University of Ghana http://ugspace.ug.edu.gh
UNIVERSITY OF GHANA 
REGIONAL INSTITUTE FOR POPULATION STUDIES 
 
  
WOMEN’S EDUCATIONAL ATTAINMENT AND HEALTH-SEEKING 
BEHAVIOUR FOR CHILDHOOD ILLNESSES IN GHANA 
  
BY 
 
AWURA ADWOA ANTWIWAA AMOFA 
(10460190) 
  
THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN 
PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF MA IN 
POPULATION STUDIES DEGREE 
  
OCTOBER, 2020 
 
 
 
 
University of Ghana http://ugspace.ug.edu.gh
ACCEPTANCE 
Accepted by the College of Social Science, University of Ghana, Legon in partial fulfillment 
of the requirement of the Master of Arts Degree (Population Studies). 
                  
  
 
 
       28TH OCTOBER 2020 
 
PROF. STEPHEN OWUSU KWANKYE     DATE 
       (THESIS SUPERVISOR) 
 
 
                                                                                                                                         
 
 
 
 
                                                                        
 
  
                                                    
i 
 
University of Ghana http://ugspace.ug.edu.gh
DECLARATION 
I AWURA ADWOA ANTWIWAA AMOFA, hereby declare that this thesis submission is 
my own research work and apart from references of other people’s work that has been duly 
acknowledge, it includes no materials of previously published work by another person 
elsewhere. 
  
  
                                                               
                                                                                                    
 
    28TH OCTOBER 2020 
 
 AWURA ADWOA ANTWIWAA AMOFA                         DATE 
                (STUDENT)                                                              
               
 
  
ii 
 
University of Ghana http://ugspace.ug.edu.gh
DEDICATION 
This work is dedicated to God Almighty; You are my glory, the one who holds my head 
high. I dedicate this work also to Uncle Geyhart Tekpor and Mr. Kwesi Gyimah-Asante. 
Until we meet again!  
 
  
  
  
  
  
  
  
  
  
  
 
 
 
 
 
 
iii 
 
University of Ghana http://ugspace.ug.edu.gh
ACKNOWLEDGEMENT 
I thank God Almighty for His grace and provision made available for me and making this 
journey possible. To my mum Barbara Adisenu-Doe, the strongest woman I know. Thank 
you for your motherly support.   
I am truly grateful to my supervisor, Prof. S.O. Kwankye for his suggestions, patience and 
guidance in ensuring the completion of this dissertation. God richly bless you! To all the 
teaching and non-teaching staff of RIPS, am thankful for the knowledge and encouragement 
offered me especially Dr. Pearl Kyei.   
To all RIPS PhD students especially Charles Agyei-Asabere, Martin Wiredu Agyekum, Isaac 
Yeboah and David Atombire Adumbire for their support and assistance during the time of 
my studies at RIPS. And to all my colleagues of the 2019/2020 MA cohort, I am delighted to 
have you as friends. 
I will always remember the contributions of my brothers especially Andrew Nii Aryee 
Aryeetey and my loved ones from Kaneshie Presbyterian Church and Kaneshie Polyclinic for 
their prayers and support. I am eternally grateful.  
 
  
  
  
  
 
 
 
 
 
iv 
 
University of Ghana http://ugspace.ug.edu.gh
ABSTRACT 
Childhood illnesses are among the main health problems globally and predominantly in Asian and 
African subcontinents. About 6.0 million under-five children died in 2018 worldwide, which could 
be prevented with simple and inexpensive interventions. Educating women empowers them by 
increasing their autonomy. Women with greater power are able to influence health care decisions 
for their children compared to women with less education. Mothers’ ability to act appropriately 
and promptly to illness symptoms will reduce childhood morbidity and mortality. Assessing the 
health-seeking behaviour of mothers and identifying specific factors influencing health-
seeking are essential to advance strategies and interventions for child survival strategies in the 
country.  
The study examines the relationship between mothers’ education and health-seeking behaviour 
during childhood morbidities in Ghana, using the 2014 Ghana Demographic and Health Survey 
(GDHS) data. A sample of 638 and 1,189 children with diarrhoea and fever/cough symptoms were 
used in the study. Univariate, bivariate and multivariate statistical tools were used to analyse the 
data. 
Findings reveal that 30.1% and 25.8% of mothers sought no treatment, while 47.1% and 52.8% 
sought medical treatment and 22.8% and 21.4% sought non-medical treatment for children 
exhibiting diarrhoea and fever/cough symptoms respectively. In the multivariate analysis, mothers 
with no education were significantly more likely to seek non-medical treatment compared to 
mothers with secondary/higher education. Results also show that mother’s age, ethnicity, religion, 
region, place of residence, wealth index, working status and age of child were significant predictors 
for health-seeking behaviour for childhood illnesses in Ghana. Less educated mothers are more 
likely to use other forms of treatments such as pharmacy, drug peddlers and traditional 
practitioners instead of visits to medical facilities when children show fever/cough symptoms. The 
study suggests that improving the educational status of women will improve better health-seeking 
practices as mothers will seek appropriate treatment to increase the chances of child survival during 
ill-health. 
. 
v 
 
University of Ghana http://ugspace.ug.edu.gh
TABLE OF CONTENTS 
 
ACCEPTANCE ............................................................................................................................................. i 
DECLARATION .......................................................................................................................................... ii 
DEDICATION ............................................................................................................................................. iii 
ACKNOWLEDGEMENT ........................................................................................................................... iv 
ABSTRACT .................................................................................................................................................. v 
TABLE OF CONTENTS ............................................................................................................................. vi 
LIST OF TABLES ........................................................................................................................................ x 
LIST OF FIGURES .................................................................................................................................... xii 
CHAPTER ONE INTRODUCTION ............................................................................................................ 1 
1.1 Background of the Study ............................................................................................................... 1 
1.2 Statement of the Problem ............................................................................................................ 3 
1.3 Research Questions ...................................................................................................................... 4 
1.4 Rationale ....................................................................................................................................... 5 
1.5 Objectives...................................................................................................................................... 6 
1.6 Organization of the study ................................................................................................................... 6 
CHAPTER TWO .......................................................................................................................................... 8 
LITERATURE REVIEW ............................................................................................................................. 8 
2.1 Introduction ........................................................................................................................................ 8 
2.2 The Burden of Childhood Illnesses and Mortality ............................................................................... 8 
2.3 Strategies for Reducing Childhood Morbidity and Mortality............................................................ 11 
2.3.1 Integrated Management of Childhood Illness (IMCI) ................................................................ 11 
2.3.2 Expanded Programme on Immunization (EPI) ........................................................................... 12 
2.4. Women’s Educational Attainment and Child Health and Mortality ................................................ 13 
2.5 Pathways Women’s Education Influences Health-seeking Behaviour .............................................. 15 
vi 
 
University of Ghana http://ugspace.ug.edu.gh
2.5.1 Mother’s Education and Health-seeking Behaviour .................................................................. 17 
2.5.2 Health-seeking Behaviour for mothers and their children ........................................................ 19 
2.5.3 Fathers influence on Health-seeking Behaviour ........................................................................ 20 
2.6 Different Treatment Sources for Childhood Illness .......................................................................... 21 
2.7 Factors Associated with Health-seeking Behaviour for the Child ..................................................... 23 
2.7.1 Mother’s Age .............................................................................................................................. 23 
2.7.2 Religion....................................................................................................................................... 23 
2.7.3 Household Wealth ..................................................................................................................... 24 
2.7.4 Type of Place of Residence ........................................................................................................ 25 
2.7.5 Child Characteristics ................................................................................................................... 25 
2.7.6 Illness Symptoms and/or Perceived Severity ............................................................................. 26 
2.8 Gaps in Literature .............................................................................................................................. 27 
2.9 Theoretical Framework ..................................................................................................................... 28 
2.10 Conceptual Framework ................................................................................................................... 28 
Fig 2.2: Conceptual framework showing the relationship between women’s educational attainment 
and health-seeking behaviour for childhood morbidity. ........................................................................ 30 
2.11 Hypotheses ..................................................................................................................................... 31 
CHAPTER THREE .................................................................................................................................... 32 
METHODOLOGY ..................................................................................................................................... 32 
3.1 Introduction ...................................................................................................................................... 32 
3.2 Study Area ......................................................................................................................................... 32 
3.3 Source of Data ................................................................................................................................... 33 
3.4 Sample size and Design ..................................................................................................................... 34 
3.5 Variables and their Measurements ................................................................................................... 34 
3.5.1 Dependent Variable ................................................................................................................... 34 
3.5.2 Independent Variable ................................................................................................................ 35 
3.5.3 Control Variables ........................................................................................................................ 35 
3.6 Method of Analysis ........................................................................................................................... 38 
3.7 Limitations ......................................................................................................................................... 38 
vii 
 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER FOUR ....................................................................................................................................... 40 
BACKGROUND CHARACTERISTICS OF RESPONDENTS ................................................................ 40 
4.1 Introduction ...................................................................................................................................... 40 
4.2 Incidence of illness symptoms .......................................................................................................... 40 
4.3 Health-seeking Behaviour of Mothers .............................................................................................. 41 
4.4 Mothers’ Educational Attainment .................................................................................................... 42 
4.5 Mother’s Age ..................................................................................................................................... 43 
4.6 Marital Status .................................................................................................................................... 44 
4.7 Ethnicity ............................................................................................................................................ 45 
4.8 Religion.............................................................................................................................................. 46 
4.9 Region ............................................................................................................................................... 47 
4.10 Type of Place of Residence ............................................................................................................. 48 
4.11 Working Status ................................................................................................................................ 49 
4.12 Wealth Index ................................................................................................................................... 50 
4.13 Sex of child ...................................................................................................................................... 51 
4.14 Age of the Child ............................................................................................................................... 52 
4.15 Conclusions ..................................................................................................................................... 53 
CHAPTER FIVE ........................................................................................................................................ 54 
MOTHER’S EDUCATIONAL ATTAINMENT AND HEALTH-SEEKING BEHAVIOUR .................. 54 
5.1 Introduction ...................................................................................................................................... 54 
5.2: Mother’s Educational and Health-seeking Behaviour ..................................................................... 54 
5.3: Mother’s Age and Health-seeking Behaviour .................................................................................. 55 
5.4: Marital Status and Health-seeking Behaviour ................................................................................. 57 
5.5: Ethnicity and Health-seeking behaviour .......................................................................................... 58 
5.6: Religion and Health-seeking behaviour ........................................................................................... 59 
5.7 Region and Health-seeking Behaviour .............................................................................................. 60 
viii 
 
University of Ghana http://ugspace.ug.edu.gh
5.8: Type of Place of Residence and Health-seeking Behaviour ............................................................. 62 
5.9: Working Status and Health-seeking Behaviour ............................................................................... 63 
5.10: Wealth Index and Health-seeking Behaviour ................................................................................ 64 
5.11: Sex of Child and Health-seeking Behaviour ................................................................................... 65 
5.12: Age of Child and Health-seeking Behaviour ................................................................................... 66 
5.13 Conclusions ..................................................................................................................................... 68 
CHAPTER SIX ........................................................................................................................................... 69 
RELATIONSHIP BETWEEN MOTHERS’ BACKGROUND CHARACTERISTICS AND HEALTH-
SEEKING BEHAVIOUR FOR UNDER-FIVE CHILDREN .................................................................... 69 
6.1 Introduction ...................................................................................................................................... 69 
6.2 Influence of Background Characteristics on Health-seeking Behaviour for Diarrhoea .................... 70 
6.3 Influence of Background Characteristics on Health-seeking Behaviour for Fever/cough ................ 74 
6.4 Discussion of Results ......................................................................................................................... 78 
6.4.1 Determinants of Health-seeking Behaviour for Diarrhoea ........................................................ 78 
6.4.2 Determinants of Health-seeking Behaviour for Fever/cough .................................................... 80 
6.4.3 Differences between Health-seeking Behaviour for diarrhoea and fever/cough symptoms .... 82 
CHAPTER SEVEN .................................................................................................................................... 83 
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS ............................................................... 83 
7.1 Introduction ...................................................................................................................................... 83 
7.2 Summary of main findings ................................................................................................................ 83 
7.3 Conclusions ....................................................................................................................................... 85 
7.4 Recommendations ............................................................................................................................ 86 
REFERENCES ........................................................................................................................................... 88 
 
  
ix 
 
University of Ghana http://ugspace.ug.edu.gh
 LIST OF TABLES 
Table 3.1: Measurement of control variables ……………...........................................33  
Table 4.1: Percentage distribution of mothers by age and symptom exhibited among children                                                                                                                 
……………………………………………………………………………………………………40  
Table 4.2: Percentage distribution of mothers by marital status and symptom exhibited among 
children………………………………………………………………………………………......41    
Table 4.3: Percent of mothers by ethnicity and symptom exhibited among children……………42  
Table 4.4: Percent of mothers of under-five children by region of abode and health condition of 
children…………………………………………………………………………………………..44 
Table 4.5: Percentage distribution of mother’s place of residence and symptom exhibited among 
children…………………………………………………………………………………………..45  
Table 4.6: Percentage distribution of mothers by wealth index and symptom exhibited among 
children…………………………………………………………………………………………..47 
Table 4.7: Percentage distribution of children under five years by single age and symptom 
exhibited among children………………………………………………………………………...49                                                                                                          
Table 5.1: Percent of mothers by educational level, health-seeking behaviour and symptom of 
health condition of child…………………………………………………………………………51 
Table 5.2: Percent of mothers by age, health-seeking behaviour and symptom of health condition 
of child…………………………………………………………………………………………...52 
Table 5.3: Percent of mothers by marital status, health-seeking behaviour and symptom of health 
condition of child…………………………………………………………………………………53 
Table 5.4: Percent of mothers by ethnicity, health-seeking behaviour and symptom of health 
condition of child…………………………………………………………………………………54 
Table 5.5: Percent of mothers by religion, health-seeking behaviour and symptom of health 
condition of child…………………………………………………………………………………55 
x 
 
University of Ghana http://ugspace.ug.edu.gh
Table 5.6: Percent of mothers by region, health-seeking behaviour and symptom of health 
condition of child…………………………………………………………………………………56 
Table 5.7: Percent of mothers by place of residence, health-seeking behaviour and symptom of 
health condition of child………………………………………………………………………….57 
Table 5.8: Percent of mothers by working status, health-seeking behaviour and symptom of health 
condition of child…………………………………………………………………………………58 
Table 5.9: Percent of mothers by wealth index, health-seeking behaviour and symptom of health 
condition of child…………………………………………………………………………………59 
Table 5.10: Percentage distribution of sex of child, health-seeking behaviour and symptom of 
health condition of child………………………………………………………………………….60 
Table 5.11: Percentage distribution of age of child, health-seeking behaviour and symptom of 
health condition of child………………………………………………………………………….61 
Table 6.1 Multinomial logistic regression model showing the association between health-seeking 
behaviour for diarrhoea and background characteristics of mothers…………………………….66 
Table 6.2: Multinomial logistic regression model showing the association between health-seeking 
behaviour for fever/cough and background characteristics of respondents……………………..70 
 
 
 
 
 
 
 
