship born out
of humility, care, compassion, equality, and love.
The “Straw to Bread” project seeks to create ways for Luo people to live
independently. Most mission trips arrive in an impoverished place and provide temporary
healthcare and food supplies to the starving people. This is a temporary cure to what ails
poverty in third world regions. The “Straw to Bread” project aims to establish something
different. By building a school and creating plans to build a permanent clinic and guest
house, this project, along with Pastor Habil Ogola, has a firm dream and mission of
creating ways for the members of Bethlehem Home to be able to permanently support
and sustain themselves.
85
While this dream seems far from reach, progress has already bloomed amongst
the hopeful community. Gardens are being planted in the school’s backyard to teach the
children about growing their own vegetables. Water sources are being improved. Gutters
are being installed on the roofs of huts. Through many partnerships with people in the
U.S., several Luo young people are being sent to pharmacy school, tailoring school, and
medical school.
In 2009, a 20,000-liter cistern was built to collect drinking water and thatched
roofs were replaced with sheet metal. Students built wooden bed frames for the elders
and healthcare was provided to over 800 people living in the Nyakach Plateau area.
In 2010, rain gutters were installed to harvest rainwater for drinking, and fruit
trees were planted at homes. Goats were purchased to provide milk for drinking and
nutrition supplementation, and a swing-set was constructed for the orphans. The clinic
improved by using a medical laboratory to test blood samples for diabetes, anemia, and
malaria while “Straw to Bread” became the first in Kenya to use adjustable glasses for
people with poor vision through Global Vision 2020.
In 2011, 800 patients were seen by Dr. Lisa Baker and Dr. Bob Dimski. Business
seminars were hosted on the plateau, and there was advancement in the water tanks and
gutter systems installer around the area. Education in reproductive health issues were
trained to the “Village Mothers”, an established group of nine local women leaders.
Clinical research in asthma, high blood-lead levels, anemic patients, and hypertensive
patients continued.
86
In 2012, four physicians and two business professors will travel to the Nyakach
Plateau. A garden will be started and given to the school children and one business
professor will begin to train the orphans how to take care of the goats. Clinical research
will continue.
87
APPENDIX B
Love There, Love Here, Love Near

Here is Kenya – the poorest of the grounds
But we look closer and around
And we see through the eyes above
Oh here it is, here is the meaning of love
Love is seeing Habil with tears in his eyes

Hugging every person with joyous cries
No matter the religious, racial walls
For in this world, we accept all
Love is taking a photo with a child

Running with them through plains and wild
Teaching them new letters and words
And them teaching us games unheard
Love is witnessing plant fledglings’ growth

Of colors, avocados, mangoes, or both
Pushing new seeds into the soil
And praying and watching these farmers’ toils
Love is the mother and baby goats

And organizing scattered data notes
And milk so fresh and white and clear
And the elders giving care so dear
Love is van trips up and down

Sleeping and dreaming of a future so sound
Hearing ‘Mzungo!’ being shouted in the streets
And just being with these people in African heat
Love is gazing at the clear skies of stars

Appreciating home and friends so far
Giving head rubs and back rubs every night
Conversing of life and love by dim light
88
Love is an elder smiling so beautiful and shy
And seeing a very sick baby live not die
It is the woman with a swollen leg and desperate tears
And one physician who can quiet her fears
Love is understanding pain

Dancing with the children in the rain
Huddled under rain jackets and boots
Watching water soak in dry roots
Love is heavy rain turned into a waterfall

And moving through greens and trees so tall
It is hearing the people’s prayers and low hums
It’s how shared spirituality can make us humbly succumb
Love is understanding beauty is not the same

Depending on who you are and what you have to claim
But giving a pen, the gift of words
And seeing in a child, a mind suddenly stirred
Love is the temporary clinic on the Nyakach Plateau

And gazing at Lake Victoria and the world of Kenya below
It is watching a man not take a meal
So more orphans will eat and their hungers will heal
Love is dancing and singing with these people of faith

And despite language, being able to relate
Handing blankets and shoes and toys
To ecstatic elders, teenagers, girls, and boys
Love is working with them even though we are apart

