MANAGEMENT OF CHILDHOOD DIARRHOEA IN RURAL GHANA: 
THE CASE OF PUTE IN THE DANGME-EAST DISTRICT

BY

MARTIN HUSHIE

A THESIS SUBMITTED TO THE DEPARTMENT OF SOCIOLOGY IN 
PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE 

OF MASTER OF PHILOSOPHY AT THE UNIVERSITY OF GHANA,
LEGON.

OCTOBER, 1994

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



£48851
R j ^5'6rD£ H 95

1W, “fXwOVf\

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



DEDICATION

THIS STUDY IS DEDICATED TO MY PARENTS CORLEY AND KORKOR, AND MY
BROTHERS AND SISTERS.

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



(i)
DECLARATION

I, MARTIN HUSHIE HEREBY DECLARE THAT, THIS THESIS CONTAINS NO 
MATERIAL WHICH HAS BEEN ACCEPTED FOR THE AWARD OF OTHER DEGREE IN 
ANY UNIVERSITY, THAT TO THE BEST OF MY KNOWLEDGE AND BELIEF, THIS 
THESIS CONTAINS NO MATERIAL PREVIOUSLY PUBLISHED OR WRITTEN BY 
ANOTHER PERSON, EXCEPT WHEN DUE REFERENCE HAS BEEN MADE IN THE 
TEXT.

MARTIN HUSHIE 
31/10/94

(SUPERVISOR)
31/10/94

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



ACKNOWLEDGEMENT
(ii)

It is a worthwhile tradition to show appreciation to all people who 
contribute in diverse ways to make work of this nature 'a dream 

come true'.

I wish to thank all Lecturers of the Sociology Department, Legon, 
who were overtly or covertly associated with the study. A special 
mention must be made of my supervisor, Mr. K.A. Senah who was very 
much helpful in the initial formulation of the research problem; 
and who in spite of his numerous engagements has actively involved 
himself at all stages of the study. I wish to record my gratitude 
for his insightful comments and invaluable dynamism.

I am also particularly grateful to Dr. L.O. Gyekye of the Institute 
of Statistical, Social & Economic Research (ISSER), University of 
Ghana Legon, who was my primary source of inspiration.

The enthusiastic interest of my friends has also been a perpetual 
source of encouragement. Special recognition is given to Mr. James 
Kofi Appiah-Benhin who took time off his packed schedule to do the 
computer analysis of field data as well as typing of the first 
draft. I also feel particularly privileged to have been associated 
with Delasi Amable, Edzodzinam Tsikata, Kojo Arhinful and Richard 
Afari, all of the 1991 Postgraduate Sociology Class. Other 
indefatigable friends are Lawrence Y. Appiah-Baiden, Godfried Addo, 
A.A. Apeadu and Mr. A. Addo-Kwafo. To all of them, I am most 
grateful.

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



(iii)

The help and moral support given by Kofi Hushie, Essie and Soyoo- 
Dede, is also very much appreciated.

That this study could have been conducted at all is owed to the 
active participation and co-operation of the people of Pute. In 
particular, I wish to express my thankfulness to Mr. Sarbah - the 
Assemblyman, as well as the Chief and his elders. The willingness 
of mothers to co-operate even whilst attending to their children in 
most cases is highly commended.

My indebtedness is also owed to Pearl Adiki Puplampu - my special 
field assistant, Ruby Tetteh and Vera Agormeda who conducted most 
of the interviews.

Ms. Sylvia Osei, Mr. Jones Kpelie and Ms. Elizabeth Dadson, did a 
good job editing this work. To them I am deeply indebted.

The last but most important acknowledgement goes to the 
Imperceptible Force behind all of this "LE DIEU TOUT - PUISSANT".

Whilst acknowledging these diverse helps, I wish to state that I 
must be held responsible for any shortcomings found in this study.

M.H.

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



(iv)

ABSTRACT

The main thrust of this study is to investigate how childhood 
diarrhoea is managed at Pute - a rural community in Ghana. The 
study aims at eliciting local perceptions of childhood diarrhoea, 
including cause(s), consequences, and appropriate treatment, so as 
to unearth some of the social and cultural factors that may 
influence health-seeking behaviour in diarrhoeal episodes.

To understand fully folk concepts of diarrhoea and its treatment, 
it was necessary to describe the social and institutional setting 
within which illness episodes are managed in order to lay the basis 
for interpreting findings from the study. To this end, the social 
structure of Pute was outlined. Pute is a small Dangme-speaking 
rural community, located some 118 kilometers from central Accra. 
It is predominantly a fishing community in which descent and 
kinship groupings form the basis of social, economic, religious and 
political organisations. Contact with Western society has set in 
motion a process of change which is gradually promoting a breakdown 
of traditional cosmology.

In order to obtain in-depth information on folk nosologies of 
childhood diarrhoea and its treatment, three major methodological 
approaches were used. These are interviewing, focus group 
discussions and observation.

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



The conceptual framework that was used for organising field data on 
ethnomedical models of diarrhoeal illness is a cultural 
construction that establishes a web of relationship among social 
factors, illness experience, help-seeking and outcome.

The study has shown that folk classificatory systems for diarrhoea 
based mainly on physical notions of etiology determined to a very 
large extent therapeutic choices and hence help-seeking patterns. 
In particular, it has been shown that, the interpretations of 
specific diarrhoeal illness episodes, and specific health-seeking 
actions of mothers were not merely shaped by signs and symptoms, 
and that a wide range of factors enter into the establishment of 
illness identification and health-care decision-making. These 
include classification of a diarrhoeal ailment, perceived 
seriousness, availability of regimens and efficacy of treatment, 
all of which were found to be deeply rooted in ethnomedical models 
of diarrhoeal illness and its treatment. As a consequence, it has 
been found that, the widespread use of pharmaceuticals, especially 
antibiotics in the treatment of childhood diarrhoea should be 
considered a product of the local socio-cultural system in which 
illness episodes occur.

To this end, it is being suggested that, programme planners for the 
control of diarrhoeal diseases need to take cognisance of popular 
health culture and home care behaviour in rural settings such as 
Pute, if the promotion of ORT (including ORS) as the most effective 
modern approach to the treatment of most childhood diarrhoeas is to 
become effective.

(V)
University of Ghana     http://ugspace.ug.edu.gh



(Vi)

This thesis is in five main parts with sub-divisions under each 
chapter.

Chapter one, which sets out the introduction outlines the problem, 
objectives, the conceptual framework, the methodological approach 
to the study, the usefulness of the study as well as its broad 
limitations.

Chapter two, is devoted to a review of literature related to
diarrhoeal illness management in different cultural contexts of the 
world.

Chapter three takes a look at the social organisation of the people 
under study and their general world-view.

Chapter four presents the main findings and interpretation of
analysed field data.

Chapter five is a summary of main findings, policy implications of
the study as well as suggested areas for further research.

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



TABLE OF CONTENTS PAGE

DECLARATION .......................................  i
ACKNOWLEDGEMENT ................................... ii
ABSTRACT ..........................................  iv
LIST OF TABLES .................................... xi
LIST OF FIGURES ................................... xiii
APPENDIX .........................................  xiv
CHAPTER ONE

INTRODUCTION ................................ 1
1.0 Statement of the Problem ....................  1
1.1 Objectives of the Study .........................  11
1.2 Conceptual Framework ............................  13
1.3 Key Concepts ................................  19
1.4 Methodology .................................  20

1.4.0 Field Setting ..............................  20
1.4.1 Data Collection Techniques ..................  21
1.4.2 Sampling .............................. 24
1.4.3 Pre-Testing ........................... 26
1.4.4 Procedure for Fieldwork ..............  27
1.4.5 Data Processing and Analysis .........  27
1.4.6 Field Problems Encountered ............  2 8

1.5 Implications of the Study ................... 28
1.6 Study Limitations ...........................  29
1.7 Outline of Thesis ........................... 30

(vii)

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



(viii)

CHAPTER TWO
REVIEW OF LITERATURE .........................  32

2.0 Introduction ................................. 32
2.1 Summary ......................................  49

CHAPTER THREE
THE SOCIAL STRUCTURE OF PUTE .................  51

3 . 0 Introduction ..................     . 51
3.1 Geographical Location .......................  51
3.2 Historical Links ............................  53
3.3 Popultion Structure .........................  54
3.4 Economic Activities .........................  55
3.5 Kinship and Family Life .....................  57
3.6 Political System ............................  66
3.7 Religion ....................................  67
3.8 Medicine ....................................  73
3.9 Social Change ...............................  75
3.10 Summary .....................................   80

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



CHAPTER FOUR
MANAGEMENT OF CHILDHOOD DIARRHOEA ...........  81

4.0 Introduction ................................. 81
4.1 Socio-demographic characteristics of the Respondents 81

4.1.0 Maternal Education ....................  81
4.1.1 Religious Background ..................  82
4.1.2 Occupational Background................  82
4.1.3 Maternal Age............................ 83
4.1.4 Parity ................................. 83

4.2 Illness Taxonomy ............................  85
4.3 Ethnoclassification of Diarrhoea ............ 87
4.4 Perceptions and Treatment of Diarrhoea ...... 96

4.4.0 Prevalence of Diarrhoea ...............  96
4.4.1 Recognition of Diarrhoea .............. 97
4.4.2 Feeding During Diarrhoea ..............  100
4.4.3 Fluid Intake during Diarrhoea ......... 104
4.4.4 Perceived Consequences of Diarrhoea .... 107
4.4.5 Perceptions of Dehydration ............ 109

4.5 Patterns of Help-Seeking and Treatment
in Diarrhoeal Episodes .......................  110

4.6 The Quest for Therapy in Diarrhoeal Episodes ... 116
4.7 The Context of Diarrhoeal Illness Management ... 12 0
4.8 Hierarchy of Resort in Diarrhoeal Episodes ...  124
4.9 Decision-Making in the Quest for Therapy .....  125
4.10 Oral Rehydration Therapy ......................  131

(ix)University of Ghana     http://ugspace.ug.edu.gh



4.11.1 Knowledge and Use of ORT ..............  132
4.11.2 Correlates of ORS Use .................  136
4.11.3 Education and Knowledge and Use of ORS. 137
4.11.4 Religion by Use of ORS.................  140
4.11.5 Parity and Use of ORS..................  141
4.11.6 Maternal Age and Use of ORS ...........  143
4.12 Summary ................................  145

CHAPTER FIVE

(x)