 
 
xi 
 
University of Ghana http://ugspace.ug.edu.gh
LIST OF FIGURES 
Figure 2.1: Global distribution of causes of under-five mortality, 2012………………………..11 
Figure 2.2: Conceptual Framework showing the relationship between women’s educational 
attainment and health-seeking behaviour for childhood morbidity………………………..........27 
Figure 3.1: Map of Ghana showing administrative regions……………………………..............29 
Figure 4.1: Percentage distribution of under-five children by illness symptoms……………….37 
Figure 4:2: Percent of mothers of under-five children by health condition of children and 
treatment-seeking status…………………………………………………………………………38 
Figure 4.3: Percent of mothers of under-five children by level of education and health condition 
of children………………………………………………………………………………………..39 
Figure 4.4: Percent of mothers of under-five children by religion and health condition of 
children…………………………………………………………………………………………..43 
Figure 4.5 Percentage distribution of mother’s working status by symptom exhibited among 
children…………………………………………………………………………………………..46 
Figure 4.6: Percentage distribution of under-five children by sex and symptom exhibited among 
children………………………………………………………………………………………….48 
 
 
 
 
                                     
  
   
 
 
xii 
 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER ONE 
INTRODUCTION 
1.1 Background of the Study 
Health-seeking behaviour is defined as “any action undertaken by individuals who perceive 
themselves to have a health problem or to be ill for the purpose of finding an appropriate remedy” 
(Olenja, 2003 p. 61). The health-seeking behaviour is preceded by a decision-making process 
influenced by individual socio-demographic and economic characteristics and/or household 
behaviours, community norms, expectations, and provider-related characteristics (Niraula, 1994; 
Bedford and Sharkey, 2014; Kolola et al., 2016). Thus, the nature of health-seeking is not 
homogenous but an interplay of socio-cultural, economic, and provider factors leading to a final 
choice of care-seeking options. 
Childhood illnesses are among the main health problems globally and predominantly in Asian 
and African subcontinents (WHO, 2016). About 6.0 million under-five children died in 2018 
worldwide, which could be prevented with simple and inexpensive interventions (UNICEF, 
2019). The type of health care provider that is sought also differs according to the type of illness. 
With a single mild symptom such as fever, home remedies or folk prescriptions are used (Shaikh 
and Hatcher, 2005). Whiles with a more prolonged period of illness with multiple symptoms, 
allopathic medicine is more likely to be utilized (de Silva et al., 2001; Kolola et al., 2016). 
Similarly, medical pluralism is prevalent in developing countries and is often based on trial and 
error, perceived effectiveness, the uncertainty of illness causation, and the expectation of rapid 
results (D'Souza, 1999; Shaikh and Hatcher, 2005; Abdulraheem, and Parakoyi, 2009). 
1 
 
University of Ghana http://ugspace.ug.edu.gh
Globally, women's educational attainment has improved over the years, and studies suggest that it 
plays an essential role in the health outcomes of women and children (Glewwe, 1999; Shapiro 
and Tenikue, 2017; Dimbuene et al., 2018; Frempong-Ainguah et al., 2018). In their study, 
Caldwell and McDonald (1982) found that educated women had self-confidence and are open to 
practise new behaviours. High educational attainment for females has been shown to reduce 
fertility through (i) delaying age at first marriage and first birth (ii) increasing interest in the 
quality of children relative to quantity and (iii) increasing knowledge and use of modern 
contraceptive methods (Martin, 1995).  
One pathway through which education impacts health is through the increase of health 
knowledge (LeVine et al., 1994; Glewwe, 1999). In-school lessons are taught on hygiene and 
nutrition, and health behaviours such as washing hands before meals are encouraged. Literacy and 
numeracy skills acquired in school is another mechanism that aid mothers absorb health 
information well such as in diagnosing and treating child health problems (Glewwe, 1999; Som, 
2014). There are several studies done in many developing countries on health that show the 
impact of mothers' education on child survival as mothers use prenatal and child health services 
such as vaccination for children (Das Gupta, 1990; Hobcraft, 1993; Buor, 2003). 
Another advantage the educated woman has is financial independence and autonomy (Hill, and 
King, 1995; Acharya et al., 2010; Bour, 2004). In a study of women's participation in primary 
health care programmes in Ghana, Livingstone (1994) asserts that finances are one of the most 
crucial factors in health care utilization by women in developing countries. The educated woman 
is more likely to get a job that carries good pay. Education hence gives the woman the 
economic empowerment to make health-seeking behaviours related decisions freely (Bour, 
2004). It is a tool to demolish the barrier to the use of health care services. 
2 
 
University of Ghana http://ugspace.ug.edu.gh
In Ghana, health-seeking behaviour of mothers for common childhood illnesses may be 
different across the regions, especially because of differences in socio-economic factors. 
Assessing the health-seeking behaviour of mothers and identifying specific factors influencing 
health-seeking are essential to advance strategies and interventions for child survival strategies 
in the country. The study seeks to examine, using secondary data, the extent to which women’s 
educational attainment impacts health-seeking behaviour among children under-five years in 
Ghana. In addition, the study also seeks to identify relevant socio-economic factors that predict 
treatment-seeking by women for common childhood illnesses. 
1.2 Statement of the Problem 
Education worldwide has been thought to provide knowledge about the causes of ill health and 
promote a change in an individual’s behaviour towards health care (MacKian, 2003). The type of 
symptoms experienced for the illness and the number of days of illness are significant determinants 
of health-seeking behaviour and choice of care provider (Shaikh and Hatcher 2005). How women 
reach health care decisions can have a profound effect on morbidity and mortality in the family 
and is, therefore, worthy of continued study. According to the WHO, child mortality and morbidity 
could be reduced by 20% if there is appropriate health-seeking behaviour (UNICEF, 2014). 
Progress towards attaining Millennium Development Goal 4: reducing under-five mortality by 
two-thirds between 1990 and 2015 was not achieved in Ghana. Furthermore, according to Kayode 
et al. (2016), Ghana was below the targeted rate, particularly with respect to neonatal mortality.  
In Ghana, the under-five mortality in 2018 was 47.9 per 1000 live births (UNICEF, 2019) which 
is still high. 
3 
 
University of Ghana http://ugspace.ug.edu.gh
Attainment of the Sustainable Development Goal 3, which is "Ensure healthy lives and promote 
wellbeing for all ages" cannot be achieved without the adoption of positive health behaviour and, 
most importantly, for women as they are primarily responsible for the welfare of the family. "The 
benefits of improving healthcare-seeking behaviour are tremendous, particularly in settings where 
social services and public health resources are limited" (Wado, 2013 p. 1). 
In Ghana, recent studies on health-seeking behaviour for common childhood diseases have been 
done using small samples, resulting in little information on the variability of health-seeking 
behaviour of mothers across the country. Also, with improvement in educational opportunities for 
women, it is assumed that highly educated women are better able to understand the symptoms that 
require treatment for childhood illnesses. However, it is not known whether the increase in 
educational attainment of women in Ghana has an influence on their health-seeking behaviour for 
their children during illness episodes. It is to bridge such a research gap and discover a strategy for 
sustainable development for children's health that it is considered necessary and timely to examine 
the relationship between mothers’ education and health-seeking behaviour during childhood 
morbidities in Ghana, using the 2014 Ghana Demographic and Health Survey (GDHS) data. 
1.3 Research Questions 
The research seeks to answer the following questions: 
i. Do women vary in their health-seeking behaviour relative to illness symptoms for their 
under-five children? 
ii. What is the association between a mother's educational level and health-seeking behaviour 
regarding their under-five children in Ghana? 
4 
 
University of Ghana http://ugspace.ug.edu.gh
iii. What other factors predict mothers’ health-seeking behaviour for their under-five children? 
1.4 Rationale 
Given their natural role in reproduction and nurturing of children, the health status of children 
greatly depends on their mother’s level of empowerment. Globally, access to formal education is 
one of the means to empowering women. Women with greater power through either education, 
wealth accumulation or social standing are able to influence health care decisions for their children 
(Das Gupta, 1990).  A healthy child constitutes the nucleus of a healthy population which is an 
asset to the socio-economic development of every nation. This explains why various global bodies 
such as the UN, WHO and UNICEF often initiate various policies, programmes and agendas such 
as the International Conference on Population and Development (ICPD) and the Sustainable 
Development Goals (SDGs), among others, meant for improving the health, growth and 
development of women and children in most cases.  
The topic is, therefore, very important and timely because of the increasing childhood morbidity 
and mortality in Asia and sub-Saharan Africa in particular. Mothers’ ability to act appropriately 
and promptly to illness symptoms will reduce childhood morbidity and mortality (Sreeramareddy 
et al., 2006). Similarly, a proper understanding of health-seeking behaviour will improve treatment 
compliance, and enhance health promotion strategies. The study adds to the existing literature 
pertaining to the different dimensions of health-seeking behaviour of mothers during childhood 
illnesses in Ghana. Furthermore, results from this study will provide policy makers additional 
knowledge to help shape policy implementation such as the Integrated Management of Childhood 
Illness (IMCI) in reducing infant mortality rate in Ghana. 
5 
 
University of Ghana http://ugspace.ug.edu.gh
1.5 Objectives 
The general objective of the study is to investigate the association between mothers’ educational 
level and health-seeking behaviour with respect to childhood illnesses in Ghana for the purpose of 
strengthening education and empowerment of women for achieving the SDG Goal 3 aimed at 
ensuring healthy lives and wellbeing across all ages. 
The specific objectives are to: 
i. examine the variation in the health-seeking behaviour of women relative to illness 
symptoms for their under-five children. 
ii. assess the relationship between mothers’ education and health-seeking behaviour relative 
to illness symptoms for their under-five children in Ghana. 
iii. identify and examine other factors that predict the health-seeking behaviour of mothers 
with respect to their under-five children. 
1.6 Organization of the study 
This research work is structured in seven chapters. The first chapter introduces the background of 
the study, the problem statement, research questions, study objectives and rationale. The second 
chapter reviews related literature on common childhood illnesses, health-seeking behaviour 
amongst mothers with respect to their children under-five years, theoretical and conceptual 
framework and the study’s hypotheses. Chapter Three discusses the methodology used in the study 
and data limitations. Chapter Four captures the description of background characteristics of the 
respondents. Chapters Five and Six present the results and discussion of key findings of the 
bivariate and multivariate analyses respectively. The final chapter provides a summary of main 
6 
 
University of Ghana http://ugspace.ug.edu.gh
findings, conclusions and recommendations for further research and policy intervention in the 
management of childhood illness in Ghana. 
 
 
  
7 
 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER TWO 
LITERATURE REVIEW 
2.1 Introduction 
This chapter examines empirical studies related to childhood illnesses, women's education, and 
health-seeking behaviour of women when their under-five children are ill, and the theoretical 
framework used in the study. The reviewed literature and theoretical underpinning of the study 
also informed the conceptual framework and hypotheses that are presented. 
2.2 The Burden of Childhood Illnesses and Mortality 
Phenomenal progress in child survival has taken place in the past decades internationally, with 
millions of children having better chances of surviving today. Notwithstanding this development, 
about 5.3 million children under five years died in 2018 (UNICEF, 2019). This translates into 
about 14,000 children dying every day. According to Sreeramareddy et al. (2006), out of all 
childhood deaths in less developed countries, the leading causes are infectious diseases, 
specifically acute diarrhoea diseases (ADD) and acute respiratory infections (ARIs).  In sub-
Saharan Africa, the leading causes of childhood deaths are ARI, diarrhoea, and malaria (Fosu, 
1994). 
Globally, ARIs such as pneumonia and bronchiolitis constitute a significant source of morbidity 
and mortality and the leading cause of under-five mortality (Ujunwa and Ezeonu, 2014 and 
Tazinya et al., 2018). Nearly 1.3 million under-five children die from ARIs worldwide. One-third 
of the deaths are from low-income countries (Tazinya et al., 2018). In 2010, a study on hospital 
admission found that 12 million episodes of hospital admissions were the result of ARIs in young 
8 
 
University of Ghana http://ugspace.ug.edu.gh
children with about 0.3 million deaths. Ninety-nine percent of deaths resulting from ARIs were 
reported in developing countries (Nair et al., 2013). From available health services statistics in 
developing countries, the extent of ARI morbidity is approximately the same as in developed 
countries. However, mortality from ARI, particularly in infants and young children, is 30 or more 
times higher than in developed countries (Leowski, 1986). Some predictors of ARIs in children 
include infection with HIV, passive smoking, exposure to wood smoke, and contact with a person 
having ARI (Ujunwa and Ezeonu, 2014). 
Diarrhoea is defined as “the passage of three or more liquid stools per day, it is usually a symptom 
of an infection in the intestinal tract caused by a variety of bacterial, viral and parasitic organisms” 
(WHO, 2017 p. 1). Diarrhoeal disease is the second leading cause of death in children under five 
years. The World Health Organization (WHO) estimates that about 525,000 children die from 
diarrhoea each year despite the availability of simple and effective treatment. According to Godana 
and Mengiste (2013), two million people die from diarrhoeal diseases in developing countries, 
with children under-five being the most affected. For these children, 90 percent of deaths are 
associated with unsafe drinking water, poor sanitation, and hygiene. In Ghana, the Ministry of 
Health shows that diarrhoeal diseases constitute 84,000 deaths annually, with 25 percent of the 
cases being children under the age of five (Osumanu, 2007). Despite this high number of deaths 
from these causes, there has been some progress. From 2000 to 2017, the annual number of 
diarrhoea deaths among children under-five has declined by 60 percent. A lot more lives could be 
saved through basic interventions; rotavirus vaccination, access to safe drinking water, proper 
hygiene and sanitation (WASH) practices like hand washing and the comprehensive use of oral 
rehydration salts (ORS) and zinc supplements during diarrhoea treatment (UNICEF, 2017) 
9 
 
University of Ghana http://ugspace.ug.edu.gh
Fever or “hot body” is a common symptom of malaria and other acute infections caused by 
bacteria, viruses, or parasites; it is one of the leading causes of death in infancy and childhood in 
many developing countries. Nations where malaria is endemic like Ghana treat fever with anti-
malarial medication (Som, 2014). In 2018, the WHO reported the highest number of malaria cases 
to be in Africa (213 million or 93%), followed by South-East Asia with 3.4% and Eastern 
Mediterranean with 2.1%. Six sub-Saharan African countries make up more than half of all malaria 
cases worldwide: Nigeria (25%), the Democratic Republic of Congo (12%), Uganda (5%) and 
Cote d'Ivoire, Mozambique and Niger (4% each) (World Malaria Report, 2019). Nearly 405,000 
malaria deaths occurred in 2018; children under age five were mostly affected, making up 67% of 
all malaria deaths (WHO, 2019).  Malaria is a crucial cause of death among adults and children in 
Ghana and is responsible for most hospital attendance in the country. It is also the leading cause 
of workday loss due to illness in the country (GSS, 2009; Abdul-Aziz, Harris and Munyakazi, 
2012). Much as malaria is preventable and curable, the disease remains a public health problem in 
Ghana, affecting all age groups, pregnant women and under-five children being the most at risk 
(GSS 2009). 
Figure 2.1 shows the global distribution of causes of under-five mortality in 2012, with pneumonia, 
diarrhoea and malaria accounting for 17%, 9% and 7% child deaths respectively. 
 