It is forever leaving in Kenya my heart
And together uplifting our spirits to a level above
Oh here it is, here is the meaning of love
89
REFERENCES
1. Abdhalah K Ziraba, Jean C Fotso, and Rhoune Ochako, “Overweight and obesity
in urban Africa: A problem of the rich or the poor?,” BMC Public Health 9
(2009): 465.
2. Addo, Juliet, Liam Smeeth, and David A. Leon. “Hypertension in sub-Saharan

Africa - A Systematic Review RID G-2195-2010.” Hypertension 50.6 (2007):
1012-1018.
3. Agyemang, C et al. “Blood Pressure Patterns in Rural, Semi-urban and Urban

Children in the Ashanti Region of Ghana, West Africa.” Bmc Public Health 5
(2005): n. pag.
4. Agyemang, C, MA Bruijnzeels, and E Owusu-Dabo. “Factors Associated with

Hypertension Awareness, Treatment, and Control in Ghana, West Africa.”
Journal of Human Hypertension 20.1 (2006): 67-71.
5. Agyemang, Charles. “Rural and Urban Differences in Blood Pressure and

Hypertension in Ghana, West Africa.” Public Health 120.6 (2006): 525-533.
6. Alberts, M et al. “Prevalence of Cardiovascular Diseases and Associated Risk

Factors in a Rural Black Population of South Africa.” European Journal of
Cardiovascular Prevention & Rehabilitation 12.4 (2005): 347-354.
7. Amusa, L. O., and D. T. Goon. “Blood Pressure Among Overweight Children

Aged 7-13 Years in 10 Rural Communities in South Africa: The Tshannda
Longitudinal Study.” Pakistan Journal of Medical Sciences 27.3 (2011): 664-667.
Print.
90
8. Ayoola, Omolola O. et al. “Maternal Malaria, Birth Size and Blood Pressure in
Nigerian Newborns: Insights into the Developmental Origins of Hypertension
from the Ibadan Growth Cohort.” Plos One 6.9 (2011): n. pag.
9. Bochud, M et al. “High Heritability of Ambulatory Blood Pressure in Families of

East African Descent RID A-3981-2010 RID F-4477-2011.” Hypertension 45.3
(2005): 445-450.
10. Boschitsch, E., S. Mayerhofer, and D. Magometschmgg. “Hypertension in

Women: The Role of Progesterone and Aldosterone.” Climacteric 13.4 (2010):
307-313.
11. Brewster, LM, JF Clark, and GA van Montfrans. “Is Greater Tissue Activity of
Creatine Kinase the Genetic Factor Increasing Hypertension Risk in Black
People of sub-Saharan African Descent?” Journal of Hypertension 18.11 (2000):
1537-1544.
12. Cappuccio, Francesco P. et al. “Body Size and Blood Pressure - An Analysis of
Africans and the African Diaspora RID D-3028-2009 RID F-4477-2011.”
Epidemiology 19.1 (2008): 38-46.
13. Center for History and New Media. “Zotero Quick Start Guide.”
14. Crook, ED. “The Genetics of Human Hypertension.” Seminars in Nephrology

22.1 (2002): 27-34.
15. Crowley, Leonard. An Introduction to Human Disease. 7. Boston: Jones and

Bartlett Publishers, 2007. Print.
16. "Distribution serum creatinine." Up To Date. n. page. Web. 8 Apr. 2012.

<http://www.uptodate.com/index>
91
17. Ehret, GB. “Genetic Variants in Novel Pathways Influence Blood Pressure and
Cardiovascular Disease Risk.” Web. 5 Feb. 2012.
18. Ejike, Chukwunonso E. C. C. et al. “Blood Pressure Patterns in Relation to

Geographic Area of Residence: a Cross-sectional Study of Adolescents in Kogi
State, Nigeria.” Bmc Public Health 8 (2008): n. pag.
19. Ely, Daniel et al. “Review of the Y Chromosome, Sry and Hypertension.”
Steroids 75.11 (2010): 747-753.
20. Ergul, S et al. “Racial Differences in Plasma Endothelin-1 Concentrations in