5.0 SUMMARY AND CONCLUSION 147

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



LIST OF TABLES
3.1 Population Distribution by Ages and Sexes.... 55
3.2 Distribution of Economically Active Population 56
3.3 School Attendance by Ages and Sexes .........  75
4.1 Level of Maternal Education .................  82
4.2 Religious Affiliation of Mothers ............  82
4.3 Occupational Distribution of Mothers ........ 83
4.4 Age Distribution of Mothers .................  83
4.5 Parity of Mothers ...........................  84
4.6 Childhood Diseases at Pute ..................  86
4.7 Folk Taxonomy of Diarrhoeal Illness .........  88
4.8 Mothers' Definition of Diarrhoea by Number

of Stools Observed......................  99
4.9 Signs and Symptoms Associated with Diarrhoea. 100
4.10 Mothers' conception of a child being more (or less)

hungry during diarrhoea ..................  102
4.11 Food Intake during diarrhoea ................  102
4.12 Mothers' concept of whether a child is more

(or less) thirsty during diarrhoea .......  105
4.13 Fluid intake during diarrhoea .....   106
4.14 Fluids given during a recent diarrhoeal episode 107
4.15 Perceptions of dehydration .................. 110
4.16 First line of Action ........................  112
4.17 Type of self-medication by traditional

treatments ................................. 113

(xi)

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



4.18 Type of self-medication by biomedical
treatments ....................................  113

4.19 Specific pharmaceuticals used in the
treatment of Diarrhoea ........................ 114

4.20 Second line of Action ............................ 115
4.21 Self-medication by biomedical treatments  116
4.22 Reasons for first action taken on recognition

of diarrhoea ..................................  12 6
4.23 Reasons for second action taken "when

diarrhoea did not Stop" ........................ 131
4.24 Knowledge and use of ORS by level of Maternal

Education .....................................  138
4.25 Significant correlations between socio-demographic

characteristics of mothers and use of ORS . 138
4.26 Use of ORS and Religious Background ..............  140
4.27 Parity by Use of ORS .............................  143
4.2 8 Age-group of Mothers by use of ORS ............... 144

(xii)

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



(xiii)

LIST OF FIGURES

1 Cultural Construction of Diarrhoeal Illness: 
Interrelationship of Social Factors, Illness 
Experience, Help-seeking and Outcome.......  17

2 Cultural Context of ORT ....................  18

3 Map of the Dangme East District Showing the 
Location of the Study Area - Pute......... 52

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



(xiv)

APPENDIX

A Interview Schedule ............................  160
B Guide Questions for Focus Group Discussions .... 170
C References .....................................  172

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



CHAPTER 1
1

INTRODUCTION
1.0 Statement of the Problem
The prime concern of this study is to investigate the social and 
cultural contexts in which childhood diarrhoea is managed in rural 
Ghana. Pute, a rural community in the Dangme East District of the 
Greater Accra Region of Ghana has been chosen as a case-study. 
This study is devoted to understanding folk concepts of childhood 
diarrhoea, including local interpretations of cause(s), course and 
treatment, and how these influence health-seeking behaviour in 
diarrhoeal episodes.

Diarrhoeal diseases pose a major worldwide health problem, and are 
frequently associated with poverty, malnutrition and infection, 
killing between 5 and 7 million people each year. It is also 
estimated that about one billion episodes of diarrhoea occur each 
year in the 3 regions of Africa, Asia and Latin America, resulting 
in approximately 5 million deaths of children less than 5 years of 
age (Snyder and Merson 1982, WHO 1990) . Diarrhoea is also known to 
complicate several other diseases; the severity of measles, 
malnutrition and malaria is heightened by associated attacks of 
diarrhoea. (Scrimschaw et al. 1966; Martorell et al. 1975; Rowland 
et al. 1977; Mata 1978; Koster et al. 1981) . Many poverty-related 
factors are seen to be important in the prevalence of these 
conditions. Inadequate and overcrowded housing increases the 
transmission rates of many diseases. Insanitation, lack of

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



drainage facilities and inadequate supply of potable water results 
in a contaminated environment, creating ideal breeding sites for 
insect vectors. Furthermore, various factors including inadequate 
nutrition and illiteracy contribute to low levels of domestic 
hygiene and hence the prevalence of diarrhoeal diseases.

In Ghana, diarrhoeal diseases have been one of the serious public 
health problems. They are second only to malaria among children 
under five years seen at outpatient clinics. They account for 6.7 
percent of total attendances. It is also estimated that 10 percent 
of diarrhoeal diseases are fatal and that 70 percent of these 
fatalities are due to acute dehydration. In addition, 10 percent of 
deaths in hospitalised children under five years are attributable 
to diarrhoea alone or in association with measles, malaria and 
malnutrition (MOH 1992). Britwum et al. (1986) reported a mean
incidence of diarrhoeal diseases of 1.9 episodes per child per year 
in a rural Ghanaian community, with the highest incidence occurring 
in children between the ages of 7 and 12 months.

According to the Ghana Demographic Health Survey (GDHS 1988), the 
two week diarrhoeal incidence rate in children under five years was 
26 percent. In 1992 this figure was put at 30.2 percent (MOH 1992) . 
Afari et al. (1988) also showed an annual diarrhoeal incidence rate 
of 4.5 episodes per child per year. This amounts to approximately 
10 million episodes of diarrhoea per year in children under five.

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



The reduction of morbidity and mortality associated with childhood 
diseases in developed countries was achieved through comprehensive 
approaches. This entailed major improvements in living conditions 
through the provision of facilities such as pipe-borne water, safe 
sewerage disposal, higher standards of housing, health-care, home 
hygiene, improved nutrition and universal education for all 
segments of society. The Alma Ata Declaration of 1978 which 
proposed a "Health for All by the Year 2 000" through Primary Health 
Care (PHC) in developing countries was conceived in a similar vein. 
Yet such holistic approaches have often been considered unrealistic 
as a long-term prospect in most developing countries in view of the 
economic recession and shrinking health budgets that confront them. 
Consequently international health experts, aid agencies and 
governments are focusing on Selective Primary Health Care in which 
a few high risk groups are targeted with a few carefully selected, 
cost-effective interventions, though the verticality of current 
programmes has been the target of harsh criticisms (Gish 1982, 
Unger and Killingsworth 1986) . To implement this new strategy, the 
emphasis has been on the promotion of immunisation and Oral 
Rehydration Therapy (ORT) as simple projects in the attempt to 
improve child health.

International attention to the promotion of ORT for childhood 
diarrhoea began with the establishment of the World Health 
Organisation (WHO) Programme for the Control of Diarrhoeal Diseases 
(CDD) in 1978, as part of global commitment to Primary Health Care

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



(PHC). The objectives of the programme are to reduce diarrhoeal 
mortality in infants and children worldwide by one-half, and to 
reduce morbidity and malnutrition associated with diarrhoeal 
illness.

The Government of Ghana has since 1983 implemented a CDD programme 
as part of a nation-wide PHC programme. The programme emphasises 
five main strategies for the prevention and control of diarrhoea. 
These are

1. effective case management;
2. encouragement of breastfeeding, weaning practices 

and nutrition;
3. proper environmental sanitation;
4. provision of good drinking water and,
5. immunisation of children against measles.

The focus of this programme is the promotion of ORT which has now 
become the mainstay of the treatment for life-threatening 
diarrhoeal diseases both epidemic cholera and childhood
diarrhoeas. ORT is the use of appropriate fluids including Oral 
Rehydration Salts (ORS) as the most effective treatment for 
patients who are suffering from dehydration due to diarrhoea.

Under the multilateral sponsorship of the United States Agency for 
International Development (USAID), the United Nations Children's 
Fund (UNICEF), the MOH of Ghana and a local private drug

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



manufacturing firm (DANAFCO), the production and marketing of ORS 
packets was formally launched in April 1988 to promote home 
management of childhood diarrhoea through oral rehydration. 
Hitherto, ORS packets were being donated by UNICEF to the MOH for 
free distribution through its service outlets. ORS packets 
approved by WHO and UNICEF contain:

3.5g sodium chloride,
1.5g potassium chloride,
2.9g trisodium citrate, and 
20g anhydrous glucose.

Promotional strategies in Ghana include public education, 
commercial advertising and sales and use of mass-media. Currently 
the consumer price per packet is about £120 (US$0.13).

The focus of this programme has been home treatment of acute 
diarrhoea with oral solutions of packaged rehydration salts (ORS) 
or household sugar and salt solution (SSS) to replace body fluids 
and prevent life-threatening dehydration.

The introduction of CDD programme activities and other diarrhoea- 
related intervention programmes in many developing countries since 
the early 1980s has led to considerable increase in access to ORS 
and use of ORT at the global level (WHO/CDD 1990) . ORS is being 
produced in about half of the developing countries. Surveys 
conducted in a number of countries have indicated that on the

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



average, the hospital admission rate for diarrhoea dropped by 61 
percent after the introduction of ORT and the average case fatality 
rate was reduced by 71 percent (Cleason and Merson 1990).

Since the introduction of the CDD programme in Ghana, two impact 
assessments of it have taken place in 1989 and 1992 respectively. 
The 1992 survey revealed among others that awareness of ORS among 
mothers had increased from 56 percent in 1989 to 80 percent in 
1992 . Access to ORS had also increased only 16 percent of mothers 
had to travel more than 5 kilometres to obtain ORS as compared to 
25 percent in 1989. Also treatment practices had improved as 31 
percent of mothers gave ORT in 1992 as compared to 19 percent in 
1989 (MOH 1989, 1992) .

Despite the progress made, appropriate case management of childhood 
diarrhoea tends to be the exception rather than the rule and as a 
result, CDD programmes have not been able to achieve fully their 
set goals and objectives. It is the view of the present study that 
programme planners for diarrhoeal diseases have not adequately 
considered the implications of indigenous health beliefs and 
practices surrounding diarrhoeal illness in rural communities for 
their programme work. During the past decade however, it has 
become increasingly recognised that the ultimate control of 
diarrhoeal diseases depends on a comprehensive understanding of 
local beliefs and practices that relate positively or negatively to 
its transmission (Elmendorf and Isely 1983, Nations 1982, de Zoysa

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



1984) . Noting the importance of the mother in childhood diarrhoeal 
episodes WHO observed: "There is an urgent need to understand her 
present attitudes, perceptions and practices regarding diarrhoea as 
well as those of other community workers" (WHO 1982) . In most 
cases, the tendency has been for research funds to be allocated to 
more conventional areas of disease control such as epidemiology, 
biomedical research and health economics, while cultural factors 
are to a very large extent neglected.