 
 
 
 
10 
 
University of Ghana http://ugspace.ug.edu.gh
Fig 2.1: Global distribution of causes of under-five mortality, 2012 
 
Source: UNICEF analysis based on IGME 2013, WHO and CHERG (2013) 
2.3 Strategies for Reducing Childhood Morbidity and Mortality 
2.3.1 Integrated Management of Childhood Illness (IMCI) 
During the mid-1990s, the IMCI was developed by the World Health Organization (WHO) and 
United Nations Children's Fund (UNICEF) to reduce under-five mortality due to diarrhoea, 
pneumonia, malaria, measles, and malnutrition (Arifeen et al., 2009). The IMCI programme was 
later expanded to include interventions to increase child survival at household, community and 
health facility levels, with three components: health worker training, health-systems improvement, 
and enhancing family and community health practices (Armstrong Schellenberg et al., 2004). 
Beyond enhancing health provider skills in managing childhood mortality, the IMCI seeks to 
improve mothers and or caregivers’ health-seeking behaviour. Health professionals are trained to 
educate mothers about signs of illness and counsel them to seek prompt care, home care during 
illness and after recovery (Sreeramareddy et al., 2006). 
Countries like India have adopted the Integrated Management of Childhood Illness Programme by 
including a neonatal element and renamed it the Integrated Management of Neonatal and 
11 
 
University of Ghana http://ugspace.ug.edu.gh
Childhood Illness (IMNIC) (Shewade, Bharti and Shewade, 2012). Its goal is to improve early 
recognition and treatment-seeking for sick neonates and children (Mazumder et al., 2004). 
Findings from studies on the effects of IMNCI show a positive impact on child health indicators. 
For example, a study done in Bangladesh found that after introducing IMCI interventions such as 
improved facility care and community activities, care-seeking behaviour for sick children at 
medical facilities has improved (Arifeen et al., 2009). Following the implementation of the 
Integrated Management of Neonatal and Childhood Illness Programme in India, treatment-seeking 
and exclusive breastfeeding practices have improved, in addition to a decline in hospital 
admissions reporting on morbidities such as diarrhoea and pneumonia during infancy (Mazumder 
et al., 2004). In Tanzania, facility-based IMNCI programmes implemented in two district facilities 
were able to reduce under-five mortality by 13 percent (Armstrong Schellenberg et al., 2004). 
Mukunya et al. (2014) also make a case that inadequate knowledge on community-integrated 
management of childhood illnesses (C-IMCI) of caretakers was associated with wasting and 
stunting among children in Northern Uganda. 
2.3.2 Expanded Programme on Immunization (EPI) 
One strategy to prevent diseases, especially among children, is immunization. The World Health 
Organization's Expanded Programme on Immunization (EPI) aims to significantly reduce 
childhood mortality through immunization against the six killer diseases endemic in developing 
countries: diphtheria-pertussis-tetanus (DPT), poliomyelitis, tuberculosis, and measles. Ada 
(2007) asserts that the emergence of vaccines has reduced the number of deaths due to infectious 
diseases. It is an essential health service for children, reducing under-five deaths by 70% (Mashal 
et al., 2007). Timely vaccination of children is a proven method for saving lives from vaccine-
preventable diseases. This will help achieve some targets of Sustainable Development Goal 3, 
12 
 
University of Ghana http://ugspace.ug.edu.gh
aiming to ensure healthy lives and promote well-being for all. According to UNICEF (n.d.), the 
EPI in Ghana has helped reduce infant mortality, with a significant fall in morbidity due to vaccine-
preventable diseases such as measles and poliomyelitis. Ghana in 2011 was certified as having 
attained elimination status for maternal and neonatal tetanus. 
2.4. Women’s Educational Attainment and Child Health and Mortality 
Among the various factors that are assumed to be determinants of child morbidity and mortality, 
one of the single biggest factors seems to be the educational attainment of mothers. For many 
individuals, high educational attainment is an important social goal. The Millennium Development 
Goal (MDG) 2, which advocates universal primary education, and the Sustainable Development 
Goal 4, which focuses on quality education, are several indicators that demonstrate the vital role 
of education in national growth and development. Education is a significant component of policy 
spending in almost all countries, across the three realms of the Human Development Index 
(education, life expectancy, and per capita income). There has also been a growing understanding 
within the development community about how education plays a crucial role in economic growth. 
The above assertion is not a shocking revelation as education of the human resource has been 
touted to have trickle-down effects on all the other sectors of an economy. Apart from the general 
inherent value of education and the significant role it plays in economic development, there is a 
causal correlation between education and a variety of health consequences, with findings in many 
studies in public health, on the link between educational attainment and child health showing 
strong associations between the educational attainment of mothers and the health and mortality 
rates of their children. 
13 
 
University of Ghana http://ugspace.ug.edu.gh
As put forward by Mark Fischetti, a writer who oversees Scientific American's print and online 
coverage of energy, environment, and sustainability, higher educational attainment of women 
enhances their knowledge on health and diseases and facilitates better use of health care services 
such as prenatal care, immunization and basic hygiene which are critical in improving health 
outcomes for under-five children (Fischetti, 2011). 
Through their education and resultant literacy, women stand a better chance of making the best 
decisions for their children in terms of their education, life choices, and, most importantly, the 
health of the children. Educated women are more likely to be working women and hence, are likely 
to have the economic stability to support their families, including the children. 
Caldwell’s (1979) paper on Nigeria highlighted the importance of mother’s education for child 
survival. In his paper, he referred to many studies that found vast variations in child survivorship 
by educational attainment of mothers but indicated that few works have tried to explain this fact. 
Caldwell (1979) suggests three explanations on how maternal education might boost child 
survival: first, educated mothers move from traditional or ‘fatalistic’ acceptance of ill-health and 
take advantage of other medicinal or curative child care practices; educated mothers are effective 
in manipulating the modern world, including seeking prompt and appropriate health care during 
illness; and a shift in family decision making, allowing the educated woman to participate and 
make contributions on family issues. 
In a 2010 report to examine the relationship between child mortality and educational levels of 
women, the researchers, after collecting data from 915 censuses and surveys, concluded that the 
rise in the education of women had and would have significant public health ramifications over 
the next few decades, leading to more substantial declines over total fertility, and that more 
educated women are more likely to use health services and effective preventive interventions, 
14 
 
University of Ghana http://ugspace.ug.edu.gh
and the increasing educational achievement of women in all areas of sub-Saharan Africa, except 
for the Sahel, is likely to contribute to greater acceptance of new maternal, newborn and child 
approaches, interventions and health-seeking behaviours (Gakidou et al., 2010). 
Similar to this research is one conducted and published by Buor (2003) where he looked into the 
education of mothers and childhood health and mortality in Ghana and found that there was a 
clear inverse association between the education of mothers and childhood mortality. This high 
child mortality associated with women with little or no education may be the product of high 
child morbidity and the lack of health knowledge on the part of the mother to prevent it or treat 
it. Thus, the higher the attainment of education of women, the lower the mortality levels were. 
According to him, children of women with higher education face lower mortality at all levels of 
childhood mortality, with the exceptions being neonatal mortality and infant mortality in which 
the primary-educated mother’s children outnumbered those without schooling (Buor, 2003). 
2.5 Pathways Women’s Education Influences Health-seeking Behaviour 
The health status and health-seeking behaviour of women have a significant impact on both the 
women and their respective children. Many studies have been done about the health-seeking 
behaviour of women, and this results from the fact that women are the primary caregivers at home 
for children (Bour, 2003). Biomedical care plays a critical role in reducing maternal and child 
deaths, and the utilization of maternal and child health services must increase to improve mother 
and child survival (Greenaway et al., 2012). 
Health knowledge is well represented in the published works as a possible medium through which 
formal education is correlated with greater use of modern care services (Fosu, 1994; LeVine & 
Rowe, 2009). Schooling exposes women to diseases causation, hygiene, nutrition, and health care 
15 
 
University of Ghana http://ugspace.ug.edu.gh
practices such as hand washing (Greenaway et al., 2012). The exposure to schooling enhances the 
mother’s knowledge on the importance of modern health care (Fosu,1994). In Rowe et al.’s (2005) 
study, they found that the health knowledge of mothers is closely related with a number of healthy 
behaviours, such as the use of preventive facilities including antenatal care, taking iron tablets 
during pregnancy, and consequently, enhancing child-seeking behaviour, even after birth. 
Schooling also enhances human capital through literacy and problem-solving skills (Ross & 
Mirowsky, 1999). Higher education teaches people to think logically and rationally, see many 
sides of an issue, and analyze and solve problems. This is commonly referred to as the 
‘information-transfer hypothesis’ (Baker et al., 2011). Literate mothers can read and understand 
printed health materials and interact better with health professionals (LeVine et al., 2004). This 
means that educated mothers are in a better position to appreciate the importance of health and 
how to address the challenges associated with poor health and, therefore, can seek appropriate 
health or treatment whenever it becomes necessary compared to their non-literate counterparts. 
Findings from studies in Ghana and Botswana show that maternal education had a strong 
correlation with health services utilization. These countries also had quite high adult female 
literacy. It was also found that educated mothers effectively communicate with health workers 
(Fosu, 1994). Thomas (1999: p. 172) concludes that ''Women with better comprehension skills 
may be better able to access and assimilate information in the community. They may thus be better 
informed than their peers and, therefore, better able to use community services effectively''. 
Studies further hypothesize that formal schooling's content influences women's belief about 
disease causation and cure and promotes the use of modern health care services (Elo, 1992; 
Obermeyer, 1993). Hill et al. (2003) identified strong health beliefs in traditional medicine as 
hindrances in maternal ability to recognize illness symptoms and health-seeking for childhood 
16 
 
University of Ghana http://ugspace.ug.edu.gh
illness in Ghana. Most developing countries have “two or more well-defined systems of medicine 
and health care” (Igun, 1979), leading to pluralistic health treatment (D’Souza, 2003). Caldwell 
and McDonald (1982) claim that women with higher education were exposed to health knowledge 
and more inclined to adopt modern medicine for childcare. 
In these modern times, however, the use of such indigenous health-seeking behaviours seems to 
transcend superstitious beliefs, but may be economically tied or driven, as some women with no 
formal education, most of them without employment and may be engaging in some menial jobs 
are unable to adopt better health-seeking practices for their children and themselves as it may 
require the expending of funds. Education, thus, empowers women through access to better income 
and participation in family decision-making (Bour, 2003). The greatest impediment to the use of 
healthcare is income or lack of money (Elo, 1992; Adulraheem and Parakoyi, 2009). Women with 
income can freely afford medical care for their children. Similarly, in most developing countries, 
women who are educated do demonstrate some autonomy and have greater decision-making power 
at the household level and increases the likelihood of seeking prompt care for childhood illnesses 
(Ojanuja and Gilbert, 1992; Thind and Andersen 2003). All the above benefits of education 
increase women’s value for good health and their demand for modern healthcare services (Fosu 
1994). 
2.5.1 Mother’s Education and Health-seeking Behaviour 
Previous studies in developing countries show that maternal education and health-seeking 
behaviour have a positive relationship (Sreeramareddy et al., 2006; Adulraheem and Parakoyi, 
2009). A study carried out by Fosu (1994) using the 1988/89 Demographic and Health Surveys 
found maternal education had significant effects with health services utilization in Ghana and 
Botswana.  Children of mothers in Ghana with no education were 42% less likely to be taken to a 
17 
 
University of Ghana http://ugspace.ug.edu.gh
medical facility while children of mothers with primary education were 58% less likely to be taken 
to a medical facility compared with children whose mothers had secondary or higher education. 
Education enhances the mother’s access to better job opportunity with higher wages making her 
self-reliant. In rural Nigeria, Adulraheem and Parakoyi, (2009), found low utilization of health 
care associated with mothers with less education although several public and private orthodox 
facilities were available. This could be explained by the high illiteracy and non-belief in orthodox 
healthcare. 
Despite claims that the mother's educational attainment positively influences health-seeking 
behaviour, results from Sir Lanka found no significant association between mother’s education 
and care-seeking for children experiencing common childhood illness (de Silva et al., 2001). This 
is buttressed by findings from Pillai et al. (2003) which found a negative relationship between 
mother’s education and health-seeking behaviour. The reason for this is that educated mothers wait 
for illness symptoms to worsen before attending medical facilities, knowing they were financially 
capable. 
A large body of literature relates mother’s education to child nutrition, immunization, morbidity, 
health-seeking behaviour, and survival. According to Vikram and Vanneman (2020), the influence 
of maternal education on the various child health outcomes may vary since “not all of the 
determinants may be under a mother’s control, especially for indicators that are multifactorial in 
origin” (Vikram and Vanneman, 2020, p. 58). The paper reports that maternal education influences 
child health services such as prenatal care and immunization, as these are under the control of the 
mother. However, for child health outcomes such as stunting, morbidities, and survival, other 
factors interact. Other predictors that influence child health outcomes include environmental 
18 
 
University of Ghana http://ugspace.ug.edu.gh
factors, contagion, and health care supply. Thus, mother's education cannot adequately protect the 
children from these interaction factors. 
2.5.2 Health-seeking Behaviour for mothers and their children 
Many women die due to maternal causes which includes postpartum haemorrhage, obstructed 
labour, hypertensive disorders, and unsafe abortions (Rueda-Clausen et al., 2011). Women are at 
further risk of morbidity and mortality due to low maternal health-seeking behaviour (Akeju et al., 
2016; Kifle et al., 2017). In urban Nigeria “women commonly utilize several obstetric health care 
providers in complement during pregnancy: traditional birth attendants, faith-based providers, and 
orthodox practitioners (nurses, midwives, community health workers, doctors, specialists)” (Akeju 
et al., 2016 p.73).  
Different factors have been found to be associated with the health-seeking behaviours of mothers. 
Majority of determinants categorized as demographic and socio-economic factors such as maternal 
age, educational level, occupation, religion, marital status, women’s autonomy, family size, 
household income, other factors associated are patriarchal culture, health service accessibility, 
cultural norms and beliefs (Abosse. et al., 2010; Akeju et al., 2016; Kifle et al., 2017). There are 
similar determinants of health-seeking behaviour of mothers for themselves and their children, for 
instance mothers age, education, religion, income, place of residence and marital status 
(Adulraheem and Parakoyi, 2009; Diaz et al., 2013; Kolola et al., 2016). 
A number of studies have found that the educational status of mothers has significant influence on 
the health-seeking behaviour of mothers for themselves and their children (Sreeramareddy et al., 
2006; Abosse et al., 2010; Abubakar et al., 2013; Kifle et al., 2017). Notable benefits of educating 
women is its influence on women’s autonomy and income. Education is known to enhance 
19 
 
University of Ghana http://ugspace.ug.edu.gh
women’s autonomy in healthcare decision-making; increasing their knowledge about health issues 
for themselves and children. Women’s autonomy is considered essential in deciding when and 
where to seek treatment. (Osamor and Grady 2016). Financial constraint is another barrier to 
proper healthcare and mothers who are educated are more likely to participate in income-
generating activities to afford healthcare services. Women who are financial empowered can 
independently access resources and make prompt decisions concerning health issues for 
themselves and their children Bour, 2004; Acharya et al., 2010). 
2.5.3 Fathers influence on Health-seeking Behaviour 
Mothers have historically been seen as responsible for healthcare decisions for their children 
(Funk, et al., 2020), however, multiple studies in sub–Saharan Africa attest that fathers are the 
main decisions makers and financial providers in health-seeking for children (Okoko et al., 2006; 
Abubakara at al., 2013; Ellis et al.,2013; Funk, et al., 2020).  For instance, Abubarkar et al. (2013) 
in Kenya found that fathers were the ultimate decision makers on when and where to seek treatment 
for the sick child. Funk et al. (2020) also reports the various roles fathers play during children 
illness episode (i) arranging resources to seek care (ii) (co)- deciding where to seek care (iii) 
accompanying the child to the health facility. 
Plantin et al., (2011) conducted a study to examine how European men’s involvement in pregnancy 
and childbirth can be related to positive health outcomes for the men, their partners and children. 
The study revealed that several research supported the idea that men contribute to better health 
outcomes for themselves, their partners and children. For example, Findings reveal that men who 
become fathers and are involved during pregnancy show better physical and physical and 
psychological health. Dudgeon & Inhorn, (2004) reports that maternal and infant mortality is 
reduced as fathers participate in maternal and child health programs (with respect to pregnancy 
20 
 