Individuals with Essential Hypertension.” Hypertension 28.4 (1996): 652-655.
Print.
21. Evensen, Kari Anne Indredavik et al. “Effects of Preterm Birth and Fetal Growth
Retardation on Cardiovascular Risk Factors in Young Adulthood.” Early Human
Development 85.4 (2009): 239-245.
22. Fezeu, L, A P Kengne, et al. “Ten-year Change in Blood Pressure Levels and
Prevalence of Hypertension in Urban and Rural Cameroon.” Journal of
Epidemiology & Community Health 64 (2009): 360-365. Web. 24 Nov. 2011.
23. Fezeu, L., B. Balkau, et al. “Waist Circumference and Obesity-related

Abnormalities in French and Cameroonian Adults: The Role of Urbanization and
Ethnicity.” International Journal of Obesity 34.3 (2010): 446-453.
24. Fisher, NDL et al. “Age, Gender, and Non-modulation - A Sexual Dimorphism in
Essential Hypertension.” Hypertension 29.4 (1997): 980-985. Print.
25. Fontbonne, A. et al. “Anthropometric Characteristics and Cardiometabolic Risk

Factors in a Sample of Urban-dwelling Adults in Senegal.” Diabetes &
Metabolism 37 (2011): 52-58. Web. 24 Nov. 2011.
92
26. Fourie, Carla M. T. et al. “The Influence of Cardiac and Vascular Responses on
Baseline Cardiovascular Parameters in Black African Children.” Ethnicity &
Disease 18.2 (2008): 187-191. Print.
27. Gillum, RF, ME Mussolino, and JH Madans. “Body Fat Distribution and
Hypertension Incidence in Women and Men - The NHANES I Epidemiologic
Follow-up Study.” International Journal of Obesity 22.2 (1998): 127-134.
28. "Global Database on Body Mass Index." WHO (World Health Organization),
2006. Web. 8 Apr 2012. <http://apps.who.int/bmi/index.jsp>.
29. Gombet, Thierry, Olivier Steichen, and Pierre-Francois Plouin. “Hypertensive

Disease in Subjects Born in sub-Saharan Africa or in Europe Referred to a
Hypertension Unit : a Cross-sectional Study.” Bulletin De L Academie Nationale
De Medecine 191.8 (2007): 1745-1754. Print.
30. Hallberg, P et al. “Gender-specific Association Between Preproendothelin-1

Genotype and Reduction of Systolic Blood Pressure During Antihypertensive
Treatment - Results from the Swedish Irbesartan Left Ventricular Hypertrophy
Investigation Versus Atenolol (SILVHIA).” Clinical Cardiology 27.5 (2004):
287-290.
31. Hayes, SN, and SJ Taler. “Hypertension in Women: Current Understanding of

Gender Differences.” Mayo Clinic Proceedings 73.2 (1998): 157-165. Print.
32. Hernandez, Adrian V. et al. “Effect of Rural-to-urban Within-country Migration

on Cardiovascular Risk Factors in Low- and Middle-income Countries: a
Systematic Review.” Heart 98.3 (2012): 185-194.
33. Huxley, Virginia H. “Sex and the Cardiovascular System: The Intriguing Tale of
How Women and Men Regulate Cardiovascular Function Differently.” Advances
in Physiology Education 31.1 (2007): 17-22.
34. “Hypertension - PubMed Health.” n. pag. Web. 8 Apr. 2012.
93
35. Jenson, Alexander et al. “Assessment of Hypertension Control in a District of
Mombasa, Kenya.” Global Public Health 6 (2011): 293-306. Web. 24 Nov. 2011.
36. Kaufman, JS et al. “Relationship Between Blood Pressure and Body Mass Index
in Lean Populations.” Hypertension 30.6 (1997): 1511-1516. Print.
37. Kraja, Aldi T. et al. “Genetics of Hypertension and Cardiovascular Disease and
Their Interconnected Pathways: Lessons from Large Studies.” Current
Hypertension Reports 13.1 (2011): 46-54.
38. Kuklinska, A. M. et al. “High-sensitivity C-reactive Protein and Total Antioxidant

Status in Patients with Essential Arterial Hypertension and Dyslipidemia.”
Advances in Medical Sciences 54.2 (2009): 225-232.
39. LAVILLE, M et al. “Epidemiologic Profile of Hypertensive Disease and Renal