Successful community-based programmes for the control of diarrhoeal 
diseases depend on mothers or other caretakers knowing how to 
manage diarrhoea with food and fluids and recognising when children 
need treatment by a health worker. In the quest for therapy, 
families may resort to various approaches both modern and 
traditional health regimens. The choice of approach however depends 
on the ethnoclassification of diarrhoea, what is seen as its cause 
and the availability of health care (Paredes 1992) .

To improve health, public health programmes often concentrate on 
strategies for convincing people to comply with professional 
recommendations. In the case of diarrhoea, the emphasis has been on 
the promotion of ORT - the discovery of which is regarded as 
"potentially the most important medical advance in this century" 
(WHO 1991).

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



Although pragmatic by biomedical standards, these recommendations 
may seem "irrational" in the community because they are at odds 
with local beliefs and practices. Health professionals who 
understand local perceptions of diarrhoeal illness are better able 
to communicate with the people they serve. They can understand how 
puzzling biomedical explanations might sound in the community and 
they can explain recommendations for prevention and treatment in a 
manner that makes them acceptable within the context of local 
beliefs and practices.

Moreover, certain characteristics of diarrhoea management (ie ORT 
and dietary practices) make cultural knowledge particularly 
relevant. These include:-

1. the fact that the management of diarrhoea is home- 
based and controlled by the child's mother or other 
caretakers as opposed to services located outside 
the household;

2. the fact that it is focused on a specific category 
of maladies for which local terminology and 
explanatory models exhibit extreme diversity; and,

3 . the fact that existing traditional therapies for 
digestive disorders generally and diarrhoea in 
particular are everywhere well developed and 
firmly well rooted in ethnomedical systems. Thus 
diarrhoea treatment lends itself to contextual 
analysis.

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



Although various surveys provide a dramatic confirmation that 
diarrhoeal disease is a major health problem in Ghana (Britwum et 
al. 1986; Afari et al. 1988; GDHS 1988), they necessarily focus to 
a large extent on broad socio-demographic data and general 
morbidity and mortality patterns without delving into the complex 
complementary area of local beliefs and practices associated with 
childhood diarrhoea. There are a few studies touching on this area 
of enquiry (Adjei 1988, Abu 1988). Yet these are scattered and 
largely unpublished.

The main task of this study is to understand folk concepts of 
diarrhoea and its treatment. Attempts have been made to raise 
culturally relevant issues that might be missed by macro surveys 
but might have important implications for community-based health 
programmes for the control of diarrhoeal diseases.

Mothers' ethnomedical models of childhood diarrhoeal illness and 
concepts of appropriate treatment have been examined in terms of 
their most receptive and resistant points to modern approaches to 
the prevention and treatment of diarrhoea.

The analysis made in this study is based on the view that, the 
socio-cultural environment (including political, economic and 
psychological aspects) is only one among a multiplicity of factors 
influencing illness behaviour and for that matter health-seeking 
behaviour. Other factors are biological and physical. Cultural

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



beliefs and attitudes have been found to be very important in 
shaping response to diseases (Maina-Ahlberg 1984) . Fosu (1984) 
observed that, ethnomedical concepts of disease causation affected 
therapeutic choices in Ghana.

Although diarrhoea is not found among children alone, this study
focuses on infant and childhood diarrhoea for two important
reasons. Firstly, it is widely recognised that acute diarrhoeal
diseases are a leading cause of morbidity and mortality in children
under five years of age especially in developing countries.
Published studies in Africa, Asia and Latin America have shown that
children under one year of age may have on the average 5 episodes
of diarrhoea a year while children between 1 and 2 years old may
even have more episodes than their counterparts aged between 3 and
4 years. (Snyder and Merson 1982, Parker 1984, WHO 1988) . This
fact has been documented in Ghana by Afari et al. (1988); Britwum
et al. (1986) and the GDHS (1988) . Secondly, most of the
diarrhoeal diseases which kill children can be averted through the
appropriate use of ORT. The relevance of ORT in diarrhoeal
diseases is best expressed in the summary of an international
conference on ORT:

"The experience reported at this conference unequivocally 
confirms that, oral rehydration therapy can reduce 
mortality, sometimes drastically in communities, clinics 
and hospitals; promote child growth and sound nutrition; 
lessen the morbidity burden; reduce hospitalisation 
attendance, duration of stay and cost; and generate 
auxilliary benefits such as minimising the indiscriminate 
use of ineffective or harmful drugs (Chen 1983)".

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



1.1 Objectives of the Study
In pursuing the research problem, the study focuses on the 

following issues: -
1 . to highlight some of the traditional health beliefs and 

practices associated with childhood diarrhoea at Pute;
2. to show the extent to which these beliefs and practices 

influence the choice of both modern and traditional health 
care regimens for the treatment of diarrhoea;

3. to make appropriate recommendations based on the findings of 
the study;
Specifically the following questions are framed to guide the 
study:-

1. What is the folk classification of childhood diarrhoeal 
disease with respect to types and causes at Pute?

2. What features of a diarrhoeal ailment do mothers- consider 
important in the process of diagnosis?

3. What indigenous beliefs exist about food and fluid intake 
during diarrhoea?

4. What are the perceived consequences of childhood diarrhoea?
5. What are the perceptions of dehydration due to diarrhoea?
6. To what extent are local concepts of childhood diarrhoea 

related to specific health care regimens that are used in 
diarrhoeal episodes and hence patterns of help-seeking?

7. What are the factors underlying the decision to use or not to 
use specific health care regimens in the quest for therapy?

8. Do mothers at Pute know of ORS? What do they think it is used 
for? Have they ever used it? What are the reasons for non-use 
if they know or have heard of it before?

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



The following hypotheses are framed to guide the discussions 
in later chapters:-

(1) The choice of modern scientific approaches to the treatment of 
childhood diarrhoea depends on folk nosologies, etiologies and 
consequences as well as treatment for various diarrhoeal 
illness categories.

(2) The use or non-use of ORT (including ORS) is a function of the 
educational level of mothers or other caretakers of the child. 
Educated mothers are more likely to use ORT as a remedy for 
the prevention and treatment of diarrhoea than uneducated 
mothers.

An attempt will be made to test these hypotheses by analysing the 
beliefs and practices of the people of Pute regarding diarrhoeal 
illness and its treatment.

A study of this nature requires a setting that will permit an in- 
depth examination of behaviours during illness episodes. In this 
respect, Pute1, a Dangme rural community in the Greater Accra 
Region of Ghana was chosen for the case-study. An important 
characteristic feature of Pute that makes it suitable for an 
investigation of this nature is that it has a high incidence of 
diarrhoeal illness and is therefore in great need of diarrhoeal 
control efforts. Elsewhere in this study, other reasons have been 
outlined to show why this particular community was chosen.

Chapter Three gives a detailed description of this 
rural community

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



1.2 Conceptual Framework
A study such as this and which is aimed at investigating the socio­
cultural context of diarrhoeal illness and treatment requires a 
cultural construction that establishes a web of relationship among 
social factors, illness experience, help seeking and outcome. To 
this effect, Weiss' (1988) proposed conceptual framework has been 
adapted for use. This framework is based on concepts developed by 
medical anthropologists in the contextual studies of diarrhoeal 
illness. Weiss places these concepts under four major headings as 
follows: patterns of distress; explanatory models/patterns of help- 
seeking and specific treatments.

Patterns of distress refer to the range of symptoms that the local 
people complain about and the ways they experience the effects of 
the illness. Traditional people may base their distress patterns on 
any one or more of the following features of the diarrhoeal 
illness:

1. quality of the diarrhoeal stool;
2. other signs and symptoms;
3. perceived seriousness;
4. perceived vulnerability to complications or other
kinds of distress,-
5. level of anxiety;
6. disturbed interpersonal relationships;
7. spiritual obligations, concerns and supports;
8. local ideas about illness that refer to diarrhoea as 

a primary or incidental feature.

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



In contrast, clinicians considering the signs and symptoms will 
diagnose a microscopic pathogen or describe tissue pathology or 
pathophysiology.

Explanatory models characterise the meaning people attribute to 
illness as they try to explain it. This concept specifies perceived 
causes of diarrhoeal illness - particularly physical, social, 
supernatural, humoral and other explanations - with reference to 
the underlying system of beliefs in different cultural settings. 
Different explanatory models specify the following causes for 
diarrhoeal illness in many cultures:
1. foods that are fatty, not cooked adequately, heavy, etc;
2. imbalance of heat and cold that may be associated with foods,

exposure to drafts or seasonal changes;
3. normal or poor quality breast milk;
4. physical factors, such as a fall or poor caretaking;
5. supernatural causes including possession of sorcery or evil

eye;
6. pollution from exposure to or inauspicious contact with

ritually impure persons or things;
7 - moral misbehaviour, including deeds of the sick person or a

sick child's parents, especially promiscuous sex and sexual 
intercourse or pregnancy whilst breastfeeeding;

8. natural consequence of milestones especially teething,
crawling and walking;

9. infection which may be associated with hygiene and sanitation 
(but which may be difficult to distinguish from ideas about 
pollution).

14University of Ghana     http://ugspace.ug.edu.gh



In every culture, especially traditional societies, people seek 
help for distress from various sources. Help providers are 
associated with modern or traditional medical systems, religious 
healers or community leaders or other local institutions. 
Preferences for medical help-seeking may vary according to features 
of the illness; socio-demographic characteristics of the individual 
family; the reputation, availability, and prior experiences with 
various sources of medical help or other factors.

Specific treatments refer to the diverse group of medical help 
providers in the community, knowledgeable relatives and others in 
the social network who may recommend an equally diverse array of 
treatments. They fall under the following broad categories:
1. adjusting diet and fluid intake or withholding foods or 

fluids;
2. changing breastfeeding routines;
3. cleansing the gastrointestinal tract with enemas, 

purgatives and emetics;
4. local herbal remedies;
5. ritual and devotional practices to promote spiritual 

healing;
6. other locally sanctioned interventions (eg. abdominal 

massage, manipulation of the soft palate and skin over 
fontanel, cutting gums for teething diarrhoea etc;

7. cosmopolitan medicines, including antimotility agents, 
adsorbents, and antibiotics;

8. ORT (including ORS).

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



As figures 1 and 2 (see overleaf) show, each of these sets of 
variables which together constitute the 'cultural construction of 
illness', represents a facet of the complex relationship between a 
disease and the cultural context in which it occurs.