University of Ghana http://ugspace.ug.edu.gh
and birth), by offering better preparation in case of obstetric emergencies. In conclusion, Plantin 
et al., (2011) discloses that men are exclude from maternal health services and suggest that a 
broader and more multi-disciplinary scope is required to involve men in maternal and child 
healthcare services, as their involvement has positive influences on the entire family. 
2.6 Different Treatment Sources for Childhood Illness 
Health-seeking is a complex process strongly influenced by socio-economic status health beliefs 
and there are several types of barriers to appropriate healthcare (Niraula, 1994; Hill et al., 2003; 
Bedford and Sharkey, 2014). Barriers to health care include visits to traditional healers, financial 
constraint, perceiving the illness as not serious and the expectation that illness would recover soon 
(Kolola et al., 2016). The importance of appropriate health-seeking is recognized in programmes 
such as IMCI which focus on teaching caregivers to recognize symptoms.  
In many sub-Saharan Africa settings, home management and spiritual/ traditional healers are 
common health care sources before health facility treatment (Charles et al., 2008; Bedford and 
Sharkey, 2014). Health seeking for children with diarrhoea, pneumonia and fever symptoms in 
Serria Leone found that some children were given traditional treatments at home due to preference 
and the lack of alternatives (Diaz et al., 2013). In rural Ghana, mother’s illness classification was 
found to be a significant barrier to care-seeking. Some local illnesses such as asram were 
categorized as ‘not-for-hospital’ because they have unclear biomedical equivalents (Hill et al., 
2003). And due to traditional beliefs mothers are more likely to use traditional medicine to treat 
childhood illness. According to Hill et al.(2003) the use of  local terminology is more  likley 
increase appropirate health seeking among mothers.  
21 
 
University of Ghana http://ugspace.ug.edu.gh
Recent studies show that many more children are taken to medical health facilities during 
childhood illness (de Silva et al., 2001; Diaz et al., 2013; Kolola et al., 2016). In Sri Lanka, health-
seeking behaviour at medical health care services was found as a possible explanation for low 
childhood mortality (de Silva et al., 2001). Results from the study show that 65% of children with 
illness episodes were taken to medical facilities for care and treatment. A study done in rural 
Guatemala reports that "modern medical care plays a major role in the treatment of infectious 
illness among children, with visits to pharmacists, doctors and the staff at government health 
facilities occurring much more frequently than visits to curers and other traditional practitioners” 
(Goldman and Heuveline, 2000: p.145). The reason for this trend could be the increase of formal 
education among mothers and the avaliablity of modern healthcare facilities in developing 
countries leading to exposure to new information and change in behaviour (Caldwell, 1979). 
Studies also report that some mothers do not seek treatment for children who experience 
childhood illnesses (Goldman and Heuveline, 2000; Thind, R. Andersen 2003; Kolola et al. 
2016). Thind and Cruz (2003) found that maternal education, sex of child, asset index and place 
of residence influence the non-use of health care services. Contrastingly, findings from Sri Lanka 
show that socio- economic status of the mother nor child characteristics have any significant 
relationship with health seeking behaviour of mothers for their sick children. However, health 
seeking is responsive to perceived symptoms and severity of the illness, mothers do not seek 
treatment because they perceive that the illness was not serious (de Silva et al., 2001). Kolola et 
al. (2016) report similar findings that perceived severity of the illness was associated with an 
increased odds of health seeking behaviour for childhood illnesses 
 
22 
 
University of Ghana http://ugspace.ug.edu.gh
2.7 Factors Associated with Health-seeking Behaviour for the Child 
The factors associated with health-seeking behaviour include mother’s age, religion, household 
wealth, place of residence, child characteristics and illness symptoms. These factors are 
discussed below. 
2.7.1 Mother’s Age  
Age influences mothers’ decisions to seek treatment outside the home. Taffa and Chepngeno 
(2005) found age of mother to be significant to health-seeking behaviour after ages 35 years.  Older 
mothers were less likely to seek treatment compared to mothers aged 20 -34 years. Again, older 
mothers (40 years and older) tend to seek less medical care for their children than younger women. 
This case may result from previous child-rearing customs and practices which emphasize the role 
of alternative therapies (Adulraheem and Parakoyi, 2009; Taffa and Chepngeno 2005).  
In his study Fosu (1994), suggest that children of younger women are more likely to visit medical 
facilities than children of older women since modern medicine have become more available for 
younger women. For example, young mothers (under 20 years) in Uganda were 61% more likely 
to seek medical treatment compare to older mothers (35 years and above). It is assumed that 
exposure to formal education provides younger mothers knowledge on modern health care 
facilities, and enhance better interaction with health professionals influencing their choose of 
health care during ill-health as oppose to older mothers. 
2.7.2 Religion  
Some studies have shown no significant relation between religion and care-seeking for sick under-
five children (Thind and Cruz, 2003; Pillai et al., 2003; Diaz et al., 2013). Even though no 
23 
 
University of Ghana http://ugspace.ug.edu.gh
association was found, Muslim or Hindu mothers in India had a lower probability of seeking 
medical care compared to Christian mothers (Pillai et al., 2003). Similar findings were found in 
Sierra leone, Muslim mothers were more likely not to seek treatment from medical health facilities 
for under-five children compared to other religious groups (Diaz et al., 2013). 
However, significant associations between religion and health care utilization were found in 
Zimbabwe and Ghana (Fosu, 1994). In Ghana being a Christian had a positive relationship with 
health care use, while in Zimbabwe, Christian mothers had a negative relationship the use of health 
services. 
2.7.3 Household Wealth  
Financial access has been cited as a major barrier to healthcare (Elo, 1992; Hill et al., 2003).  For 
instance, Adulraheem and Parakoyi (2009) found household income and maternal participation in 
income generating activity as a predictor for health-seeking behaviour by mothers. Their study 
revealed that even after the child visited the hospital, lack of funds prevented the child from 
receiving the appropriate prescribed drugs. Ching’s (1995) study in the Philippines reported that 
larger household size was correlated with less use for private medical facilities. His findings 
postulated that the family members' income per capita is reduced due to the larger family size. 
Again, households with high economic status was found to seek less care, especially for mild 
sickness (Pillai et al., 2003). This is plausible as richer families may only visit the health centers 
when illness symptoms continue to persist. 
On the other hand, some studies found no association between household wealth and health-
seeking behaviour.  In Ethiopia, caregivers tend to seek care for under-five children only when 
illness symptoms worsened irrespective of their monthly income (Kolola et al., 2016). Again, 
24 
 
University of Ghana http://ugspace.ug.edu.gh
maternal participation in income- generating activity was not significantly associated with more 
health care seeking for child illnesses (Taffa and Chepngeno (2005). 
2.7.4 Type of Place of Residence  
Studies shown that, children of mothers living in rural areas are less likely to use allopathic 
medicine during childhood illnesses than those residing in urban areas. Thind and Anderson (2003) 
reported a significant effect of place of residence and care-seeking behaviour of mothers. Their 
results show that children of mothers living in rural areas were less likely to use medical services 
than those living in urban areas. This can be attributed to the differences in socio-economic factors 
such as lack of health centers or infrastructure like roads in rural dwellings acting as barriers to 
healthcare use for women. Although type of place of residence was fairly significant, Pillai et al., 
(2003) found that families in rural areas seek care from alternative providers than families in urban 
areas because they have to travel long distances to access medical care. However, after adjusting 
for other variables type of place of residence showed no significant association with health care 
seeking behavior (Kolola et al., 2016). This could be as a result of many health post available in 
both rural and urban areas. 
2.7.5 Child Characteristics 
Healthcare decisions for boys and girls have been found to be different due to gender bias.  Pillai 
et al. (2003) reported from Kerala state (India) that gender was associated with the decision to seek 
care and the choice of allopathic over alternative medical systems. Similar findings in Nepal show 
that more males were reported to be ill and subsequently sought treatment for at allopathic facilities 
compared to females (Pokhrel et al. 2005). According to Adulraheem and Parakoyi (2009), they 
found that child age and sex were significantly associated with healthcare behaviour of the mother. 
25 
 
University of Ghana http://ugspace.ug.edu.gh
Although not statistically significant, mother’s healthcare seeking behaviour for male children was 
a little higher compared to females. And for child’s age, healthcare-seeking behaviour was found 
highest at age 0–12 months and gradually declines for older age groups for both sexes.  Taffa and 
Chepngeno (2005) also found that compared to children less than one year, mothers sought less 
treatment for children aged 1 and 2 years old. Pillai et al., (2003) also found families of younger 
children (less than 6 months) more likely not seek treatment. They explained that young children 
have frequent illness which also resolve on their own. 
On the contrary, Sreeramareddy et al. (2006) reported no correlation between the sex of the child 
and the mother's health-seeking behaviour. Generally, younger children are vulnerable to severe 
consequences of ARI and diarrhoea compared to older children because of their level of immunity. 
Mothers in rural Sir Lanka suggested that younger children who are ill should seek treatment 
promptly, on the account that they are incapable of expressing their illness (de Silva et al., 2001).  
2.7.6 Illness Symptoms and/or Perceived Severity 
The health-seeking behaviour of mothers is driven by symptomology. Perceived symptoms and 
perceived severity are the proximate determinants of health services utilization (Thind and 
Andersen 2003). Healthcare was strongly associated with particular illness symptoms; for 
example, mothers tend to respond quickly to diarrhoeal diseases than cough (Adulraheem and 
Parakoyi, 2009). A study by Sreeramareddy et al. (2006) revealed that children with one illness 
symptom had a lower probability to be taken for hospital care because many of these symptoms 
resolve on their own with time. Goldman and Heuveline (2000) also found that families are much 
more likely to seek biomedical health care when a child experiences fever and gastrointestinal 
symptoms, such as vomiting or diarrhoea relative to respiratory and other symptoms. A possible 
26 
 
University of Ghana http://ugspace.ug.edu.gh
reason for this could be that mothers find frequent episode of cough symptoms as normal illness 
which will pass with time. 
On the other hand, Thind and Andersen (2003) found no significant relationship between illness 
symptoms and the use or non-use of healthcare services. From their findings predisposing and 
enabling factors were the driving force for the utilization of healthcare services in Dominican 
Republic rather than illness symptoms. In Ghana, Hill et al., (2003) argues that poor recognition 
of illness symptoms is important barrier to health-seeking behaviour, as 60% of childhood illness 
episodes were never taken to medical facilities. Their study recognizes the IMCI to provide 
caregivers the education to identify illness symptoms and seek early treatment.  
2.8 Gaps in Literature  
Existing literature on the relationship between mother’s educational attainment and health-
seeking behaviour for childhood illness has showed varied results. Many studies have found a 
significant relation between maternal education and health-seeking behaviour while others found 
no significant effect. Additionally, evidence from literature has identified maternal child 
characteristics, socio-cultural and economic factors predicting health-seeking behaviour for 
childhood illnesses. Much work has been done on health-seeking behaviour globally and in sub-
Saharan Africa. In spite of that, there are very few studies that have examined maternal 
education and health-seeking behaviour for under-five children in Ghana. Women’s educational 
attainment has improved over the years and has been found as the greatest predictor for health-
seeking behaviour for child health outcomes. The study therefore seeks to examine how mother’s 
education influences health-seeking behaviour for under-fives in Ghana, to contribute to the 
global discourse. 
27 
 
University of Ghana http://ugspace.ug.edu.gh
2.9 Theoretical Framework 
This research is established on the Andersen and Newman’s (1973) Health Utilization Model. The 
Model describes the determinants of health services utilization by categorizing them into three 
components as predisposing factors, enabling factors and need factors. These three factors 
demonstrate individual and societal determinants in the utilization of health care services. 
The predisposing component includes (i) demographic variables such as age, sex and marital 
status; (ii) social structure variables include characteristics such as education, race, occupation, 
family size, ethnicity and religion; (iii) health beliefs are attitudes, values, and knowledge that 
diseases, wellbeing and health systems influence health care utilization. 
The enabling component suggests that some resources must exist for health care utilization to 
occur. These include family, income, health insurance, cost and availability of health care while 
the need component considers perceived ill-health as the most proximate cause of health care 
utilization. It consists of the illness symptoms exhibited. 
2.10 Conceptual Framework 
This study seeks to identify whether mother’s educational attainment influences health seeking 
behaviour for childhood illness in Ghana. Adapting the Andersen and Newman’s (1973) Health 
Utilization Model, the study separates mother’s educational level as the independent variable and 
other predisposing, enabling and need factors as control variables in this study. Reviewing existing 
literature and using the Health Utilization Model, the variables in Figure 2.2 have been identified 
as having a direct or indirect impact on the type of care sought for illness.  
28 
 
University of Ghana http://ugspace.ug.edu.gh
Due to data limitation, some variables in the Andersen and Newman’s (1973) framework were not 
used in this study. Variables such as beliefs about health systems, knowledge about diseases, and 
cost and availability of health care could not be investigated using the GDHS data because no 
questions on them were asked in the survey. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
29 
 
University of Ghana http://ugspace.ug.edu.gh
Fig 2.2: Conceptual framework showing the relationship between women’s educational 
attainment and health-seeking behaviour for childhood morbidity. 
 
 
INDEPENDENT 
 
VARIABLE 
DEPENDENT VARIABLE 
 
Mother’s educational 
Health-seeking behaviour 
 attainment 
• Type of treatment 
 
sought 
CONRTOL 
 - No treatment 
VARIABLES 
- Medical facility 
 
Mother characteristic - Non-medical facility 
 Age  
Ethnicity 
 
Religion  
 Marital status  
 
 
Child characteristics 
 Sex 
Age 
 
Enabling resources 
 
Working status 
 
Wealth index 
 Place of residence 
Region of abode 
 
Need 
Illness symptoms 
 
 
Sour ce: Adapted from Andersen and Newman's (1973) Health Utilization Model 
   
 L .    , 
30 
 
University of Ghana http://ugspace.ug.edu.gh
2.11 Hypotheses 
The following are the proposed hypotheses based on the theorical framework and literature review: 
i. Health seeking behaviour of mothers for children who experience diarrhoea symptoms 
differs by educational level.  
ii. Health seeking behaviour of mothers for children who experience fever/cough symptoms 
differs by educational level.  
 
  
31 
 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER THREE 
METHODOLOGY 
3.1 Introduction 
This chapter presents the research design and methodology used in the study. It begins with the 
background information of the study area, source of data, sample design and selection, 
categorization of variables, method of analysis, and limitations associated with the data. 
3.2 Study Area 
Ghana is a developing country located on the West African coast, bordered by three French-
speaking countries: Togo on the east, Burkina Faso to the north, and Côte d'Ivoire on the west. In 
2014, Ghana's population was approximately 27 million (GSS, 2013). The country was 
subdivided into ten administrative regions with 216 districts at the time the survey was 
conducted.  
 