Risk-factors in Black-africa.” Journal of Hypertension 12.7 (1994): 839-843.
Print.
40. Lawes, CMM et al. “Blood Pressure and the Global Burden of Disease 2000. Part
1: Estimates of Blood Pressure Levels.” Journal of Hypertension 24.3 (2006):
413-422.
41. Leccia, G et al. “Sex-related Influence of Body Size and Sexual Maturation on
Blood Pressure in Adolescents.” European Journal of Clinical Nutrition 53.4
(1999): 333-337.
42. Lodenyo, HA, SO McLigeyo, and EN Ogola. “Cardiovascular Disease in Elderly

In-patients at the Kenyatta National Hospital, Nairoibi-Kenya.” East African
Medical Journal 74.10 (1997): 647-651. Print.
94
43. Longo-Mbenza, B., E. Lukoki Luila, and J. R. M’Buyamba-Kabangu.
“Nutritional Status, Socio-economic Status, Heart Rate, and Blood Pressure in
African School Children and Adolescents.” International Journal of Cardiology
121.2 (2007): 171-177.
44. Makgae, P. J. et al. “Somatotype and Blood Pressure of Rural South African
Children Aged 6-13 Years: Ellisras Longitudinal Growth and Health Study.”
Annals of Human Biology 34.2 (2007): 240-251.
45. Maria Sanchez-Zamorano, Luisa et al. “Body Mass Index Associated with
Elevated Blood Pressure in Mexican School-aged Adolescents.” Preventive
Medicine 48.6 (2009): 543-548.
46. Marina A Njelekela et al., “Gender-related differences in the prevalence of

cardiovascular disease risk factors and their correlates in urban Tanzania,” BMC
Cardiovascular Disorders 9 (2009): 30.
47. Mathenge, Wanjiku, Allen Foster, and Hannah Kuper. “Urbanization, Ethnicity
and Cardiovascular Risk in a Population in Transition in Nakuru, Kenya: a
Population-based Survey.” BMC Public Health 10 (2010): 569. Web. 24 Nov.
2011.
48. Mazicioglu, Mustafa Mumtaz et al. “Anthropometric Risk Factors for Elevated
Blood Pressure in Adolescents in Turkey Aged 11-17.” Pediatric Nephrology
25.11 (2010): 2327-2334.
49. Mensah, G. A. “Epidemiology of Stroke and High Blood Pressure in Africa.”

Heart 94.6 (2008): 697-705.
50. Minniti, A. et al. “Comparison of Physical and Psychological Status in Younger

and Older Overweight-obese Women.” Nutrition Metabolism and
Cardiovascular Diseases 21.12 (2011): 909-914.
95
51. Monyeki, K. D., and H. C. G. Kemper. “The Risk Factors for Elevated Blood
Pressure and How to Address Cardiovascular Risk Factors: a Review in
Paediatric Populations.” Journal of Human Hypertension 22.7 (2008): 450-459.
52. Monyeki, K. D., H. C. G. Kemper, and P. J. Makgae. “Relationship Between Fat

Patterns, Physical Fitness and Blood Pressure of Rural South African Children:
Ellisras Longitudinal Growth and Health Study.” Journal of Human
Hypertension 22.5 (2008): 311-319.
53. Monyeki, KD, HCG Kemper, and PJ Makgae. “The Association of Fat Patterning
with Blood Pressure in Rural South African Children: The Ellisras Longitudinal
Growth and Health Study.” International Journal of Epidemiology 35.1 (2006):
114-120.
54. Motswagole, B. S. et al. “The Sensitivity of Waist-to-height Ratio in Identifying

Children with High Blood Pressure.” Cardiovascular Journal of Africa 22.4
(2011): 208-211.
55. Muraguri, PW, SO McLigeyo, and JK Kayima. “Proteinuria, Other Selected

Urinary Abnormalities and Hypertension Among Teenage Secondary School
Students in Nairobi, Kenya.” East African Medical Journal 74.8 (1997): 467-473.
Print.
56. Naicker, S. “Burden of End-stage Renal Disease in sub-Saharan Africa.” Clinical