These variable sets proposed by Weiss (1988) to characterise 
diarrhoeal illness-related beliefs and practices patterns of 
distress, explanatory models, help seeking and treatment practices 
provide an appropriate framework for organising field data on 
ethnomedical models of diarrhoeal illness in the community under 
study, and for generating hypotheses that relate perceptions and 
the experience of illness to relevant public health outcomes.

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



17

Social'' ?ontexM Patterns
ofdistress i

Humoral
Spiritual
Physical
Pollution
Milestones
Infection

Socioeconomic 
General education 
Health education 
Sanitation & hygiene 
Feeding & breastfeeding 
Health & nutrition 
Environment

Biomedical diagnosis 
Incidence 
Course 
Outcome

Explanatory
Models

Family
Community leaders 
Local healers 
Hospital clinic 
Private physician

Signs (e.g., stool quality) 
Symptoms
Perceived seriousness 
Features ol dehydration 
Designated (oik category

Fig. I. Cultural construction o f  diarrheal illness: interrelationship o f  social factors, illness experience, help
seeking and outcome.

Source : Weiss M. G. ,  Cultural Models of Diarrhoeal
Illness : Conceptual Framework and Review.
Soc. Sci, Med., Vol. 27, p p .6, 1988.

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



Fig. 2. C u l t u r a l  co n tex t o f  O R T.

Source: Weiss M .G., Cultural Models of 
Diarrhoeal Illness : Conseptual Framework and 
Review, Soc. Sci, Med., Vol. 27, No. 1 pp. 7, 
1988.

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



1.3 Key Concepts
It is appropriate that in a study of this nature the various
concepts used are defined in order to do away with any ambiguity 
in the mind of the reader. Specifying the meanings of the concepts 
will also facilitate the reading and understanding of the study. 
These concepts are therefore defined below.
1. DIARRHOEA

A brief survey of diarrhoeal diseases literature depicts
varying levels of heterogeneity in the use of the term 
diarrhoea. (Black et al. 1982). For the purposes of this
study however, WHO's recommended definition will be used. 
Diarrhoea is defined clinically as the excretion of three or 
more liquid or semi liquid stools during a 24 hour period. 
(WHO 1988) . Two main forms of diarrhoea are recognised by 
WHO: acute and persistent. Acute diarrhoea refers to episodes 
that last less than one week. Persistent diarrhoea on the 
other hand is an episode which starts acutely but which lasts 
at least 14 days (Dialogue on diarrhoea, No. 48 1992) .

2. TRADITIONAL MEDICINE -
This term is used here to describe health care that does not 
fall within the allopathic 'modern1 and scientific system. 
Under this system, there is no conceptual separation between 
natural and supernatural entities, although it has as a key 
feature, the use of magico-religious acts and concepts. As
Twumasi has noted in traditional medicine "tradition is
important not in the sense of empirical experience but through
its validation of the power of the unseen world" (Twumasi 1975

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



p.10). Traditional health care regimens used in the 
prevention and treatment of diarrhoea include local remedies 
such as herbs, enemas, purgatives, visiting faith healers, 
spiritualists (diviners or fetish priests) and traditional 
herbalists.

3. MODERN SCIENTIFIC OR ALLOPATHIC APPROACHES TO THE 
PREVENTION AND TREATMENT OF DIARRHOEA -
These include the use of home medication with pharmaceuticals, 
seeking advice and prescribed treatment from a 
pharmacist/chemical seller, health worker in government or 
private practice and the use of ORT (including ORS). 
Practitioners of these approaches have as an important 
underpinning a naturalistic world view of the causes of 
diseases and ill-health in general. As a result the 
establishment of causalities in supernatural terms is assumed 
to have no place in modern therapeutic choices.

1.4 METHODOLOGY
1.4.0 FIELD SETTING.
To investigate the central thesis that the choice of modern 
approaches to the treatment of childhood diarrhoea is a function of 
folk concepts of cause(s), consequences and treatment of various 
forms of diarrhoea, it became necessary to move into a rural area. 
The Dangme East District readily came to mind as the investigator 
was familiar with this area and besides, such familiarity has been 
found to enhance the quality of in-depth studies (Janzen 1978; 
Stone and Camp 1984; Bleek 1987) . In deciding on the particular

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



community to be studied however clinical records at the Ada Foah2 
Health Post were reviewed, and Pute was found to be one of the 
communities in the district with a high incidence of diarrhoea, 
which therefore made it a suitable setting for the study. Apart 
from these considerations, it was also felt that the chosen 
community should have identifiable rural features to make it easily 
discernible from an urban community. In this respect, lack of 
pipe-borne water supply, electricity, good accessible roads, and 
its smallness (it has a population of less than 5000) were 
considered typical rural features.

Furthermore, this micro-approach was adopted for two other 
important reasons; time and financial constraints as well as to 
allow for a deeper study of beliefs and practices surrounding 
childhood diarrhoeal illness.

1.4.1 Data Collection Techniques
The data upon which this study is based were gathered mainly from 
mothers with children under age five residing in Pute.
Aware of the limitations of using standard survey methods to elicit 
valid information on health beliefs among a rural community, three 
major complementary methods were used in this study.

The primary method of data collection was interviewing using a 
schedule which was designed following a review of literature 
related to the management of childhood diarrhoea in various

2 Ada Foah is the administrative capital of the Dangme 
East District.

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



cultural contexts of the world. The schedule consisted of both 
close-ended and open-ended questions. The open-ended questions 
made it possible to elicit free and detailed responses as well as 
recording of information spontaneously. Topics covered in the 
interview schedule include socio-demographic characteristics of 
respondents, beliefs, perceptions and attitudes to diarrhoea with 
particular reference to types, their causes and treatment options; 
case management of diarrhoea which also included questions 
related to prevalence of diarrhoea, food and fluid practices and 
knowledge and use of ORS/SSS. Taking into consideration the 
problems of recall usually encountered in household morbidity 
surveys, questions regarding case management were framed around a 
point prevalent measure (the proportion of children under age five 
whose mothers report that they have had diarrhoea in the 24 hours 
preceding the interview) and a period prevalent measure (the 
proportion with diarrhoea in the two weeks preceding the 
interview). Where the diarrhoea occurred in the remote past, the 
technique used to facilitate recall was to centre questioning 
around the "last diarrhoeal episode". Thus responses were linked 
to a previous experience of diarrhoea usually recent in time 
rather than an abstract event.

A major shortcoming of data from the interview schedule is that 
they tend to reflect reported behaviour in the sense that there may 
be variations in what people say they do and what they actually do 
(Bentley 1987) . In order to minimise the distortion between 
'reported' behaviour and 'actual' behaviour, this first approach of

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



interviewing was combined with observation and informal questioning 
of mothers. To this end, actual diarrhoeal episodes were followed 
through to recovery. Where possible, visits were made daily or 
every other day until the diarrhoea subsided. Conducting 
interviews among families in which the child was ill with diarrhoea 
made it possible to observe and to sympathetically comprehend 
interactions among mother, child and other caretakers and thereby 
focusing the behaviours of interest first hand. In all, the 
observation technique which has been found useful in the study of 
small-scale largely illiterate communities, provided the necessary 
social and cultural background for interpreting data derived from 
the interview schedules.

As a further validity check on self-reported information, three 
focus group discussions of mothers randomly selected from three 
different age groups 15-24, 25-34 and 35 and above were conducted. 
Age was used as the main criterion for making the groups 
homogeneous. This ensured that mothers felt at ease and spoke. 
All participants had been previously interviewed. The investigator 
served as the main facilitator whilst two other members of the 
research team served as recorders. The questions for the focus 
group discussions were treated as an outline of discussion which 
only provided a general direction. Hence, they were not rigidly 
followed. To initiate a discussion on the issues of interest, the 
group was led to talk about general topics to build up mutual 
trust. It was only when this had been achieved and the discussion 
was running smoothly and all members were actively participating 
that issues related to the management of childhood diarrhoea were

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



raised. It was this part of the discussion that was recorded on a 
voice-activated portable tape-recorder. Sessions which were 
conducted in Dangme3 lasted between 60-80 minutes and they were 
held in a classroom at the village's primary school. Each group 
consisted of eight members. At the end of the group sessions, the 
recordings were listened to by the main facilitator and reports 
were written on the salient points.

Furthermore, the investigator held informal interviews with some 
knowledgeable people like the local chief, fetish priest/priestess, 
herbalists, the local Assemblyman and a Chemical Seller in the 
village. They were all relied on to describe and to report on the 
culture of the people at Pute.

Besides, secondary data on the anthropological studies of 
diarrhoeal illness (mainly from journals) in different cultural 
contexts were utilised where necessary.

1.4.2 Sampling
In order to obtain relatively unbiased results at minimum cost, the 
following procedure was used in selecting mothers with children 
under age five for questioning. First it was important to assess 
the population universe of mothers in the community having children 
under age five. In the absence of current population data on the 
village, a house to house head count of these mothers was carried 
out together with the number of children under age five that lived

Dangme is the local language of the people of Ada.

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



in each house. The head count which was done in May 1993 revealed 
that there were approximately 291 mothers who had a total number of 
383 children under age five and lived in 151 houses. There were 
however about 9 houses without children under age five which have 
therefore been excluded from the sampling universe. On the average 
there were about two mothers with 1.3 children in each house. In 
order to obtain a sample size representative of Pute community as 
a whole, the family was used as the sampling unit. It is defined 
here as a mother plus any children she had borne who are living 
with her in the same household. A significant proportion of 
households contained multiple nuclear families. Questionnaires 
were administered to one family in each household. The mode of 
interaction in this community made it unnecessary for our purpose 
to interview more than one family per household since the 
interviews were not private other family members who were around 
indirectly participated. In all 143 mothers from all houses having 
children under five were interviewed. This represents about 50 
percent of mothers with children under age five residing in Pute 
and is therefore expected to provide a good representative sample. 
In selecting mothers, priority was given to those whose children 
were currently experiencing diarrhoea or had had the most recent 
episode. It was felt that household behaviour for an on-going 
diarrhoea will probably be reported more accurately than for past 
or future episodes.

Furthermore, in order to obtain a representative number of 
diarrhoeal episodes to be followed, it was decided that half of all 
new episodes recorded during household interviews in a particular

25
University of Ghana     http://ugspace.ug.edu.gh



day will be randomly selected for study. Out of the 47 on-going 
diarrhoeal episodes that were recorded over the period of the 
study, 23 were monitored.