 
 
 
 
 
 
 
32 
 
University of Ghana http://ugspace.ug.edu.gh
Fig 3.1 Map of Ghana showing administrative regions 
 
Source: GSS et al., (2015) 
3.3 Source of Data 
The primary source of data for this study is the 2014 Ghana Demographic and Health Survey 
(GDHS) conducted by the Ghana Statistical Service in collaboration with the Ghana Health 
Service (GHS) and National Public Health Reference Laboratory (NPHRL) of the GHS, with 
funding from USAID and other development partners. The GDHS is a nationally representative 
survey that gathered data on demographic, socioeconomic, and health information on men and 
women in their reproductive ages and children under the age of five years. The primary objective 
of the survey was to collect, analyse, and disseminate information on socioeconomic 
characteristics, educational attainment, maternal and child health, nutrition, family planning, child 
mortality, women’s empowerment indicators and knowledge and behaviour related to HIV and 
AIDS, and sexually transmitted infections (STIs). 
33 
 
University of Ghana http://ugspace.ug.edu.gh
3.4 Sample size and Design 
The sampling frame used in the 2014 GDHS was from the 2010 Ghana Population and Housing 
Census and excluded nomadic and institutional populations such as persons in hotels, barracks, 
and prisons. Participants were selected using a two-stage sampling technique. The first stage 
involved the selection of enumeration areas with a total number of 427. Cluster sampling technique 
was employed to select 216 urban areas and 211 rural areas for the survey. The second stage 
involved a systematic sampling of 30 households from each enumeration area. The total sample 
size was 12,831 households. Men aged 15-59 years and women in their reproductive ages 15 – 49 
years were eligible for the survey. 
The 2014 GDHS consisted of three sets of questionnaires: the household, women's and men's 
questionnaires. The questionnaires were initially produced in English and later translated into 
Akan, Ga, Ewe and Dagbani. A total number of 4,388 males and 9,396 females responded to the 
questionnaires. The survey had a total number of 5,884 under-five children but for this study, only 
children who had diarrhoea, fever and/or cough two weeks preceding the survey were included in 
the analysis. 
3.5 Variables and their Measurements 
This section provides a description of the variables used in the study and how they were measured. 
3.5.1 Dependent Variable 
The outcome variables in this study, which is ‘type of treatment sought’ was gleaned from the 
question "Where did you seek advice or treatment?" The sources of treatment were categorized as 
‘No treatment’, ‘Medical health facility’ (including government hospital/clinic, private 
34 
 
University of Ghana http://ugspace.ug.edu.gh
hospital/clinic, health post/CHPS, mobile clinic and field worker), and ‘Non-medical health 
facility’ (Pharmacy, chemical/drug store, traditional practitioner, provision shop/market and drug 
peddler).  
3.5.2 Independent Variable 
Mother's educational attainment was measured using the level of education, and the question asked 
in the study was "What is the highest level of school you attended?" The responses were 
categorized as ‘No education’, ‘Primary’, ‘Secondary and Higher’. However, in this study, 
educational attainment is recategorized as ‘No education’, ‘Primary’, ‘Secondary/Higher’. Due to 
the small  number of cases Secondary and Higher was merged into one. 
3.5.3 Control Variables 
To better understand the association between a mother's educational attainment and the health-
seeking behaviour, other demographic and socioeconomic variables were included to explain the 
relationship between the dependent and the independent variables. Mother's personal 
characteristics such as age, ethnicity, religion, marital status, occupation, wealth index, type of 
place of residence, the region of residence, and child characteristics such as age and sex of the 
child were included as part of the control variables in the model. 
Table 3.1 below shows the measurement of the control variables used in the study. For this study, 
some variables were re-categorized. Religion was re-grouped into Christian, Islam, Other. 
Catholic, Anglican, Methodist, Presbyterian, Pentecostal/Charismatic and Other Christian were 
put under one category as Christians. People belonging to the category ‘Traditional or Spiritualist’ 
and ‘no religion’ were added to the “other” category. From literature, there is no significant 
difference among the various Christian denomination groups however studies have found 
35 
 
University of Ghana http://ugspace.ug.edu.gh
significant differences when Christian groups are categorized as one (Fosu, 1994 and Diaz et al., 
2013).  Ethnicity was re-categorized into Akan, Ga/Dangme, Ewe, Mole Dagbani, Gurma, and 
others. Mande, Guan, Grusi, and other ethnic groups were merged to form 'other' because their 
individual sample sizes were small compared to the other listed ethnic groups. Mothers’ marital 
status was recodified into never married, currently married, and formerly married. Respondents 
who had never been in any union at the time of the study were categorized under never married. 
Those who were either married or living with their partners make up the currently married 
category, and those who were either widowed, divorced, no longer living together, or separated 
make up the formerly married category. Working status was also re-categorized into 'Working' and 
'Not Working'; those who responded as 'don't know' were added to the 'Not Working' group 
because there was a higher possibility that they were unemployed than employed. The other 
variables used in this study were left as they were measured in the GDHS data set. 
 
  
 
 
 
 
 
 
 
 
 
36 
 
University of Ghana http://ugspace.ug.edu.gh
Table 3.1: Measurement of control variables 
Variables Measurement 
Mother’s age Mother’s age is categorized as 15-24, 25-34 and 35-49 
 
Ethnicity Akan, Ewe, Ga/Dangme, Mole Dagbani, 
 
Others (Guan, /Grussi/Gurma/Mande others) 
Religion Christian, Islam, Other 
 
Marital status Never married, Currently married and Formerly married 
 
Working status Working or Not working 
 
Wealth index Wealth index categorized as Poorest, Poorer, Middle, 
 Richer and Richest 
Type of place of residence Urban or Rural 
 
Region Region of residence are categorized into the ten regions in 
Ghana: Western, Central, Greater Accra, Volta, Eastern, 
 
Ashanti, Brong Ahafo, Northern, Upper West, and Upper 
East Regions 
Child age Age in years categorized as 0, 1, 2, 3 and 4 
 
Child sex Male or Female 
 
Illness symptoms Diarrhoea and Fever/Cough 
Source: Author’s Construct, August 2020 
 
37 
 
University of Ghana http://ugspace.ug.edu.gh
3.6 Method of Analysis 
The descriptive and inferential analysis used in the study was carry out using the statistical analysis 
software package IBM SPSS Statistics (Version 23). The univariate stage showed the proportions 
of respondents with the various characteristics. The descriptive statistics showed the frequencies 
and percentages of each variable in tables and graphs. 
The bivariate analysis was conducted to evaluate the factors associated with the outcome variable 
(health-seeking behaviour). The independent variable (mother's educational level), demographic 
and socio-economic characteristics of respondents were cross-tabulated against the dependent 
variable.  At a 0.05 significant level, the strength of the associations between the independent 
variables and dependent variable was tested using the chi square test together with its 
corresponding degrees of freedom. 
In the multivariate analysis, relationships between the dependent variable and independent 
variables were examined simultaneously. Multinomial logistic regression analysis was employed 
in determining the extent to which the main independent and control variables (demographic, 
socio-economic characteristics) influenced the health-seeking behaviour among children with 
common childhood morbidity. Because the dependent variable has three categories a multinominal 
logistic regression was used. 
3.7 Limitations 
The study was not without limitations. Health beliefs and health services-related factors could not 
be examined using the GDHS data as questions on them were not asked. Lack of data on mother’s 
health-seeking behaviour for herself when sick did not allow for a comparison between health-
seeking behaviour when she is directly affected and when the child is the one taken ill. This 
38 
 
University of Ghana http://ugspace.ug.edu.gh
resulted from the use of secondary data where the researcher is unable to introduce new variables 
or measures which originally were not part of the data. The use of primary data could lend itself 
to the collection of data as would be most appropriate in the form that would inform intended 
analysis in response to the research questions. This is a challenge generally presented when using 
secondary data sources for such studies that may require variables and their 
definitions/measurements which the available dataset may not contain.  
 
 
 
 
 
 
 
 
 
 
 
 
 
39 
 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER FOUR 
BACKGROUND CHARACTERISTICS OF RESPONDENTS 
4.1 Introduction 
Chapter four reports on the univariate analysis of all the variables employed in this study.  The 
background characteristics of mothers and children are described using charts and tables, providing 
information on the percent distribution of respondents by the various categories of the variables. 
4.2 Incidence of illness symptoms 
Figure 4.1 shows the percentage distribution of under-five children that were reported with illness 
symptoms.  Mothers reported that 88.8% and 79.1% of children did not exhibit diarrhoea and 
fever/cough symptoms respectively.  On the other hand, 11.2% and 20.1% of the children were 
reported to have exhibited diarrhoea and fever/cough symptoms respectively two weeks prior to 
the interview. Similar findings were found in Nepal, Sierra Leone and Ethiopia, where more 
children experience fever symptoms compared to diarrhoea (Sreeramareddy et al., 2006; Diaz et 
al.,2013; Kolola et al., 2016) 
 
 
 
 
 
 
 
40 
 
University of Ghana http://ugspace.ug.edu.gh
 
Fig 4.1: Percentage distribution of under-five children by illness symptoms 
 
Source: Generated from 2014, GDHS. 
 
4.3 Health-seeking Behaviour of Mothers 
Out of 638 children that had diarrhoea, it was reported that 30.1% of their mothers did not seek 
treatment, 47.1% sought treatment at medical facility and 22.8% sought treatment at non-medical 
facility. Mothers also reported that 25.8% of them with children having fever/cough symptoms did 
not seek treatment, 52.8% sought treatment at medical facility while 21.4% sought treatment from 
non-medical facilities (Figure 4.2). The results show that majority of mothers (69.9% for diarrhoea 
and 74.2% for fever/cough) sought treatment than no treatment during childhood illness.  
 
 
41 
 
University of Ghana http://ugspace.ug.edu.gh
Fig 4.2: Percent of mothers of under-five children by health condition of children and 
treatment-seeking status 
 
Source: Generated from 2014, GDHS. 
 
 
4.4 Mothers’ Educational Attainment 
The analysis presented in Figure 4.3 reveals that a relatively higher proportion of mothers have 
attained secondary/higher education. For children who experienced diarrhoea symptoms, 47.5% 
of their mothers had secondary/higher education while 19.3% and 33.2% had primary and no 
education respectively. About 53% of mothers of under-five children with fever/cough symptoms 
had secondary/higher education, 19.3% attained primary education and 28.2% had no education. 
The educational level of the mother most likely predicts her knowledge on basic hygiene practices 
and care for their children (Glewwe, 1999).   
 
 
42 
 
University of Ghana http://ugspace.ug.edu.gh
Fig 4.3: Percent of mothers of under-five children by level of education and health condition 
of children 
 
Source: Generated from 2014, GDHS. 
 
4.5 Mother’s Age   
Table 4.1 shows the percentage distribution of mothers by age. The modal age group of mothers 
was 25-34 with 47.5% and 51.7% of them reported to have children with diarrhoea and fever/cough 
symptoms respectively. The least proportion of mothers were in the age group 15-24. The age 
distribution shows that a big proportion of mothers interviewed are under age 35, indicating that 
the country has a youthful age structure.  
 
 
43 
 
University of Ghana http://ugspace.ug.edu.gh
Table 4.1: Percentage distribution of mothers by age and symptom exhibited among children 
  Diarrhoea Fever/Cough 
Mother’s 
ages Frequency Percent Frequency Percent 
15-24 152 23.9 231 19.5 
25-34 303 47.5 614 51.7 
35-49 183 28.6 343 28.9 
Total 638 100.0 1189 100.0 
Source: Generated from 2014, GDHS. 
 
4.6 Marital Status 
From Table 4.2, mothers with children exhibiting diarrhoea and fever/cough symptoms (82.5% 
and 83.4% respectively) were currently in union followed by 10.6% and 8.9% for never in union 
and 6.9% and 7.85% for formerly married. The results, therefore, show that a huge proportion of 
the women were in union. This is expected, as in Africa, we have universal marriage where almost 
every woman is expected to marry in her life-time (Thomas Hakansson, 1985). 
 
 
 
 
 
 
 
44 
 
University of Ghana http://ugspace.ug.edu.gh
Table 4.2: Percentage distribution of mothers by marital status and symptom exhibited 
among children 
  Diarrhoea Fever/Cough 
Marital status Frequency Percent Frequency Percent 
Never in union 68 10.6 106 8.9 
Currently in union 526 82.5 991 83.4 
Formerly union 44 6.9 92 7.8 
Total 638 100.0 1189 100.0 
Source: Generated from 2014, GDHS. 
4.7 Ethnicity 
Based on homogeneity in geographical proximity, traditional lineage systems and other cultural 
practices, ethnicity is categorised into four groups. From Table 4.3, close to half of mothers were 
from the Akan ethnic group (comprising of Asante, Akyem, Akuapem and others), which is the 
largest ethnic group in Ghana (GSS, 2014) constituting 45.0% for diarrhoea and 44.6% for 
fever/cough symptom, followed by Mole Dagbani and Other. The least represented ethnic group 
was Ga/Dangme with 6.1% and 8.2% for mothers with under-five children with diarrhoea and 
fever/cough respectively.  
 
 
 
45 
 
University of Ghana http://ugspace.ug.edu.gh
 
 
Table 4.3: Percent of mothers by ethnicity and symptom exhibited among children 
  Diarrhoea Fever/Cough 
Ethnicity 
Frequency Percent Frequency Percent 
Akan 287 45.0 530 44.6 
Ga/Dangme 39 6.1 97 8.2 
Ewe 55 8.6 164 13.8 
Mole Dagbani 147 23.0 226 19.0 
Other 110 17.3 172 14.5 
Total 638 100.0 1189 100.0 
Source: Generated from 2014, GDHS. 
4.8 Religion 
Religion is an important factor of human behaviour in the Ghanaian society, influencing 
individuals’ beliefs and ways of life. Majority of mothers of children with diarrhoea and 
fever/cough symptoms (67.3% and 75.5% respectively) identified themselves as Christians 
followed by mothers belonging to Islam. The smallest proportions of mothers indicated their 
belonging to the Other Religion (Figure 4.4). It is evident that Christianity is the largest religion 
in Ghana. 
 
 
 
 
46 
 
University of Ghana http://ugspace.ug.edu.gh
Fig 4.4: Percent of mothers of under-five children by religion and health condition of 
children 
 
Source: Generated from 2014, GDHS. 
 
4.9 Region  
From table 4.4, the Ashanti Region had the highest proportion of mothers with children 
experiencing illness symptoms (22.1% for diarrhoea and 20.6% for cough) while the Upper West 
Region had the lowest proportion (3.4% for diarrhoea and cough each). The Ashanti Region at the 
time was the largest region in the country, and, therefore, this is to be expected. 
 
 
 
 
 
 
47 
 
University of Ghana http://ugspace.ug.edu.gh
Table 4.4: Percent of mothers of under-five children by region of abode and health 
condition of children 
  Diarrhoea Fever/Cough 
 Region Frequency Percent Frequency Percent 
Western 38 5.9 82 6.9 
Central 51 8.0 112 9.4 
Greater Accra 63 9.9 185 15.6 
Volta 29 4.5 82 6.9 
Eastern 80 12.5 151 12.7 
Ashanti 141 22.1 245 20.6 
Brong Ahafo 82 12.8 91 7.6 
Northern 107 16.8 157 13.2 
Upper East 26 4.1 43 3.6 
Upper West 22 3.4 41 3.4 
Total 638 100.0 1189 100.0 
Source: Generated from 2014, GDHS. 
 
4.10 Type of Place of Residence 
Place of residence in this study is categorized as urban or rural area. In Ghana, an urban area refers 
to a settlement with a population of 5000 or more whereas a settlement with less than 5000 people 
is defined as a rural area. From Table 4.5, more than half of mothers with children who experience 
diarrhoea (59.9%) and fever/cough symptoms (55.6%) were rural dwellers.  This shows that a 
greater number of sick children live in rural areas. 
 
48 
 
University of Ghana http://ugspace.ug.edu.gh
Table 4.5: Percentage distribution of mother’s place of residence and symptom exhibited 
among children 
  Diarrhoea Fever/Cough 
Place of 
residence Frequency Percent Frequency Percent 
Urban 256 40.1 527 44.4 
Rural 382 59.9 662 55.6 
Total 638 100.0 1189 100.0 
Source: Generated from 2014, GDHS 
 
4.11 Working Status 
Figure 4.5 shows that a large majority of mothers with children experiencing diarrhoea (78.0%) 
and fever/cough symptoms (83.5%) were working at the time of the survey. About one-third of 
mothers were not working. Working mothers are more likely to have access to funds to seek health 
care when their under-five children are sick (Bour, 2004).  
 