Nephrology 74 (2010): S13-S16. Print.
57. Njelekela, Marina A. et al. “Gender-related Differences in the Prevalence of

Cardiovascular Disease Risk Factors and Their Correlates in Urban Tanzania
RID F-4062-2010.” Bmc Cardiovascular Disorders 9 (2009): n. pag.
58. Oladapo, O. O. et al. “A Prevalence of Cardiometabolic Risk Factors Among a

Rural Yoruba South-western Nigerian Population: a Population-based Survey.”
Cardiovascular Journal of Africa 21.1 (2010): 26-31. Print.
96
59. Paradis, G et al. “Blood Pressure and Adiposity in Children and Adolescents.”
Circulation 110.13 (2004): 1832-1838.
60. “Prevalence of Hypertension and Controlled Hypertension --- United States,

2005--2008.” Web. 8 Apr. 2012.
61. de Ramirez, S Stewart et al. “Prevalence and Correlates of Hypertension: a Cross-

sectional Study Among Rural Populations in sub-Saharan Africa.” Journal of
Human Hypertension 24 (2010): 786-795. Web. 24 Nov. 2011.
62. Regitz-Zagrosek, V et al. “Gender Aspects in Heart Failure - Pathophysiology and

Medical Therapy.” Archives Des Maladies Du Coeur Et Des Vaisseaux 97.9
(2004): 899-908. Print.
63. Reyes, D, SQ Lew, and PL Kimmel. “Gender Differences in Hypertension and

Kidney Disease.” Medical Clinics of North America 89.3 (2005): 613-+.
64. van Rooyen, Johannes M. et al. “Early Cardiovascular Changes in 10- to 15-year-
old Stunted Children: The Transition and Health During Urbanization in South
Africa in Children Study.” Nutrition 21.7–8 (2005): 808-814. Web. 3 Feb. 2012.
65. van der Sande, MAB. “Cardiovascular Disease in sub-Saharan Africa: a Disaster
Waiting to Happen.” Netherlands Journal of Medicine 61.2 (2003): 32-36. Print.
66. Schutte, AE, HW Huisman, et al. “Associations Between Arterial Compliance and
Anthropometry of Children from Four Ethnic Groups in South Africa: The
THUSA BANA Study.” Blood Pressure 12.2 (2003): 97-103.
67. Schutte, AE, JM van Rooyen, et al. “Dietary Risk Markers That Contribute to the
Aetiology of Hypertension in Black South African Children: The THUSA BANA
Study.” Journal of Human Hypertension 17.1 (2003): 29-35.
97
68. Seedat, YK. “Hypertension in Developing Nations in sub-Saharan Africa.”
Journal of Human Hypertension 14.10-11 (2000): 739-747.
69. ---. “Is the Pathogenesis of Hypertension Different in Black Patients?” Journal of
Human Hypertension 10 (1996): S35-S37. Print.
70. Stevens, J et al. “The Body Mass Index-mortality Relationship in White and
African American Women.” Obesity Research 6.4 (1998): 268-277. Print.
71. Steyn, NP et al. “Weight and Health Status of Black Female Students.” South
African Medical Journal 90.2 (2000): 146-152. Print.
72. Tibazarwa, Kemi et al. “A Time Bomb of Cardiovascular Risk Factors in South
Africa: Results from the Heart of Soweto Study ‘Heart Awareness Days’ RID A-
6207-2008.” International Journal of Cardiology 132.2 (2009): 233-239.
73. Tiffin, Nicki et al. “Computational Analysis of Candidate Disease Genes and
Variants for Salt-Sensitive Hypertension in Indigenous Southern Africans RID
D-2810-2009.” Plos One 5.9 (2010): n. pag.
74. Tostes, Rita C. et al. “Endothelin, Sex and Hypertension.” Clinical Science 114.1-
2 (2008): 85-97.
75. Twagirumukiza, Marc et al. “Current and Projected Prevalence of Arterial