Given the time constraints facing the investigator, three female 
interviewers were recruited to help in collecting information from 
mothers. They were all sixth formers and Dangme-speaking. They 
were trained in the conduct of focus group discussions, 
interviewing and qualitative research techniques using role playing 
exercises for one week. The goal of this training was to enable 
interviewers to become familiar with the aims of the study and the 
purpose and meaning of each question. Since the interview schedule 
was in English, appropriate translations of the questions into 
Dangme were carried out to ensure that questions were asked 
uniformly by all. All the interviewers stayed at Pute throughout 
the survey. This enabled them to interview women engaged in fish- 
related activities and traders who left home very early and 
returned home very late in the evening.

1.4.3 Pre-testing
The interview schedules and the guide questions for the focus group 
discussions were pre-tested at Ocanseykope and Totimekope - all 
rural communities in the Dangme East District. Pre-testing the 
interview schedules provided the investigator with the means for 
detecting and solving unforeseen problems not previously 
anticipated. These included rephrasing, re-wording and changing 
the sequence of questions. It also showed the need for new 
questions and the elimination of some old ones.

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



1.4.4 Procedure for Fieldwork 27 
The investigator's first visit to Pute to begin the study was in 
May 1993 in which a census of mothers having children under age 
five was conducted. During this first visit, the chief and his 
elders caused a "gong gong" to be beaten at a fee of <?800. This 
was to "announce the arrival in their midst of a young man from 
Legon-Accra and three ladies from Ada-Foah to study the health 
problems of our children. They will be visiting your houses to ask 
you questions so we entreat you to give them your maximum co­
operation" . The actual survey on the management of childhood 
diarrhoea could not take place immediately after the first survey 
due to interruptions by funeral celebrations. Hence it was 
conducted in the latter part of June 1993. The findings are 
reported in Chapter four.

1.4.5 Data Processing and Analysis
In presenting the data collected, statistical analysis of the study 
is limited to percentage frequency distributions, cross-tabulations 
and significant correlations between some important variables which 
have been used for illustrative purposes in an effort to 
communicate the results of the study in a simple non-technical 
manner. All close-ended questions were pre-coded whilst answers to 
open-ended questions were listed, placed into meaningful categories 
and assigned codes after the survey for subsequent analysis. The 
analysis of data was carried out by a computer expert at the Health 
Social Sciences Research Unit of the Institute of Statistical, 
Social and Economic Research University of Ghana - Legon. The 
software used for the analysis is the Statistical Analysis System 

(SAS).

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



1.4.6 Field Problems Encountered
Some of the major problems encountered in the study were:-
1. the feeling of some mothers that the investigation was

irrelevant to their pressing health needs. Hence their 
reluctance to provide complete answers. Whenever such a
problem arose, the purpose of the study was re-echoed in a
polite and convincing manner which made most mothers to co­
operate ;

2. the absence of some mothers, which led to postponement of 
interviews to be carried out at an appointed time - usually in 
the early hours of the morning or late in the evening;

3. the problem of different meanings between interviewers and
respondents: granted that the majority of mothers were
illiterates, it was quite difficult to translate exactly their 
thoughts or views into English without slight changes, whilst 
recording responses. This problem was minimised by listening 
attentively and repeating what had been written down to them 
as a cross-check on information given;

4. and finally the difficulty of conducting interviews privately. 
Most of the time, relatives or neighbours often refused to 
leave the scene of the interviews. Hence only a few were not
conducted under public scrutiny. A lot of patience and tact
on the part of interviewers became a sine qua non for the
interview process to go on.

1.5 Implications of the Study
The understanding of household management of childhood diarrhoea
has one important applied purpose - it would help to obtain

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



information about the most receptive and resistant points to 
intervention programmes for diarrhoeal diseases. This will enable 
health planners to design appropriate programmes to communicate 
needed educational information either to reinforce adaptive 
behaviours or modify maladaptive practices in the management of 
childhood diarrhoea. In the long run this will help them to promote 
improved and acceptable health care for rural people.

It is also hoped that the study would inspire more rigorous micro­
studies of diarrhoea-related behaviour as well as those of other 
diseases that demand cultural knowledge of specific populations in 
other ethno-geographic regions of Ghana.

Finally this study will strengthen empirically the theory of 
health-seeking behaviour in Ghana.

1.6 Study Limitations
The following limitations of the study are notable.
First, due to the small number of diarrhoeal episodes that were 
followed (23) it is not known whether the management of childhood 
diarrhoea as a whole was captured in full.

Furthermore, there is the important issue of the extent to which 
the findings can be generalised. This limitation arises mainly from 
the decision to confine the study to only one rural community 
although there exists broad geographical, ecological, demographic, 
cultural and linguistic variability in Ghana. However, the central 
questions which this study addresses are not peculiar to the

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



specific case chosen but bears on people of other ethno-geographic 
areas in Ghana as well as other communities in the district if not 
in the whole country.

1- 7 Outline of Thesis
This thesis comprises five main chapters with sub-divisions under 
each chapter.

Chapter one (1) , which is the introduction sets forth the problem, 
objectives, conceptual framework, the methodological approach to 
the study, the practical contributions that the study can offer as 
well as its broad limitations.

Chapter two(2) deals with a review of literature on the management 
of childhood diarrhoeal illness in different cultural environments.

Chapter three (3) is devoted to a description of the research site 
with particular reference to the social organisation of the people, 
their way of life and general world-view.

Chapter four (4) is concerned with a presentation of findings and 
the analysis of the data collected.

Chapter five (5) is a summary of findings, conclusions and 
implications of the study for public health programmes for the 
control of diarrhoeal diseases as well as suggested areas for 
further research.

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



Copies of interview schedules used in the study, a list of 
references consulted, as well as other related information are 
found in the appendix.

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



CHAPTER TWO 
REVIEW OF LITERATURE

2.0 Introduction
In this chapter, researches on childhood diarrhoeal diseases and 
other related issues in different cultural settings will be 
reviewed. The studies that are focused on are mainly 
anthropological and epidemiological in nature. Apart from 
providing a general background to the study as a whole, the review 
will help to establish more firmly the rationale of the study, and 
lay the basis for understanding discussions in subsequent chapters.

The development of interest in anthropological approaches to the 
study of diarrhoeal illness has been traced to three main factors: 
PHC development, methodological contribution and theoretical issues 
(Coreil and Mull 1988) .

Over the years, efforts to implement PHC programmes in many 
developing countries have led to an increased awareness of the 
importance of cultural information about the constituent population 
served by specific programmes. (Foster 1984). There is greater 
recognition that many problems encountered in the promotion of 
community-based PHC programmes emanate from lack of information on 
local health beliefs and practices, insensitivity to cultural 
factors as well as other issues that require detailed knowledge 
about cultural dispositions and local ecological limitations.

With regard to methodological contributions, anthropological 
perspectives have been seen to have moved beyond the usual practice

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



of merely providing a 'background view' of the cultural context to 
an in-depth cultural analysis and interpretation of specific 
illness nosologies and care practices that are likely to influence 
the adoption of new health behaviours.

Finally, one of the most important theoretical issues that has been 
found to be associated with the development of interest in 
anthropological diarrhoeal research has been the fact that for the 
first time in history, ORT and dietary management as therapeutic 
health practices (rather than preventive), based primarily in the 
home were being promoted on a massive scale worldwide by 
international health organisations. Granted that the home 
treatment of diarrhoea raises a lot of issues which anthropology is 
particularly suited to address, it is not surprising that, since 
1980 when ORT has been the dominant focus of international health 
programmes, many researchers have within the past decade 
investigated maternal beliefs, attitudes and behaviour related to 
diarrhoeal diseases (mainly childhood diarrhoea and cholera, ORT, 
dietary management, breastfeeding and other pertinent issues in 
developing countries). These studies have constituted part of 
public health efforts to improve home management of childhood 
diarrhoea.

Bentley (1988) conducted an in-depth anthropological study of 
childhood diarrhoea in three (3) rural communities in a North 
Indian state of Haryana to show variations in the household 
management of childhood diarrhoea. Using ethnographic techniques, 
structured interviews and participant observation, the study

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



revealed that mothers in the three study villages had a clearly 
defined indigenous classification system for diarrhoea. Diarrhoea 
was referred to as 'dust' and could be categorised into between two
(2) and five (5) types that were related to either a physical or 
spiritual characteristics. The commonly listed types were 'Khooni- 
dust' ('bloody diarrhoea'), 'pani dust' ('watery diarrhoea'), 
'phate-phate' ('bits and pieces'), 'hare dust' ('green diarrhoea') 
and 'pila dust' ('yellow diarrhoea').

Mothers attributed the causes of diarrhoea to hot weather (66 
percent) , bad food (33 percent) , overeating: (30 percent) teething 
(24 percent), cold weather (23 percent), 'hot' food (16 percent), 
hot breastmilk (15 percent), evil eye (12 percent) and dirty water 
or germs (4 percent).

A great majority of mothers in Bentley's study who believed 
diarrhoea could be treated preferred 'doctors' treatment (86 
percent). Some of the mothers (3 percent) felt that home treatment 
alone was sufficient; less than 5 percent said both 'doctor' and 
home treatment should be done and less than 1 percent believed 
exorcism should be sought as a treatment choice. In the fifty (50) 
diarrhoeal episodes that were followed, medical treatments were 
used in 62 percent of cases and the most striking finding was the 
widespread use of antibiotics and in many cases the use of multiple 
antibiotics during a single episode. In the sample of episodes 
followed, there was no relationship between the presence of blood 
and whether or not an antibiotic was given.

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



One important aspect of Bentley's (1988) study was to determine the 
extent to which mothers were withholding food and fluids as is 
widely reported in professional literature. Although the survey 
data depicted traces of this 'common view' the more commonly 
observed pattern was a shift in diet towards foods mothers 
considered "helpful" away from foods they considered "harmful". On 
the whole however, normal dietary patterns were maintained whilst
mothers continued to breastfeed normally and did not decrease
fluids during diarrhoea.

In addition, acceptance and sustained use of ORT was found to be 
inversely related to an understanding of the function of ORT.
Eighty-one percent of mothers who had previously used ORT but do
not intend to use it again believed it did not "stop the 
diarrhoea". Mothers tended to attribute anti-diarrhoeal qualities 
to ORT and did not understand what it was meant to do - replace 
body fluids lost from stool output.