 
 
 
 
 
 
 
49 
 
University of Ghana http://ugspace.ug.edu.gh
Fig 4.5: Percentage distribution of mother’s working status by symptom exhibited among 
children 
 
Source: Generated from 2014, GDHS. 
 
4.12 Wealth Index 
From Table 4.6, the highest wealth index (richest) had the least proportion (10.8%) of mothers 
with under-five children exhibiting diarrhoea symptoms while the lowest wealth index (poorest) 
had the highest proportion (26.4%) of mothers with children with these symptoms. 
For children who experienced fever/cough symptoms, mothers in the poorer wealth index had the 
highest proportion (22.4%) while mothers in the richer and richest wealth index had the least 
proportion of 17.3% each. 
 
 
 
50 
 
University of Ghana http://ugspace.ug.edu.gh
Table 4.6: Percentage distribution of mothers by wealth index and symptom exhibited among 
children 
  Diarrhoea Fever/Cough 
Wealth index Frequency Percent Frequency Percent 
Poorest 168 26.4 267 22.4 
Poorer 164 25.7 257 21.6 
Middle 134 20.9 254 21.4 
Richer 104 16.3 205 17.3 
Richest 69 10.8 206 17.3 
Total 638 100.0 1189 100.0 
Source: Generated from 2014, GDHS. 
 
4.13 Sex of child 
The results as shown in Figure 4.6 indicate that a higher proportion of male children experienced 
both diarrhoea and fever/cough symptoms compared to females. This could result from 
behavioural differences by sex at childhood ages with male children more exposed to risks of 
infection as they are more likely than females to roam about picking unhygienic items outside the 
home without supervision by parents or care-givers (van Lunzen, and Altfeld, 2014). 
 
 
 
 
 
51 
 
University of Ghana http://ugspace.ug.edu.gh
Fig 4.6: Percentage distribution of under-five children by sex and symptom exhibited among 
children 
 
Source: Generated from 2014, GDHS. 
 
4.14 Age of the Child 
The age of a child is an important predictor in the study of health-seeking behaviour among under-
five children (Taffa and Chepngeno 2005). Table 4.4 presents the ages of children in years; 
majority of children are between the ages of 1 and 2.  
 
 
 
 
52 
 
University of Ghana http://ugspace.ug.edu.gh
Table 4.7: Percentage distribution of children under five years by single age and symptom 
exhibited among children 
  Diarrhoea Fever/Cough 
Age 
Frequency Percent Frequency Percent 
0 120 18.8 206 17.4 
1 187 29.3 289 24.3 
2 176 27.6 274 23 
3 77 12.1 234 19.6 
4 78 12.3 186 15.7 
Total 638 100.0 1189 100.0 
Source: Generated from 2014, GDHS. 
 
4.15 Conclusions 
From the analysis presented above, 30.1% and 25.8% of children who experienced diarrhoea and 
fever/cough symptoms sought no treatment, while 47.1% and 52.8% sought medical treatment and 
22.8% and 21.4% sought non-medical treatment. A relatively high proportion of mothers of sick 
under-five children had attain secondary/higher education. The modal age group of mothers was 
25-34 years, a big proportion of mothers were currently in union. A greater number of sick children 
live in rural Ghana and more than half of sick under-five children were males.  
 
 
 
 
53 
 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER FIVE 
  MOTHER’S EDUCATIONAL ATTAINMENT AND HEALTH-SEEKING 
BEHAVIOUR 
5.1 Introduction 
Chapter five examines the association between background characteristics and health-seeking 
behaviour among the respondents using cross tabulations with special focus on women’s 
educational attainment. Using the chi-square test, an alpha value of 0.05 was set to show the 
association between a women’s educational attainment alongside other control variables and 
health-seeking behaviour. Consequently, an alpha value less than 0.05 indicates a statistically 
significant association between the independent variables and health-seeking behaviour while an 
alpha value greater than or equal to 0.05 indicates no association. The chapter presents two models 
of bivariate analysis. This arises from the differences in illness symptoms experienced by under-
five children (i.e., diarrhoea and fever/cough). 
5.2: Mother’s Educational and Health-seeking Behaviour 
 Results from Table 5.1 reveal no statistically significant association between mother’s education 
and health-seeking behaviour for both diarrhoea and fever/cough symptoms, suggesting that 
health-seeking behaviour of the mother does not vary significantly with her educational attainment. 
For diarrhoea symptoms, a higher proportion of mothers with no education sought medical 
treatment (52.7%) compared to no treatment (42.2%) and non-medical facility (45.1%).  However, 
for fever/cough symptoms, mothers with secondary/higher education had the highest proportion 
seeking medical treatment (54.0%). Similar findings are reported in Kerala-India and Jeldu 
54 
 
University of Ghana http://ugspace.ug.edu.gh
District-Ethiopia mothers’ educational status was not found to significantly influence health-
seeking behaviour for childhood illness (Pillai et al., 2003; Kolola et al., 2016). 
 
Table 5.1: Percent of mothers by educational level, health-seeking behaviour and symptom 
of health condition of child 
Health-seeking behaviour  Health-seeking behaviour  
  diarrhoea (%) fever/cough (%) 
Mother's  No   Non-  No   Non-
education treatment Medical medical  Total treatment Medical  medical  Total 
No 
education 27.0 52.7 20.3 212 22.5 53.3 24.2 336 
Primary 30.8 42.2 27.0 123 27.3 48.7 24.0 229 
Secondary/ 
Higher 31.9 45.1 22.9 303 27.0 54.0 19.0 624 
X2= 4.677 df= 4 p=0.322     X2=6.658 df= 4 p=0.155   
Source: Generated from 2014, GDHS. 
 
5.3: Mother’s Age and Health-seeking Behaviour 
Mother’s age was found to have no statistically significant association with health-seeking 
behaviour for children with diarrhoea symptoms. A higher proportion of mothers in the various 
age groups, however, reported to have sought treatment for their children with diarrhoea symptoms 
(Table 5.2). For instance, 51.1% of mothers age 25-34 sought treatment at medical facility whiles 
55 
 
University of Ghana http://ugspace.ug.edu.gh
22.7% sought treatment at non-medical facility. In contrast, mother’s age was found to be 
significantly associated with treatment-seeking behaviour for fever/cough episodes. Majority of 
mother’s aged 34 and below (46.9% and 57.9%) sought medical treatment for their children with 
fever/cough symptoms. Findings support studies that mother’s age influences her decisions to seek 
treatment. Taffa and Chepngeno (2005) found that mothers aged 20-34 years were more likely to 
seek biomedical treatment for sick children compared to older mothers 35 years and above.   
 
Table 5.2: Percent of mothers by age, health-seeking behaviour and symptom of health 
condition of child 
Health-seeking behaviour Health-seeking behaviour 
  diarrhoea (%)   fever/cough (%)   
  Non-   Non-
Mother's  No Medical medical  No Medical medical 
age treatment facility facility Total treatment facility facility Total 
15-24 32.8 47.0 20.3 152 30.0 46.9 23.1 231 
25-34 26.2 51.1 22.7 303 23.1 57.9 19.0 614 
35-49 34.3 40.4 25.3 183 27.7 47.6 24.7 343 
X2=6.765 df= 4 p=0.149   X2=13.852 df= 4 p= 0.008   
Source: Generated from 2014, GDHS. 
 
56 
 
University of Ghana http://ugspace.ug.edu.gh
5.4: Marital Status and Health-seeking Behaviour 
From Table 5.3, marital status was not statistically significant with health-seeking behaviour. 
About half (48.9%) of currently married mothers sought treatment at medical facility for diarrhoea 
symptoms followed by never married mothers (39.9%). For fever/cough symptoms, a higher 
proportion of never married mothers (52.9%) and currently married mothers (53.2%) sought 
treatment at medical facility compared to formerly married mothers. 
 
 
 
 
 
 
 
 
 
57 
 
University of Ghana http://ugspace.ug.edu.gh
Table 5.3: Percent of mothers by marital status, health-seeking behaviour and symptom of 
health condition of child 
Health-seeking behaviour Health-seeking behaviour 
  diarrhoea (%)   fever/cough (%)   
 Non-  Non-
Marital  No  Medical medical  No  Medical medical 
Status treatment facility facility Total treatment facility facility Total 
Never in 
union 38.0 39.9 22.1 68 32.5 52.9 14.6 106 
Currently 
married 28.8 48.9 22.4 526 25.0 53.2 21.7 991 
Formerly 
married 34.1 36.5 29.4 44 26.1 47.5 26.4 92 
X2=5.202 df=4 p=0.267   X2= 5.948 df= 4 p= 0.203   
 
Source: Generated from 2014, GDHS. 
 
5.5: Ethnicity and Health-seeking behaviour 
Bivariate analysis showed no significant effects of ethnicity on health-seeking behaviour for 
diarrhoea symptoms. However, ethnicity had a significant association with health-seeking 
behaviour for fever/cough symptoms. Mothers of Mole Dagbani had the highest proportion of 
mothers seeking treatment at a medical facility. In contrast, Ga/Dangme mothers had the highest 
proportion that sought no treatment while Ewe mothers reported the highest proportion seeking 
non-medical treatment for children who exhibited fever/cough symptoms. Taffa and Chepngeno 
(2005) also reported that maternal ethnicity positively influences health-seeking for childhood 
illnesses. 
58 
 
University of Ghana http://ugspace.ug.edu.gh
Table 5.4: Percent of mothers by ethnicity, health-seeking behaviour and symptom of health 
condition of child 
Health-seeking behaviour Health-seeking behaviour 
  diarrhoea (%)   fever/cough (%)   
  Non-   Non-
 No Medical medical  No Medical medical 
Ethnicity treatment facility facility Total treatment facility facility Total 
Akan 30.2 43.5 26.3 287 26.0 51.1 22.9 530 
Ga/Dangme 28.8 43.0 28.2 39 41.5 45.1 13.4 97 
Ewe 31.8 45.5 22.7 55 27.1 47.6 25.3 164 
Mole 
Dagbani 30.0 51.6 18.5 147 15.3 61.7 23.0 226 
Other 29.6 52.6 17.9 110 28.9 55.5 15.6 172 
X2= 6.965 df=8 p=0.540     X2=32.352 df= 8 p= 0.000   
Source: Generated from 2014, GDHS. 
 
5.6: Religion and Health-seeking behaviour 
The results depicted in Table 5.5 confirm that religion has a statistically significant association 
with health-seeking behaviour relative to diarrhoea and fever/cough symptoms. This supports 
findings by Fosu (1994) that religion had significant effects on health care use. Results in Ghana 
from Fosu (1994) found that Christian mothers were more likely to seek treatment at health facility 
compared to other religions. In this study, mothers belonging to the Islam religion had the highest 
proportion seeking medical facility compared to mothers belonging to Christian and Other religion 
for both childhood illness symptoms. 
 
59 
 
University of Ghana http://ugspace.ug.edu.gh
Table 5.5: Percent of mothers by religion, health-seeking behaviour and symptom of health 
condition of child 
Health-seeking behaviour  Health-seeking behaviour 
  diarrhoea (%) fever/cough (%) 
 No   Non-  No  Non-
Religion treatment Medical  medical  Total treatment  Medical  medical  Total 
Christian 30.2 47.3 22.5 430 27.1 52.2 20.8 898 
Islam 25.1 52.8 22.1 167 18.1 58.3 23.6 215 
Other 48.9 21.7 29.4 42 32.4 44.2 23.4 76 
 
X2= 14.522    df= 4     p= 0.006   X2= 9.956    df= 4     p= 0.041  
Source: Generated from 2014, GDHS. 
 
5.7 Region and Health-seeking Behaviour 
According to Table 5.5, there are regional variations in health-seeking behaviour. Region of abode 
of mothers was found to be highly statistically significant in association with health-seeking 
behaviour for both illness symptoms. For diarrhoea symptoms, mothers in the Volta Region 
reported the highest percentage (36.9%) seeking no treatment; mothers residing in Western Region 
had the highest proportion (75.9%) seeking medical treatment while Ashanti Region had the 
highest proportion (33.9%) seeking non-medical facility treatment for their children having 
diarrhoea. For fever/cough illness, Upper West Region had the lowest proportion (5.2%) seeking 
treatment from non-medical facility and the highest (68.8%) at medical facility. No treatment was 
60 
 
University of Ghana http://ugspace.ug.edu.gh
highest (32.5%) among mothers of Greater Accra Region.  This is surprising as the Greater Accra 
Region has more health facilities compared to other regions of the country. 
Table 5.6: Percent of mothers by region, health-seeking behaviour and symptom of health 
condition of child 
Health-seeking behaviour Health-seeking behaviour  
   diarrhoea (%) fever/cough (%) 
  Non-   Non-
 No Medical medical  No Medical medical 
Region treatment facility facility Total treatment facility facility Total 
Western 20.1 75.9 4.0 38 16.7 67.6 15.7 82 
Central 19.4 50.4 30.1 51 22.7 65.1 12.2 112 
Greater 
Accra 33.4 39.0 27.6 63 32.5 43.5 24.0 185 
Volta 36.9 41.7 21.3 29 23.6 52.1 24.3 82 
Eastern 33.7 44.8 21.5 80 32.3 48.6 19.1 151 
Ashanti 32.8 33.3 33.9 141 27.8 44.4 27.8 245 
Brong 
Ahafo 32.1 49.6 18.2 82 20.9 61.7 17.4 91 
Northern 29.0 52.3 18.7 107 19.8 50.9 29.3 157 
Upper 
East 23.1 59.6 17.3 26 24.2 68.3 7.6 43 
Upper 
West 28.7 66.7 4.6 22 26.0 68.8 5.2 41 
  
X2= 41.637   df= 18   p= 0.001 X2= 56.969      df= 18      p= 0.000 
Source: Generated from 2014, GDHS. 
61 
 
University of Ghana http://ugspace.ug.edu.gh
5.8: Type of Place of Residence and Health-seeking Behaviour 
The chi-square test indicates that association between type of place of residence and health-seeking 
behavior for each illness symptom is not statistically significant. Comparing to mothers in urban 
areas, a higher proportion of mothers in rural areas sought treatment for diarrhoea (51.0%) and 
fever/cough (53.6%) symptoms at a medical facility. It is also observed that a higher number of 
mothers in both rural and urban areas sought no treatment for diarrhoea symptoms compared to 
fever/cough symptoms. 
 
Table 5.7:  Percent of mothers by place of residence, health-seeking behaviour and symptom 
of health condition of child 
Health-seeking behaviour  Health-seeking behaviour 
  diarrhoea (%)  fever/cough (%) 
  Non-   Non-
Place of  No Medical medical  No Medical medical 
residence treatment facility facility Total treatment facility facility Total 
Urban 34.2 41.2 24.6 256 25.9 51.7 22.4 527 
Rural 27.4 51.0 21.6 382 25.7 53.6 20.7 662 
X2= 5.728 df= 2 p=0.057     X2= 0.580 df= 2 p= 0.748   
Source: Generated from 2014, GDHS. 
 
 
62 
 
University of Ghana http://ugspace.ug.edu.gh
5.9: Working Status and Health-seeking Behaviour 
Mothers who are engaged in economic activity are most likely to afford healthcare services. 
Adulraheem and Parakoyi (2009) found that mother’s working status was a predictor for care-
seeking for her child. However, from Table 5.8, no significant association was found in the 
relationship between working status of mothers and their health-seeking behaviour.  The results 
from the analysis reveal that for both symptoms, a higher proportion of working mothers sought 
medical treatment compared to non-working mothers. 
 