Hypertension in sub-Saharan Africa by Sex, Age and Habitat: An Estimate from
Population Studies.” Journal of Hypertension 29.7 (2011): 1243-1252.
76. Westaway, Margaret S. “The Impact of Chronic Diseases on the Health and Well-
being of South Africans in Early and Later Old Age.” Archives of Gerontology
and Geriatrics 50.2 (2010): 213-221.
98
77. Woelk, GB. “Is Low Birth Weight a Risk Factor for Adult Hypertension? A
Literature Review with Particular Reference to Africa.” South African Medical
Journal 85.12 (1995): 1348-& Print.
78. YK Seedat, “Is the pathogenesis of hypertension different in black patients?,”

Journal of Human Hypertension 10 (September 1996): S35-S37.
79. Ziraba, Abdhalah K, Jean C Fotso, and Rhoune Ochako. “Overweight and
Obesity in Urban Africa: A Problem of the Rich or the Poor?” BMC Public
Health 9 (2009): 465. Web. 24 Nov. 2011.
99

Monica Sok Honorsthesis

Uploaded by

Copyright:

Available Formats

Monica Sok Honorsthesis

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Monica Sok Honorsthesis

Uploaded by

Copyright:

Available Formats

ABSTRACT

Rural Western Kenya

Director: Lisa Baker, MD PhD

In order to effectively lessen the drastically increasing morbidity of

APPROVED BY THE HONORS PROGRAM:

OF HYPERTENSION IN RURAL WESTERN KENYA

A Thesis Submitted to the Faculty of

List of Figures, Tables, and Graphs iv

communicable diseases, especially hypertension

Hypertension: The Current World Status

If this word, ‘hypertension’, is mentioned to anyone in suburban America, the

in developing countries. There would be no surprise – just a look of sympathy.

Unfortunately, communicable diseases are not the only pathological processes

threat is non-communicable diseases (NCD).

Cardiovascular diseases are spreading rampant throughout Africa. With

cardiovascular diseases. It is this silent and undetected hypertension that is leading to

dominating current research.

having hypertension 73.

Hypertension is considered a serious problem because of the many complications

serious complications hypertension could lead to is a cerebral hemorrhage or stroke.

artery, thus bringing damage to the brain 15, 34.

effects on the heart, blood vessels, and kidneys 15, 34.

hypertrophic cardiomyopathy, a similar condition. Another cardiac side-effect would be

dizziness, and chest pain 15, 34.

dying 15, 34.

hypertension as well. Conditions that can cause hypertension as a secondary effect

aorta, and preelampsia 15, 34.

systolic blood pressure is slightly or highly elevated. This is a hypertension primarily

vasoconstriction, the diastolic pressure remains normal 15, 34.

Other conditions that have to do with hypertension include preeclampsia and

and can lead to seizures. Metabolic syndrome refers to a combination of hypertension,

least 150mg/dl and a blood pressure of +140/90 mmHg 15, 34.

 Family history of hypertension, heart disease, or diabetes

 Over the age of 55

 Not physically active

 Diet high in saturated fats or salt

 Illicit drug usage

 NSAIDS or decongestant drugs

hypertension is severe or malignant enough, possible symptoms will be produced:

Nyakach Plateau Clinic Observations

not all that was happening in this area.

developing countries is turning into a crisis.

Thus, this research investigates the possibility of finding a strange relationship

also add to the general scientific knowledge of this devastating disease.

The Global Significance

When people hear underdeveloped or developing nations, many think of

diseases are starting to become an epidemic. Recent statistics show that

non-communicable diseases (NCDs) caused over 2.4 million deaths. Non-communicable

increase by 60% 3,4.

Particular anthropometric measurements indicate different things and can be more

(BMI) and waist-to-hip ratio (WHR).

The WHR is an extremely important variable in hypertension. According to

The Gender Differences

According to a study by de Ramirez, men had an overall higher hypertension prevalence

related to systolic blood pressures (SBP) 61.

require to become hypertensive 19,24, 75.

ethnicities. Usually, angiotensin II, involved in the renin-angiotensin-aldosterone system,

there happens to be a complementary release of nitric oxide, a vasodilator. Fisher also

Furthermore, Hallberg studied gender specific associations with the vasoconstrictor

endothelin – a contributing vascular factor to hypertension. Hallberg found that in