Some of Bentley's findings of practical relevance for diarrhoea 
intervention were that ORT should be promoted as a therapy: for the 
replacement of lost fluids. Also messages that are meant to change 
behaviours should be based on the target group's perceptions.

Green's study (1985) of beliefs and practices associated with 
childhood diarrhoea in Swaziland relied primarily on traditional 
healers and survey respondents as informants. His study revealed 
an ethnoclassification of diarrhoea into 3 main types 'umsheko', 
'kuhabula' and 'umphezulu'. Although their signs and symptoms

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



tended to overlap somewhat, they were viewed as separate syndromes 
with different causes, treatments and potentials for prevention. 
'Umsheko' is believed to be natural, of a brief duration, non­
dehydrating and caused by diet, teething and mild fevers. 
Diarrhoea initially regarded as 'umsheko' can later be regarded as 
'kuhabula' or 'umphezulu' if it persists and if symptoms such as 
sunken fontanel appear.

Among others, it was discovered that enemas were used in the 
treatment of two types of more serious diarrhoea ('kuhabula', 
'umphezulu') thereby contributing to dehydration. A kind of
fumigation was also used to treat 'kuhabula'. Herbal teas could be 
used for any illness associated with diarrhoea. Those used for 
simple 'umsheko' are believed to have stool-solidifying 
characteristics.

It was observed that most children with diarrhoea were taken to the 
clinic only after home treatments and those of indigenous health 
practitioners had proven futile, by which time a child may be 
dehydrated. In discussing the implications of the study for ORT 
programmes in the context of Swazi culture,. Green identified two 
important groups which should be made primary 'targets' for health 
promotion: traditional healers and mothers. Whilst the former are 
seen to be the opinion leaders in all matters of health including 
diarrhoeal diseases, the latter are decision-makers regarding 
therapy for childhood diarrhoea.

In an evaluative study of a diarrhoeal disease prevention and

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



control programme in Honduras, Kendall and his collaborators (1984) 
demonstrated the extent to which beliefs and practices surrounding 
diarrhoeal illness partially shaped the promotion of ORT. 
Interviews with key informants concerning diarrhoeal etiology and 
treatment revealed a number of folk illnesses associated with 
diarrhoeal symptoms believed to be causes of diarrhoea. These were 
'empacho', 'ojo' (evil eye), 'caiada de moltera1 (fallen fontanel) 
and 'lombrices1 (worms) . Episodes of diarrhoea were believed to be 
caused by these folk illnesses especially when the diarrhoea is 
severe and refractive. 'Empacho' was believed to be a painful 
condition of the intestinal tract characterised by explosive 
evacuations and flatulence. It was believed to be caused by eating 
the wrong kinds of food, or a combination of foods improperly 
balanced along notions of hot and cold foods; eating at improper 
times or missing a meal or eating foods that were improperly cooked 
or raw. To treat 'empacho' the body needs to be massaged by a 
masseur or masseuse and a purgative administered to clean the gut. 
Evil eye or 'ojo' which produces fever was felt to be the result of 
malicious penetrant rays. It is treated by bundling as for any 
fever and spraying of the skin with a number of liquids. 
Purgatives were not used and the belief was that evil eye does not 
contaminate the gut. 'Fallen fontanels' were believed to be caused 
by improper maternal handling of an infant. It was also held that 
the tissue beneath the fontanel were incompletely developed and 
that they could fall. Treatment entailed pushing up on the roof of 
the mouth and tapping on the heels of the inverted child and/or 
sucking on the fontanel.

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



'Lombrices' (worms) were considered a normal symbiote of the gut 
but caused illness when disturbed. It needs treatment by a 
purgative otherwise they could cause "worm fever" which was 
believed to be a fatal illness. Worm related diarrhoea- was 
generally believed to be benign and therefore needed no treatment 
except when it became persistent. Based on their findings, Kendall 
et al. recommended that the use of ORT may not be resisted in cases 
of diarrhoea attributed to 'ojo' or 'caiada'. But for 'empacho' 
which required the administration of a purgative, it was felt that 
mothers would resist the use of ORT for the treatment of this 
disease. A natural consequence of their findings would have been 
the promotion of ORT as a purgative, or at least as a specific 
treatment for 'empacho' . This promotional strategy for ORT was not 
approved for two reasons. Firstly, it was rejected by the Honduran 
cosmopolitan medical community and, secondly, only few mothers 
reported 'empacho' as a cause of diarrhoea.

Mull and Mull (1988) studied diarrhoea-related traditional health 
beliefs and practices among women in rural Sind-Pakistan. A total 
of 57 women with children under age 5 were interviewed to obtain 
information on the prevalence and perceived seriousness of 
diarrhoea, its causes, the ways in which it was treated; including 
ORT as well as the range of indigenous health practitioners who 
were consulted for diarrhoea and for diseases in general. They 
found that mothers' perceptions of the causes of child diarrhoea 
were closely linked to a concept of 'heat' . As used by these 
mothers, 'heat' did not normally refer to measurable temperatures 
but to a hot quality believed to be inherent in certain foods or

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



bodily states. To this end, mothers in Mull and Mull's study 
tended to consider substances perceived as 'cooling' the intestines 
as effective anti-diarrhoeal remedies. Such cooling agents 
included yoghurt, lime juice mixed with water, 'rice-water' and 
various herbal infusions. Of particular relevance to ORT 
acceptance was the fact that 72 percent of the fifty-seven (57). 
women interviewed stated that they had used a form of 'sugar water' 
and that it was useful because of it's 'cooling' effects on the 
body. Although the salt content of this solution could not be 
established, a few mothers (9 percent) held that salt worsens or 
causes diarrhoea. Nevertheless, the evidence from the study 
suggested that mothers in this area of Pakistan will not reject ORT 
outrightly simply because of it's sugar content. Another clinically 
significant finding was the perception of certain diarrhoeas as 
folk illnesses requiring traditional remedies rather than 
biomedical therapy. Two such illnesses were 'sutt' (fallen 
fontanel) and 'nazar' (evil eye). The import of this finding was 
that, if serious symptoms are attributed to a folk disease, health 
could be compromised while traditional treatments are sought. Mull 
and Mull also discovered as an aspect of folk etiology and therapy 
that certain diarrhoeas were considered 'normal' - a transition in 
the process of growing up and hence was not considered an illness. 
The belief was that such diarrhoea stops by itself by avoidance of 
the root cause. Such diarrhoea include those associated with 
'teething' ; those viewed as caused by a breastfeeding mother having 
eaten foods that were too hot or having become overheated in the 
sun; or having become pregnant. Besides, the study revealed that, 
even among mothers who had ever used ORS, a vast majority did not

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



know how the solution should be prepared and administered. Whilst 
several women described a preparation method that would have 
yielded a solution much too dilute or much too concentrated to be 
effective, others thought that it should be given in doses of 1 or 
2 teaspoonfuls, 2 or 3 times daily - confusing its administration 
with the manner in which over-the-counter syrups were administered. 
In a concluding note, Mull and Mull attributed the lack of 
enthusiasm for ORT and the subsequent usage pattern observed to 
inadequate understanding of the concept of ORT.

Malik et al. (1992) used survey methods and in-depth interviews of 
595 households in rural and urban Punjab-Pakistan, to study 
mothers' emotional response to children's illness with particular 
emphasis on diarrhoea. The main thrust of the study was to assess 
the extent to which diarrhoea was particularly distressing to 
mothers; the relationship between seriousness and children's 
symptoms; and mothers' thoughts about causes, consequences and 
treatment of various forms of diarrhoea in children.

Results of the household survey showed that mothers used various 
local names for diarrhoea; 'dast', 'julab' and 'paichish'. 
Cultural categories of diarrhoea were established by colour and 
consistency; whether blood was present in the stool; and whether 
defaecation was explosive. Mothers also associated certain causes 
and varying degrees of severity with different types of diarrhoea. 
It was also found that diarrhoea must be regarded not only as a 
discrete disease entity but as a symptom belonging to several 
popular illness categories. An analysis of the relationship

40
University of Ghana     http://ugspace.ug.edu.gh



between local cultural models of diarrhoeal illnesses to maternal 
emotional responses revealed that the latter was in part shaped by 
the illness categories to which mothers assigned a child's episode. 
In addition, maternal fears that symptoms of diarrhoea may be life- 
threatening were found to be associated with death of children, 
with treatment choices and help-seeking as a whole. Besides, a 
significantly high proportion of mothers who fear diarrhoea to be 
life-threatening to their children rather than mothers with other 
concerns choose to use Nimkol - the Pakistani ORS. The importance 
of recognising the complexity of interpretative and emotional 
processes which shape the care of children and the home treatment 
of childhood diseases is also emphasised.

Adetunji (1989) highlighted the behavioural response of parents to 
five major killer diseases in his study of a Yoruba community in 
Nigeria. Data were collected on parent's perceptions of cause, 
best treatment and the decision-making process about the killer 
diseases which were diarrhoea, measles, tetanus, pertussis and 
fever.

With regard to diarrhoea, teething was found to be the major cause. 
This was reported by about 44 percent of the total sample size of 
27 respondents. Diarrhoea occurring prior to teething was 
considered necessary because it did away with all inhibitions in 
the way of emerging teeth. The type of impediments that needed to 
be flushed out included diseases 'hiding' inside the child's 
abdomen - measles being one of these. Food-related causes were the 
second major group reported (about 30 percent). They include

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



inappropriate food for the child, overfeeding, eating too hot food, 
using dirty feeding bottles or utensils for children and a sudden 
change from one infant formula to another. The major treatments 
recommended were medicinal herbal infusions (29 percent), ORT (24 
percent) and' locally purchased antibiotics (12 percent). 
Generally, a prevalent belief was that loose motions must be 
allowed to continue for sometime before any treatment was sought. 
Foods that were cold in temperature wete to be given during 
diarrhoea whilst beans was to be withdrawn from the child's diet. 
One important finding that the study thought could have both 
positive and negative health implications was the use of medicinal 
teas. The view was that, although, these herbal remedies could be 
a useful rehydrating fluid, they could become a threat to health in 
the hands of caretakers of the child with little or no schooling 
when herbal infusions are prepared and administered under 
unhygienic conditions. The paper concludes that although some of 
the practices might have negative health implications they could 
usefully be adopted to the goal of self-reliance in medical care 
within the PHC context.