Table 5.8:  Percent of mothers by working status, health-seeking behaviour and symptom of 
health condition of child 
Health-seeking behaviour Health-seeking behaviour  
   diarrhoea (%) fever/cough (%) 
  Non-   Non-
Working  No Medical medical  No Medical medical 
status treatment facility facility Total treatment facility facility Total 
Not 
working 35.5 39.8 24.7 140 30.7 46.7 22.6 196 
Working 28.6 49.1 22.3 498 24.8 54.0 21.2 993 
X2=4.084 df= 2 p=0.130     X2= 3.831 df= 2 p= 0.147   
Source: Generated from 2014, GDHS. 
 
 
63 
 
University of Ghana http://ugspace.ug.edu.gh
5.10: Wealth Index and Health-seeking Behaviour 
Wealth index is found to be highly significant statistically in its association with health-seeking 
behaviour of mothers relative to their children with diarrhoea but not for fever/cough symptoms 
(Table 5.9). The results are generally congruent with literature that show the importance of 
household wealth in determining the use of health care. For example, mothers from the richest 
wealth index had the highest proportion seeking medical treatment (58.1%) for fever/cough 
symptoms while poorest mothers had the least proportion (47.6%) for the same purpose. In 
contrast, however, for children who experienced diarrhoea, the use of medical facility was highest 
among poorest mothers and no treatment is seen to be highest for richest mothers. This, however, 
supports findings from Pillai et al., (2003), who suggest that richer families are less likely to seek 
medical care but only visit medical facilities if symptoms persist for a longer period. 
 
 
 
 
 
 
 
 
 
 
 
64 
 
University of Ghana http://ugspace.ug.edu.gh
Table 5.9:  Percent of mothers by wealth index, health-seeking behaviour and symptom of 
health condition of child 
Health-seeking behaviour Health-seeking behaviour  
   diarrhoea (%) fever/cough (%) 
  Non-   Non-
Wealth  No Medical medical  No Medical medical 
index treatment facility facility Total treatment facility facility Total 
Poorest 27.8 53.2 19.0 168 28.8 47.6 23.6 267 
Poorer 29.5 50.2 20.3 164 23.5 50.9 25.6 257 
Middle 27.0 36.8 36.2 134 25.6 56.0 18.4 254 
Richer 33.2 42.9 23.9 104 25.5 52.4 22.0 205 
Richest 38.5 50.9 10.6 69 25.3 58.1 16.6 206 
X2=24.527   df= 8       p=0.002 X2= 10.892   df= 8     p= 0.208 
      
Source: Generated from 2014, GDHS.5.11: Sex of Child and Health-seeking Behaviour 
Table 5.10 shows the association between sex of child and health-seeking behaviour during 
childhood illnesses. There is no significant association between sex of the child and health-seeking 
behaviour of the mother. From the results, a higher proportion of female children were taken to 
medical facility compared to males. This supports findings in Western Nepal. Sreeramareddy et 
al. (2006), reported that between sex of child and health-seeking behaviour of the mother, there 
was no association. 
65 
 
University of Ghana http://ugspace.ug.edu.gh
 
Table 5.10:  Percentage distribution of sex of child, health-seeking behaviour and symptom 
of health condition of child 
Health-seeking behaviour Health-seeking behaviour 
   diarrhoea (%)  fever/cough (%) 
 Non-   Non-
Sex of  No  Medical medical  No Medical medical 
child treatment facility facility Total treatment facility facility Total 
Male 28.9 46.4 24.7 371 27.7 51.3 21.0 634 
Female 31.8 48.0 20.2 267 23.6 54.5 21.9 555 
X2= 1.964    df= 2      p=0.375 X2= 2.631   df= 2    p= 0.268 
      
Source: Generated from 2014, GDHS. 
5.12: Age of Child and Health-seeking Behaviour 
The analysis show that age of child is statistically highly significant in its association with mother’s 
health-seeking behaviour.  The results in Table 5.11 agree with Adulraheem and Parakoyi (2009) 
that a child’s age has a strong association with the mother’s health-seeking behaviour. The results 
show that mothers of children of age one year had the highest proportion seeking treatment at 
medical facility for diarrhoea (59.0%) and fever/cough (58.0%) symptoms. The use of medical 
facility is, however, lowest for children of ages three and four years: 34.2% and 33.2% respectively 
for diarrhoea and 47.8% and 48.7% for fever/cough.  
 
66 
 
University of Ghana http://ugspace.ug.edu.gh
Table 5.11:  Percentage distribution of age of child, health-seeking behaviour and symptom 
of health condition of child 
Health-seeking behaviour Health-seeking behaviour 
   diarrhoea (%)  fever/cough (%) 
Age of   Non-   Non-
child in  No Medical medical  No Medical medical 
years treatment facility facility Total treatment facility facility Total 
0 39.1 44.6 16.4 120 26.6 54.9 18.4 206 
1 22.9 59.0 18.1 187 19.5 58.0 22.4 289 
2 27.6 47.9 24.5 176 21.7 52.5 25.8 274 
3 29.7 34.2 36.1 77 35.4 47.9 16.8 234 
4 39.5 33.2 27.4 78 28.5 48.7 22.7 186 
X2= X2= 
31.548 df= 8 p= .000     25.156 df= 8 p= 0.001   
Source: Generated from 2014, GDHS. 
 
  
67 
 
University of Ghana http://ugspace.ug.edu.gh
5.13 Conclusions 
The main independent variable which is mother’s educational attainment was not statistically 
significant for both diarrhoea and fever/cough symptoms at the bivariate analysis level. This 
indicated that health-seeking behaviour does not vary with the educational level attainted by 
mothers.  
Again, other background characteristics such as mother’s age, ethnicity, religion, region of 
abode, wealth index and age of child were statistically significant predictors for health-seeking 
behaviour for childhood illnesses in Ghana.  
 
 
 
 
 
 
 
 
68 
 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER SIX 
RELATIONSHIP BETWEEN MOTHERS’ BACKGROUND CHARACTERISTICS AND 
HEALTH-SEEKING BEHAVIOUR FOR UNDER-FIVE CHILDREN 
6.1 Introduction 
This chapter presents and discusses the multivariate analysis to assess the influence of background 
characteristics of mothers and children on health-seeking behaviour during childhood illnesses. 
This chapter has three sections. The first section examines background characteristics on health-
seeking-behaviour for diarrhoea symptoms while the second examines background characteristics 
on health-seeking behaviour for fever/cough symptoms. The last section discusses key findings 
from the analysis. 
The multinomial logistic regression model is used for a nominal categorical dependent variable 
with more than two categories. For this study, the dependent variable (health-seeking behaviour) 
has three categories as no treatment, medical facility treatment and non-medical facility treatment. 
For all the models, no treatment was the reference category for the dependent variable.   
To interpret the model, the study uses the Odds Ratios obtained by taking the exponent of the beta 
estimate to understand the association between the variables. The reference category of each 
variable was labelled as RC in the analysis and was set at 1.0 for the comparison. An Odds Ratio 
greater than 1.0 represents a higher likelihood of experiencing the outcome compared to the 
reference category while an Odds ratio less than 1.0 means a less likelihood of experiencing the 
outcome compared to the reference category. 
69 
 
University of Ghana http://ugspace.ug.edu.gh
6.2 Influence of Background Characteristics on Health-seeking Behaviour for Diarrhoea 
This section uses two models to predict the factors associated with health-seeking behaviour for 
children with diarrhoea symptoms. The first model examines the effect of mothers’ education on 
their health-seeking behaviour. In the second model, mother’s education (main independent 
variable) together with the other demographic and socio-economic variables are put into the 
regression model to assess their influence on health-seeking behaviour of the mother. 
The results from table 6.1 presents the multinomial logistic regression revealing the effect of the 
background characteristics of respondents on health-seeking behaviour for diarrhoea symptoms of 
children. For model 1, the overall model level of significance was 0.310, which means the 
relationship is statistically not significant. The Nagelkerke R2 (0.009) indicates only 0.9% of the 
variation in health-seeking behaviour is explained by mother’s education. This suggests that other 
variables (demographic and socio-economic) can better explain the variations in the model. The 
results of model 1 show that there was no statistically significant relationship between mother’s 
education and health-seeking behaviour (p>0.05).  
In model 2, the Nagelkerke R2 value (0.233) shows that 23.3% of the variation in health-seeking 
behaviour was explained by the independent variables. The model significance was high (p=0.000) 
compared to Model 1.  Like in Model 1, maternal education was not found to be statistically 
significant in predicting health-seeking behaviour (p>0.05) in situations when children present 
diarrhoea symptoms.  
Children of mothers aged 25-34 were more likely (1.84) to seek treatment at a medical facility than 
seek no treatment compared to their peers whose mothers are aged 35-49 years, the results depicted 
a statistical significance at p<0.05. Also, compared to mothers belonging to Other religion, 
70 
 
University of Ghana http://ugspace.ug.edu.gh
Christian and Muslim mothers were more likely to seek medical treatment than seek no treatment 
for their sick children of diarrhoea (OR=6.022 and 8.148 respectively). Children aged one year 
were significantly more likely to be taken to medical facility than seek no treatment compared to 
children aged four years.  
With respect to non-medical facility, mother’s education was not found to have a statistically 
significant relationship with health-seeking behaviour. Wealth index showed a high level of 
significant relationship with health-seeking behaviour. Mothers from poorest, poorer, middle and 
rich wealth groups were more likely to seek non-medical treatment than seek no treatment for their 
children with diarrhoea symptoms compared to mothers of richest wealth group. Findings show 
that as wealth index increase the odds of seeking non-medical treatment decreases, suggesting that 
poorer wealth households tend to use non-medical care for sick child. However, mothers from 
middle wealth index has the greatest likelihood in seeking non-medical treatment. 
No statistically significant effect was found in respect of marital status, ethnicity, region, place of 
residence, working status and sex of child in relation to health-seeking behaviour of mother for 
children exhibiting diarrhoea symptoms in Ghana. 
 
71 
 
University of Ghana http://ugspace.ug.edu.gh
Table 6.1: Multinomial logistic regression model showing the association between health-
seeking behaviour for diarrhoea and background characteristics of mothers 
Health-seeking behaviour for diarrhoea 
    Model 1         
Medical facility vs No Non-medical facility vs No 
treatment treatment 
Exp 
(β)  95% CL Exp (β)  95% CL 
Variable   LB UB   LB UB 
Mother's education        
No education 1.378 0.913 2.08 1.046 0.634 1.727 
Primary 0.968 0.591 1.585 1.22 0.698 2.133 
Secondary/Higher (RC) 1.000 - - 1.000 -  - 
Nagelkerke R = 0.009   P-value = 0.310 
Model 2 
Medical facility vs No Non-medical facility vs No 
treatment treatment 
Exp 
(β)  95% CL Exp (β)  95% CL 
Variable   LB UB   LB UB 
Mother's education        
No education 1.318 0.712 2.438 1.162 0.567 2.381 
Primary 0.712 0.403 1.259 1.097 0.58 2.074 
Secondary/Higher (RC) 1.000 - - 1.000 -  - 
Mother's age        
15-24 1.477 0.798 2.731 0.982 0.48 2.007 
25-34 1.84* 1.143 2.963 1.33 0.769 2.3 
35-49 (RC) 1.000 - - 1.000 - -  
Marital Status        
Never in union 0.974 0.359 2.643 0.712 0.237 2.138 
Currently married 1.589 0.71 3.554 1.094 0.465 2.572 
Formerly married (RC) 1.000 - - 1.000 -  - 
Ethnicity        
Akan 0.974 0.473 2.004 1.598 0.661 3.861 
Ga/Dangme 1.446 0.464 4.513 2.015 0.525 7.736 
Ewe 1.197 0.454 3.152 1.458 0.449 4.733 
Mole Dagbani 0.532 0.267 1.06 0.884 0.371 2.102 
Other (RC) 1.000 - - 1.000 -  - 
Religion        
Christian 6.022* 2.257 16.064 1.131 0.446 2.868 
Islam 8.148* 2.929 22.667 1.97 0.697 5.571 
Other (RC) 1.000  -  - 1.000  -  - 
72 
 
University of Ghana http://ugspace.ug.edu.gh
Table 6.1 contd. 
Medical facility vs No Non-medical facility vs No 
treatment treatment 
Exp (β)  95% CL Exp (β)  95% CL 
Variable   LB UB   LB UB 
Region        
Western 1.961 0.462 8.333 1.235 0.071 21.559 
Central 1.144 0.276 4.745 9.613 0.861 107.368 
Greater Accra 0.554 0.138 2.229 7.167 0.645 79.631 
Volta 0.583 0.122 2.78 3.037 0.239 38.572 
Eastern 0.594 0.159 2.227 3.552 0.337 37.389 
Ashanti 0.492 0.143 1.7 7.787 0.801 75.739 
Brong Ahafo 0.711 0.207 2.444 3.023 0.303 30.151 
Northern 0.947 0.297 3.017 4.034 0.433 37.586 
Upper East 1.592 0.374 6.774 4.613 0.382 55.732 
Upper West (RC) 1.000  -  - 1.000  -   - 
Place of residence        
Urban 0.65 0.369 1.147 1.268 0.675 2.38 
Rural (RC) 1.000  -  - 1.000  -   - 
Working status        
Not working 0.611 0.369 1.012 1.09 0.607 1.957 
Working (RC) 1.000  -  - 1.000  -   - 
Wealth index        
Poorest 0.92 0.317 2.667 6.092* 1.595 23.277 
Poorer 0.902 0.353 2.307 4.716* 1.404 15.836 
Middle 0.825 0.356 1.913 7.198* 2.445 21.188 
Richer 0.949 0.439 2.048 2.842* 1.016 7.95 
Richest (RC) 1.000  -  - 1.000  -   - 
Sex of child        
Male 1.18 0.793 1.755 1.272 0.795 2.035 
Female (RC) 1.000  -  - 1.000  -   - 
Age of child in years        
0 1.248 0.591 2.634 0.706 0.295 1.686 
1 2.934* 1.445 5.958 1.367 0.612 3.055 
2 2.063 1.033 4.119 1.453 0.678 3.115 
3 1.359 0.587 3.148 2.088 0.876 4.978 
4 (RC) 1.000  -   - 1.000   -   - 
Nagelkerke R = 0.233   P-value = 0.000 
 CI= Confidence Interval LB= Lower Bound UB= Upper Bound N= sample size  
*P< 0.05 N=638 
73 
 
University of Ghana http://ugspace.ug.edu.gh
6.3 Influence of Background Characteristics on Health-seeking Behaviour for Fever/cough 
This section also uses two models to assess how maternal education affects their health-seeking 
behaviour regarding their children with symptoms of fever/cough. Table 6.2 shows the results of 
the multivariate analysis for health-seeking behaviour regarding children with fever/cough 
symptoms. The Nagelkerke R2 value of 0.006 reveals only 0.6% of the variation in health-seeking 
is explained in terms of mother’s education. From Model 3, the mother’s education was found to 
be significantly related to health-seeking for their children with fever/cough symptoms. The results 
show that children whose mothers had no education were more likely (OR=1.531) to seek 
treatment from non-medical facility than no treatment for their children who experience 
fever/cough symptoms compared to mothers with secondary/higher education. This suggest that 
mothers with no education are more likely to find non-medical treatment more effective than 
allopathic medicine probably due to lack of knowledge on disease causation and treatment. 
Model 4 indicates a Nagelkerke R2 value of 0.172, meaning that 17.2% of the variation in health-
seeking behaviour is explained by the independent variables. From Model 4, there was no 
statistically significant relationship between mother’s education and health-seeking behaviour 
when it comes to children with fever/cough symptoms. In contrast, a mother’s ethnicity had 
statistically significant effects on health-seeking behaviour. Mole Dagbani mothers were 1.797 as 
likely to seek medical treatment than seek no treatment compared to mothers from ‘other’ ethnic 
groups. Another predictor of health-seeking behaviour was type of place of residence. Mothers 
who reside in urban areas were less likely (OR= 0.633) to seek medical treatment than seek no 
treatment compared to mothers dwelling in rural areas. Lastly, mothers without work were less 
likely to seek treatment from medical facility than seek no treatment for their children exhibiting 
fever/cough symptoms compared to working mothers.  
74 
 