Coreil and Genece (1988) used ethnographic and survey methods in 
Haiti to study the extent to which various variables influenced the 
use of ORT, including beliefs and practices concerning diarrhoeal 
illness, socioeconomic status and exposure to health training 
programmes among rural and urban mothers. They discovered that 
mothers' beliefs about how ORT works (about whether it stops or 
replaces fluids) determined their treatment choices (home-made 
solutions versus packaged ORS), as well as the time-lag before

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



treatment was started. Prior help seeking at medical clinics was 
associated with use of ORT, but family use of traditional remedies 
(prior to treatment) was not. Their findings indicate that 
literacy and beliefs about ORT are important determinants of use in 
Haiti and should be addressed in health education programmes.

Chowdhury et al. (1988) in an epidemiological evaluation of the
usage and safety of home-made oral rehydration salts also 
investigated peoples' perceptions, beliefs and customs about 
diarrhoea, its treatment and the extent to which these affected the 
use of Lobon-gur solution (LGS) - the Bangladeshi ORS. The study 
which was conducted in 3 phases identified 4 main types of 
diarrhoea with specific causes and other peculiar characteristics. 
'Dug haga1 was believed to occur when a child suck 'polluted' 
breastmilk. 'Ajirno' diarrhoea was perceived to be caused by 
indigestion and food poisoning. Whilst the cause of 'Amasha1 
diarrhoea was unknown, it was believed to have a mucousy stool 
characteristic. The fourth type which was described as 1 severe 
diarrhoea' or 'cholera' was felt to be very dangerous and that it 
could result in dehydration and subsequent death. Among others, 
the evaluation depicted that over a half of all the diarrhoeal 
episodes recorded in the study had received some form of treatment. 
In ranking modes of treatment used, LGS was third in importance 
after other allopathic medicines, and 'Boneji' - a household herbal 
remedy. Moreover, a high usage rate of LGS was observed among 
mothers for 'severe diarrhoea' which was thought to be the most 
life-threatening and thereby demonstrating the usefulness of using 
a typology of diarrhoea based on peoples' perception.

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



In Ghana some of the studies related to the management of childhood 
diarrhoea have been conducted by Afari et al. (1988), Ghana
Statistical Service (Ghana) - GDHS (1988), Adjei (1988) and Abu 
(1988) .

Afari et al. (1988) in their survey of childhood morbidity and
treatment in two rural communities (Gomoa Fetteh and Gomoa 
Onyadze/Otsew Jukwa) paid attention to determining the under five 
annual incidence rate, mother's definition of diarrhoea, and 
treatment choices. In all, a total of 235 and 132 mothers were 
interviewed at Gomoa Fetteh and Gomoa Onyadze/Otsew Jukwa 
respectively. Among others, the study revealed that majority of 
mothers (96.6 percent) at Gomoa Fetteh and 95.5 percent at Gomoa 
Onyadze/Otsew Jukwa defined diarrhoea in a child as the passing of 
5-8 loose stools in a 24-hour period. By this definition, it was 
the view of the study that, mothers did not recognise diarrhoea 
early enough to start home treatment. About 40 percent of mothers 
at Gomoa Fetteh will treat diarrhoea with drugs like paracetamol 
and antibiotics (tetracyline and ampicillin), 28.9 percent will 
give sugar salt solution (SSS), 10.2 percent will administer enema 
and 80.1 percent will take their children to.the clinic. In Gomoa 
Onyadze/Otsew Jukwa, 29.5 percent of mothers will choose the clinic 
as a first treatment choice, 28 percent will buy drugs, whilst 21.2 
percent and 7.6 percent will give enema and SSS respectively.

Although this study mentions that a mother's concept of diarrhoeal 
disease and its treatment are shaped by cultural factors, these are 
not considered important in any effort aimed at educating mothers

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



on the appropriate management of childhood diarrhoea.
45

In the GDHS (1988) , carried out by the Statistical Service of 
Ghana, an attempt was made to,assess the prevalence of diarrhoea in 
the country as a whole and the various treatments that were used in 
these episodes. The study made use of a point and a period 
prevalent measure of diarrhoea in the last 24 hours and the past 
two weeks respectively, as well as various sociodemographic 
indicators. It was discovered among others that there was a 
curvilinear relationship of the prevalence of diarrhoea to age of 
the child. Children under six months of age were least likely to 
have had diarrhoea. The rate was found to increase up to age 12-17 
months and then falls. It was also found that the prevalence of 
diarrhoea varied little according to the sex of the child or region 
except in the Western Region of Ghana where diarrhoeal prevalence 
was lower than elsewhere. There was also no substantial 
differences between children living in urban areas and those living 
in rural areas. Diarrhoea prevalence was also found to be 
approximately the same for children of mothers with no education, 
primary education and middle school education but was lower with 
children whose mothers had higher education. Concerning treatment 
it was reported that 40 percent of children with diarrhoea in the 
last two weeks were taken to a medical facility. Moreover, 
children living in urban areas were more likely to consult a 
medical facility than rural children. In addition approximately 
one-third of children with diarrhoea in the last two weeks were 
treated with ORT - 34 percent were given ORS while 6 percent were 
given SSS. Apart from reporting a widespread use of traditional

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



medicine in the treatment of diarrhoea its' use was found to be 
more prevalent in rural than urban areas. Also 12 percent of 
children in the last two weeks received no treatment. With respect 
to knowledge about ORT while only two-fifths of women with no 
education said they had heard of the ORS packets, more than three- 
fifths of those with primary education and four-fifths of women 
with higher education knew about packets.

Another important contribution to the understanding of the 
management of childhood diarrhoea has been made by Adjei (1988). 
He investigated factors affecting the transmission of diarrhoea in 
a cross-sectional survey of Accra and Ada districts. The study 
sought to understand the extent to which household demographic, 
social and economic characteristics, health services utilisation by 
households, maternal knowledge, attitudes and practices (KAP) 
related to diarrhoeal episodes contributed to its prevalence. Five 
social factors that were considered important for the transmission 
of diarrhoea were; age of the mother, parity, ethnicity, religion 
and educational level of husband. With regard to maternal age, a 
high prevalence rate was observed in children of mothers aged 15-29 
years compared with mothers above 50 years.. Mothers with higher 
parity of 3 or 4 children had less diarrhoea in their children 
compared to those with 1 or 2 children. A higher prevalence rate 
was obtained among children of parents of the Northern and Upper 
regions living in the South. Also children of Moslems and 
Catholics tended to have higher rates compared with those Of other 
religions. Higher paternal education was found to be negatively 
associated with the transmission of diarrhoea.The KAP aspects of

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



the study highlight some maternal perceptions of childhood 
diarrhoea that are of relevance to the present study. It was shown 
among others that diarrhoea was conceptualised in terms of 
frequency of passage of stools and also in variations in its 
frequency. Lack of knowledge on the causes of diarrhoea was also 
observed to be significantly associated with diarrhoea transmission 
in children. Those who do not know what causes diarrhoea or said 
it was due to teething had a high prevalence than those who said it 
was due to germs/flies or some kind of fever. Knowledge on how to 
prevent or cure as well as what to do in case of diarrhoea was also 
seen not to be associated with diarrhoea transmission. Following 
from these observations, Adjei identified the following groups of 
people as being at the greatest risk; Children of young mothers 
(15-29 years); children of mothers whose husbands have had little 
or no education and multiparous mothers, and that these should be 
made the targets of health educational programmes.

From the review of these three studies in Ghana (ie Afari et al. 
1988, GDHS 1988 Adjei 1988), it is perceptible that their studies 
have been mainly epidemiologic in character - attempting to assess 
the incidence and distribution of disease and/or the various 
factors controlling its transmission. In such studies, the 
tendency is to rely heavily on the techniques of analytical 
epidemiology using statistical methods to demonstrate associations 
between variables and ensure the reliability of their findings 
whilst neglecting ethnomedical conceptions about diarrhoeal 
illness. Anthropologists however argue that, for their findings to 
be valid, an ethnomedical perspective is essential (Kleinman 1978) .

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



Nations (1986) cautions against reliance on decontextualised 
statistical tools: failure to consider the ethnomedical perspective 
she argues, transforms this "epidemiological" rigor into 
'quantitative rigormortis'.

One of the few studies which deal exclusively with indigenous 
concepts of diarrhoeal diseases and its treatment in Ghana has been 
Abu's (1988). This was an intervention study preceding the 
introduction of CDD programme in Ghana. In a cross-sectional 
survey of various ethno-geographic regions, Abu highlighted popular 
perceptions of diarrhoea, its causes, its effects on the body, 
associated symptoms and diseases, care of the diarrhoea patient and 
treatment, as well as the extent to which signs of dehydration were 
recognised by caretakers. The study which featured five major 
language groups in Ghana identified different names for diarrhoea 
but with the same meaning as the term diarrhoea is used in a 
layman's English. Some of these different names included 'ayamtuo' 
(Akan), 'musutsomo' (Ga). Varying types of diarrhoea like 'sunken 
eyes diarrhoea', 'dysentry' 'diarrhoea with measles ' and 'diarrhoea 
with malaria' were also identified. The dominant causes of 
diarrhoea were perceived to be first, contaminated food or water, 
second, food unsuited for a person's age or constitution, teething, 
fever, measles and the habit of not washing hands before meals. 
The literates in urban settings placed more emphasis on 
contaminated food, poor hygiene and teething than did the non­
literates. It was also discovered that four main approaches were 
used in the treatment of diarrhoea. These are local medicines, 
self-medication with pharmaceutical products, hospital treatment

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



49
and ORS.

Much as Abu's study contributed to the clarification of local 
concepts surrounding childhood diarrhoea, some degree of caution 
needs to be exercised in any attempt to apply her findings at the 
macro level. This is mainly due to the fact that her cross- 
sectional survey data could have been confounded by several 
important demographic socio-economic and social-structural/ethno- 
geographic differences which may have affected the behavioural 
outcomes reported.

In addition, most of the studies in Ghana have tended to follow the 
most common mode of rapid social data collection in applied health 
research - the survey method. This method utilises extensive 
samples with rather thin information from individuals. In 
contrast, the approach that has been used in this study is a micro 
one, emphasising local knowledge. Hence the decision to confine 
the study to only one community. Granted that there are different 
ethnic groups in Ghana, the need to investigate thoroughly 
ethnomedical models of diarrhoeal illness with particular reference 
to specific ethno-geographical areas cannot- be overemphasised. To 
this end a combination of methods have been used to accumulate 
evidence on the management of diarrhoea from various research 
approaches.

2.1 Summary
The main concern in this chapter has been to review literature 
investigating diarrhoeal illness and related beliefs and practices

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



in different socio-cultural contexts of the world.
50

It has been shown that most of the studies - especially the 
anthropological ones have generated culturally sensitive findings 
of practical significance for health professionals and programme 
planners in the promotion of ORT as the most effective way to 
prevent dehydration due to diarrhoea.

Furthermore, these studies have demonstrated the usefulness of 
relating health, illness, healing and the cultural context that 
defines them as important determinants of health-seeking 
preferences in diarrhoeal episodes.

In the next chapter, the social organisation of the people of Pute 
will be outlined to provide a basis for understanding the context 
in which diarrhoeal illness is managed.

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



CHAPTER THREE 
THE SOCIAL STRUCTURE OF PUTE

3.0 Introduction
It is important to give a descriptive profile of Pute society in 
general in order to provide a context from which to view some of 
the social characteristics of the respondents and a cultural milieu 
within which the management of childhood diarrhoea as a conceptual 
and a cognitive phenomenon would be appreciated in this study.

3.1 Geographical Location
Pute lies on longitude 0°32'E and latitude 5°47'N. It is among a 
number of fishing villages situated on the sand spit between the 
Songhor Lagoon and the Gulf of Guinea along the south-east coast of 
Ghana. (see map overleaf). It is linked by a 10 kilometer third 
class-road from Ada-Foah,the capital of the Dangme-East District.lt 
is about 118 kilometers from the centre of Accra.

Pute experiences mean annual temperatures of 27.2°C and humidity 
of 82.2 percent. Rainfall figures show that this part receives 
much lower rainfall than most parts of Ghana, the mean annual 
rainfall being between 74 and 89 centimetres.. There are two rainy 
seasons: April to June is the major season in which half the rain 
of the year is received; the minor season - a period of less rain 
covers the months of September and November. In between the rainy 
seasons, conditions of extreme drought prevail. The scantiness of 
rainfall in this region gives special importance to the existing 
surface and underground water supplies. The type of vegetation 
which occurs at Pute is short grass with small clumps of bush and

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



52

Fig.2  MAP OF THE DANGME E A S T  DISTRICT SHOWING T H B  

LOCATION OF THE STUDY AREA — P U T E

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



a few trees. The soil is poor and is of the coastal sandy type.

3 .2 Historical Links
The origins of Pute are shrouded in myth as it has no recorded 
history. Oral tradition however has it that the village was founded 
by a member of the Tekperbiawe clan4 - a clan belonging to the 
stock of Dangmes who settled in the Ada area after having migrated 
together with other groups from elsewhere3. It is held that not 
long after settling in the Ada area, a group of fishermen led by a 
member of the Tekperbiawe clan (whose name is not known) embarked 
on a fishing expedition to Pute during a major fishing season. The 
usual practice was for such migrant fishermen to return home after 
the peak fishing season. The crew of fishermen who visited Pute 
however realised that they continued to experience bumper catches 
even when the main fishing season was supposed to have ended so 
they decided to create a permanent settlement here to enable them

‘For a detailed discussion- oxu clans- at- Pute see. section on 
"Kinship and Family Life"

According to Quaye (1972), the origins of the 'Ga' and their 
eastern neighbours 'Adangme1, (comprising Ada, Kpone, Prampram, 
Ningo, Osudoku, Shai, Yilo and Manya Krobo) are shrouded in myth 
and mystery and all sorts of hypothesis have been advanced. The 
most widespread of this hypothesis first popularised by Reindorf 
(1895) and accepted by Ward and Field (1940) maintains that the Ga 
originated from Benin. In addition to this hypothesis, Reindorf 
recorded two traditional accounts of origin. One claimed that the 
Ga originated from the sea, and the other that they emigrated 
together with the Adangme, from Tetetutu or Sameh in the East 
between two large rivers, and that after crossing the River Volta, 
they dispersed over the country. Azu (1926) accepted the Sameh 
origin of the Ga and the Adangme. He described Sameh as an island 
situated on the South-West of the River Ogun adjoining 'Ladah' in 
'Dahome". He claimed that the Ga and the Adangme left Sameh 
because they were greatly oppressed by the mighty King "Akpo" of 
'Dahome.

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



continue with their work. The village was therefore named 'Pute' 
meaning 'fishing ground'. Thus Pute is of Dangme origin with the 
basic socio-cultural characteristics of the people of the state of 
Ada, now referred to as the Ada Trational Area with its seat of 
paramountcy at Big-Ada in the Dangme East District of Ghana. The 
capital of the District is however Ada-Foah. The local language 
spoken here is Dangme.

3.3 Population Structure
According to the 1984 national census report, the total population 
of Pute is 1779. (Refer to Table 3.1). About 16.6 percent of the 
population are below five years of age. The 1984 population census 
indicates that females outnumber males. The difference is clearer 
among those above 15 years. In this group. 54.4 percent are 
females whilst 45.6 percent are males. The population tends to be 
dominated by females due to high migration among young male adults. 
Table 3.1 gives a clear picture of Pute's population as.at 1984.

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



TabLe_3_,.l
Population Distribution By Ages and Sexes.

Age Groups Male Female Total
All Ages 871 908 1779
Below 1 year 25 12 37
1 - 4 143 116 259
5 - 9 154 150 S04-
10 - 14 89 81 170
15 - 24 172 181 353
25 - 44 126 175 301
45 - 64 92 119 211- -
65 and over 70 74 114

Source: 1984 Population Census of Ghana, Vol.II Census
Office, Accra (unpublished Report)

As Table 3.1 shows, Pute has a youthful population. Persons from 
the ages below 1 to 24 years form a total of 1123 or 63.1 percent 
of the total population. The preponderance of young persons in 
Pute is perhaps due to the value placed on high fertility rates as 
a yardstick for measuring social status as well as the rise in 
teenage pregnancies and subsequent births among the youth within 
the last decade.

3.4 Economic Activities
The traditional economic system of Pute has been fishing. The 1984 
population census shows this clearly. Table 3.2 shows the economic 
structure of Pute.

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



Distribution of Economically Active Population
Table 3.2

Type of Employment Male Female Total
Fishing 336 9 345
Homemaker 6 13 19
Unemployed 5 6 11
Others 74 70 144

Source: 1984 Population Census of Ghana Vol.II, Census
Office, Accra. (Unpublished)

As Table 3.2 shows, the most important economic activity in Pute is 
fishing - it employs 66.5 percent of the economically active 
population. Some of the people are described as homemakers and 
others as unemployed, yet virtually every adult person in Pute 
engages in some economic activity. There are two main sources of 
fish at Pute: the sea and the Songhor Lagoon. Sea fishing is a 
year round activity with the period from July to November being the 
peak season. In the Songhor Lagoon, fishing continues throughout 
the year and becomes very intensive for a period after the rainy 
season when the sand bars separating the lagoon from the sea is 
breached either artificially or naturally. Fishing is usually done 
with the aid of large hand-nets used from canoes. Curing and 
marketing of fish are done by women.

The sea is greatly revered by the people of Pute who depend 
directly on it for their livelihood and for wealth in particular. 
As a result it is a taboo to go fishing on Tuesday which day is set 
aside for offering sacrifices to the sea gods, mending fishing gear 
and undertaking communal self-help projects.

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



The sandy nature of the soils at Pute does not allow for any 
meaningful agricultural activity to be engaged in. The thick 
layers of beach sand along the coast of Pute are only useful for 
the cultivation of coconuts. Around a considerable part of the 
Songhor Lagoon are thick layers of clay which are for the most part 
infertile. A few vegetables however do well at Pute.

Another important source of income for the people of Pute is salt- 
mining though this is not reported specifically by the 1984 
population census. It is usually carried out during the dry season 
when portions of the lagoon dry out leaving salt crystals on the 
beds. A common sight at Pute is therefore heaps of salt stored 
under woven coconut branches to be sold during the rainy season 
when their prices begin to soar up.

Although the village does not have a market place, petty trading is 
engaged in mostly by women in front of their houses,, in their 
houses or along the major road passing through to a nearby village 
(Totopey) from Ada-Foah. The wares on sale are mostly food items 
which are normally purchased from markets at Ada-Foah and Kasseh 
(about 30 kilometres away). They include cassava/cassava dough, 
bread, groundnuts, beans, rice, garden eggs and gari.

3.5 Kinship and Family Life
The key to the understanding of the structural principles forming 
the basis of social life at Pute is through the patrilineal descent 
system, which is based on clan and lineage organisation. The 
people of Pute are basically Dangme people who trace descent

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



lineally through the male line. With the exception of a few cases 
each person becomes absorbed into his patriliny through specific 
infancy and naming rituals. Thereafter, he enjoys full living and 
home rights and more importantly, the right to be buried according 
to custom. The fact that one is a member of his paternal kin group 
defines one's rights and obligations within it. The duties and 
responsibilities within one's paternal kin group entails co­
operating with the head of the group in the performance of 
customary rituals.

In Pute, while membership ties with one's paternal kin group are 
defined in juridical customary terms, the relationship to one's 
maternal kin is more of a free though conventional nature. 
Although a person recognises the relatives with whom he is 
connected through his mother, he does not belong to them like a 
member of a corporate group. In other words, groupings based on 
matrilineal descent does not exist among the Ada.

In Pute, actual patterns of social groupings are built around the 
concept of 'We' which has three meanings in Dangme viz 'home' 
'house' (habitation) or "people of the house", conceived as a 
social unit (ie clan or lineage).

Members of a typical clan at Pute comprise all people who are 
united by a belief in a common descent through the male line back 
from three to six or even more generations. This group excludes 
children of unmarried daughters and descendants of slaves.

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



The whole of Ada-land is divided into seven main clans which are 
the Dangmebiawe, Tekperbiawe, Lomobiawe, Adibiawe, Kujragmebiawe, 
Kabiawe and Ohoewe. The clans have each been named after a 
putative apical ancestor believed to have founded it, 
distinguishing it from other groups. The clans are dispersed 
throughout the Ada area in such a way that every clan is 
represented in each community. This arises due to inter-marriages 
and migration. Consequently, as in other Ada communities, every 
Pute person belongs through the male line to one of these clans.

Every village in Ada is under the governance of a founding clan 
among the seven already mentioned. In Pute society therefore it is 
the Tekperbiawe clan which rules the village. The chief and all 
his elders are members of this clan.

In Ada-land as a whole, each of the seven clans has a male head who 
is usually resident at Big-Ada. Such a person is supposed to be 
the most senior kinsman of the oldest living generation within the 
whole clan, regardless whether his father was a first-born or not. 
H