University of Ghana http://ugspace.ug.edu.gh
Children whose mothers belong to the poorest, poorer and middle households were less likely to 
take their children with fever/cough symptoms to a medical facility for treatment than seek no 
treatment for them compared to mothers from the richest households. Children aged one year were 
1.726 times as likely to seek medical treatment than seek no treatment for their children exhibiting 
fever/cough symptoms compared to mothers with children of age four years. 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
75 
 
University of Ghana http://ugspace.ug.edu.gh
Table 6.2: Multinomial logistic regression model showing the association between health-
seeking behaviour for fever/cough and background characteristics of mothers 
Health-seeking behaviour for fever/cough 
Model 3 
Medical facility vs No Non-medical facility vs No 
treatment treatment 
Exp (β)  95% CL Exp (β)  95% CL 
Variable   LB UB   LB UB 
Mother's education             
No education 1.183 0.854 1.639 1.531* 1.035 2.264 
Primary 0.891 0.621 1.277 1.246 0.809 1.920 
Secondary/Higher (RC)  1.000  -  - 1.000  -  - 
  Nagelkerke R = 0.006   P-value = 0.158     
Model 4 
Medical facility vs No Non-medical facility vs No 
treatment treatment 
Exp (β)  95% CL Exp (β)  95% CL 
Variable   LB UB   LB UB 
Mother's education         
No education 0.964 0.608 1.529 1.269 0.727 2.213 
Primary 0.984 0.646 1.499 1.22 0.74 2.011 
Secondary/Higher (RC) 1.000  -  - 1.000  -  - 
Mother's age         
15-24 0.877 0.548 1.403 1.077 0.628 1.848 
25-34 1.348 0.955 1.903 0.967 0.64 1.462 
35-49 (RC) 1.000  -  - 1.000  -  - 
Marital Status        
Never in union 1.141 0.55 2.368 0.42 0.171 1.033 
Currently married 0.967 0.555 1.685 0.705 0.375 1.325 
Formerly married (RC) 1.000  -  - 1.000  -  - 
Ethnicity         
Akan 0.87 0.499 1.516 2.166* 1.055 4.449 
Ga/Dangme 0.61 0.297 1.253 0.706 0.267 1.868 
Ewe 0.757 0.387 1.478 2.057 0.883 4.794 
Mole Dagbani 1.979* 1.137 3.443 2.666* 1.287 5.52 
Other (RC) 1.000  -  - 1.000  -  - 
Religion         
Christian 1.265 0.692 2.314 1.064 0.526 2.151 
Islam 1.525 0.754 3.081 1.277 0.54 3.019 
Other (RC) 1.000  -  - 1.000  -  - 
 
 
 
76 
 
University of Ghana http://ugspace.ug.edu.gh
Table 6.2 contd. 
Medical facility vs No Non-medical facility vs No 
treatment treatment 
Exp (β)  95% CL Exp (β)  95% CL 
Variable   LB UB   LB UB 
Region         
Western 1.618 0.558 4.688 5.601 0.942 33.294 
Central 1.139 0.408 3.18 2.754 0.466 16.273 
Greater Accra 0.488 0.179 1.33 6.48* 1.173 35.79 
Volta 1.298 0.433 3.887 6.364* 1.065 38.027 
Eastern 0.804 0.302 2.136 4.31 0.79 23.514 
Ashanti 0.591 0.23 1.516 6.158* 1.174 32.297 
Brong Ahafo 1.429 0.53 3.852 5.001 0.894 27.968 
Northern 1.155 0.485 2.748 8.808* 1.805 42.972 
Upper East 1.381 0.483 3.949 1.764 0.248 12.555 
Upper West (RC) 1.000  -  - 1.000  -  - 
Type of Place of 
residence         
Urban 0.633* 0.414 0.968 1.28 0.776 2.112 
Rural (RC) 1.000  -  - 1.000  -  - 
Working status         
Not working 0.573* 0.38 0.863 0.854 0.524 1.392 
Working (RC) 1.000  -  - 1.000  -  - 
Wealth index         
Poorest 0.199* 0.094 0.422 1.373 0.565 3.333 
Poorer 0.376* 0.192 0.739 1.887 0.836 4.259 
Middle 0.487* 0.271 0.875 1.165 0.569 2.388 
Richer 0.674 0.403 1.127 1.388 0.739 2.606 
Richest (RC) 1.000  -  - 1.000  -  - 
Sex of child         
Male 0.82 0.612 1.098 0.886 0.623 1.26 
Female (RC) 1.000  -  - 1.000  -  - 
Age of child in years         
0 1.146 0.694 1.891 0.845 0.455 1.567 
1 1.726* 1.069 2.785 1.44 0.819 2.535 
2 1.344 0.835 2.165 1.647 0.945 2.869 
3 0.666 0.417 1.062 0.584 0.328 1.04 
4(RC) 1.000  -  - 1.000  -  - 
  Nagelkerke R = 0.172   P-value = 0.000     
 CI= Confidence Interval LB= Lower Bound UB= Upper Bound N= sample size  
*P< 0.05 N=1,189 
77 
 
University of Ghana http://ugspace.ug.edu.gh
In terms of non-medical facility, mother’s education was not significantly associated with health-
seeking behaviour. On the other hand, mothers from Mole Dagbani and Akan ethnic groups are 
more likely to seek non-medical treatment than seek no treatment for their children with 
fever/cough compared to the ‘other’ ethnic group. Mothers residing in Greater Accra, Volta, 
Ashanti and Northern regions were more likely to seek non-medical treatment than seek no 
treatment for their under-five children who experienced fever/cough symptoms. 
There was, however, no statistically significant relationship found between age of mother, marital 
status, religion and sex of child with health-seeking behaviour of mothers for children with 
fever/cough episodes. 
6.4 Discussion of Results 
This section presents the discussion of key findings while comparing them with similar studies on 
health-seeking behaviour for under-five children.  
6.4.1 Determinants of Health-seeking Behaviour for Diarrhoea 
The results from the multivariate analysis (Table 6.1) indicate that maternal education has no 
influence on health-seeking behaviour of mothers when it comes to their children with diarrhoea. 
This is similar to findings of de Silva et al. (2001) in Sir Lanka, Pillai et al., (2003) in India and 
Ustrup et al., 2014 in Malawi. This finding disproves the hypothesis that ‘health seeking behaviour 
of mothers for children who experience diarrhoea symptoms differs by educational level’. Findings 
from the study also reveal that mothers who had no education are more likely to seek medical 
treatment for their children compared to mothers with secondary/higher education. This is however 
inconsistent with findings which suggest that mothers with lower education were more likely to 
78 
 
University of Ghana http://ugspace.ug.edu.gh
use less of medical health facilities during ill health (Sreeramareddy et al., 2006). Mother’s age 
was found to have significant effects on health-seeking behaviour for children with diarrhoea.  
Mothers between the ages of 25-34 years were found to be more likely to seek treatment from 
medical facility than no treatment compared to older mothers (35-49 years).  Older mothers unlike 
younger mothers seek alternative treatment primarily due to past experiences and customs in 
childbearing that rely more on traditional medicine (Bedford and Sharkey, 2014). 
From Table 6.1, mother’s religion is seen as a significant predictor for health-seeking behaviour. 
Similar findings were found by Fosu (1994) and Diaz et al. (2013). Findings from Table 6.1 show 
that Christian and Muslim mothers were more likely to seek medical treatment compared to 
mothers belonging to the other religion. Religion having significant effects on health-seeking 
behaviour is plausible because Christian and Muslim religion encourage and organise health talks 
and programs for their members, influencing a change in their behaviour towards the use of health 
care facilities during morbidities. However, health practices by mothers in Other religion are based 
on the belief in herbs and spiritual therapies for illness treatment. 
Household wealth index has a positive relationship with health-seeking behaviour of mothers. The 
results showed that comparing to mothers in richest household, mothers in poorest, poorer, middle 
and rich households were more likely to seek treatment from non-medical facility. The study 
findings confirm that mothers from poor homes seek other forms of treatment from drug peddlers 
and traditional practitioners during ill health. These treatment sources are preferred due to its 
relatively low cost and proximity (Ustrup et al., 2014). 
 
 
79 
 
University of Ghana http://ugspace.ug.edu.gh
The age of child had significant association with the health-seeking behaviour of the mother. 
Children of age one year were more likely to be taken to medical facility compared to children 
aged four years. This is expected as mothers are quick to seek care for younger children as they 
are more likely to have severe consequence when treatment is delayed (De Silva et al., 2001). 
6.4.2 Determinants of Health-seeking Behaviour for Fever/cough 
For fever/cough symptoms, mother’s education had significant effect on health-seeking behaviour. 
From Model 3, mothers with no education were more likely to seek non-medical treatment for sick 
children with fever/cough compared to mothers who had attained secondary or higher education. 
Pillai et al. (2003) found a similar pattern among mothers in Kerala, India. In their study, mothers 
with more education are less likely to seek alternative medicine for their sick children compared 
to mothers with less education. This is possible as less educated mothers are incline to belief that 
traditional treatments are more potent than hospital-based medicine (Bedford and Sharkey, 2014). 
The hypothesis that ‘health seeking behaviour of mothers for children who experience fever/cough 
symptoms differs by educational level.’ is confirmed in this study. 
In terms of socio-cultural factors, ethnicity of the mother had some significant relationship with 
mothers’ health-seeking behaviour (Model 4). Taffa and Chepngeno (2005) also found that 
ethnicity was positively associated with health-seeking behaviour. Mothers from the Mole 
Dagbani ethnic group are more likely to seek treatment from medical facility compared to mothers 
from the ‘other’ ethnic group. Additionally, Mole Dagbani and Akan mothers are more likely to 
seek non-medical treatment compared to the ‘other’ ethnic group. This means that mothers from 
the Mole Dagbani Akan ethnic groups have a good chance to seek treatment than no treatment for 
fever/cough symptoms.  
80 
 
University of Ghana http://ugspace.ug.edu.gh
Region, type of place of residence, wealth index and working status were the significant enabling 
factors with health-seeking behaviour regarding fever/cough symptoms among children. Mothers 
residing in urban areas are also less likely to seek medical treatment for fever/cough compared to 
rural mothers. This is quite interesting as urban areas have relatively more health care facilities 
compared to rural areas but healthcare services may not be available or affordable for all urban 
dwellers. In the same light, mothers residing in the Northern, Ashanti and Greater Accra regions 
were found to be more likely to seek non-medical treatment compared to mothers in Upper West 
Region. Results require further investigation. 
Generally, literature shows that wealthier households are more likely to use health services 
compared to poorer households due to differences in the financial capability between rich and 
poorer households (Adulraheem and Parakoyi, 2009 and Fosu, 1994). The results show that poorer 
mothers will seek no treatment for their children with fever/cough symptoms compared to 
wealthier mothers. Mothers who do not seek treatment may not be well informed about the danger 
signs of illness and the availability of treatment (de Silva et al., 2001; Ustrup et al., 2014). They 
may also be hindered by financial cost for health care services. Along the same lines, mothers 
without work are less likely to seek medical treatment compared to working mothers.   
Only age of child was significant in predicting health-seeking behaviour for fever/cough among 
the child characteristics used in the study. Similar for diarrhoea symptoms, mothers of children 
age one year are more likely to seek treatment at medical facilities compared to children aged four 
years.  Other studies confirm that mothers seek better care for children below two years with 
fever/cough symptoms (Ustrup et al., 2014; Noordam et al., 2015; Adinan, et al., 2017). 
81 
 
University of Ghana http://ugspace.ug.edu.gh
6.4.3 Differences between Health-seeking Behaviour for diarrhoea and fever/cough 
symptoms 
Results from this study show some differences in health-seeking behaviour of mothers for 
childhood illnesses. More children (20.1%) experience fever/cough symptoms compared to 
diarrhoea (11.2%), again mothers with children having fever/cough symptoms were more likely 
to seek care than children with diarrhoea symptoms (74.2% for fever/cough and 69.9% for 
diarrhoea). It is possible that mothers recognise how vulnerable under-fives are to fever/cough 
symptoms and they are more likely to seek care during illness episode (Ustrup et al., 2014). 
Evidence in literature supports this finding (Akinyemi et al., 2019).  Finally, the study also 
reveals different determinants for diarrhoea and fever/cough symptoms with the exception of age 
of child. 
 
 
 
 
 
 
 
 
 
82 
 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER SEVEN 
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS 
7.1 Introduction  
This chapter presents the summary of the key findings from the analysis of mother’s education and 
health-seeking behaviour during childhood illnesses. The chapter also presents relevant 
recommendations on health-seeking behaviour of mothers based on findings. 
7.2 Summary of main findings 
Analysis of the background characteristics showed that, children who exhibited diarrhoea 
symptoms had 30.1%, 47.1% and 22.8% of mothers seeking no treatment, medical facility 
treatment and non-medical facility treatment, respectively. Nearly half (47.5) of mothers had 
secondary/higher education, a higher proportion of them were in the age group 25-34 years, 82.5% 
were currently married and almost half were from the Akan ethnic group. For children who 
exhibited fever/cough symptoms, 25.8%, 52.8% and 21.4% of the mothers respectively, sought no 
treatment, sought medical facility and non-medical facility treatment. More than half of the 
mothers had attained secondary/higher education, and Christian mothers had the highest 
representation with 75.5%. 
Objective one of the study sought to examine the variation in the health-seeking behaviour of 
women relative to illness symptoms for their under-five children. Health-seeking behaviour was 
assessed using three categories: no treatment, medical facility and non-medical facility. The results 
showed very little variation in mothers’ health-seeking behaviour for children with diarrhoea and 
fever/cough symptoms. Majority of mothers sought treatment when children exhibited childhood 
83 
 
University of Ghana http://ugspace.ug.edu.gh
illness symptoms (69.9% for diarrhoea and 74.2% for fever/cough) and a higher proportion of 
mothers also preferred medical facility (47.1% and 52.8% respectively) compared to non-medical 
facility (22.8% and 21.4% respectively). 
Objective two aimed to assess the relationship between the mother’s education and health-seeking 
behaviour relative to illness symptoms for their under-five children. With respect to diarrhoea 
symptoms, the results from the bivariate and multivariate analyses showed no statistically 
significant relationship between mother’s education and health-seeking behaviour. However, for 
fever/cough symptoms, mothers with no education were significantly more likely to seek non-
medical facility treatment for sick children compared to mothers with secondary/higher education. 
Objective three sought to identify and examine other factors that predict the health-seeking 
behaviour of mothers with respect to their under-five children. The results suggested that 
predictors of health-seeking behaviour for diarrhoea include mothers’ age, religion, wealth index 
and age of child. Mothers of age group 25-34 were more likely to seek medical treatment for their 
children with diarrhoea symptoms juxtapose mothers aged 35-49 years. Comparing to mothers of 
Other religion, Christian and Muslim mothers were more likely to seek medical treatment for their 
wards. Children of mothers from poorest, poorer, middle and rich households were also more likely 
to use non-medical facility for diarrhoea symptoms compared to mothers from richest households. 
Lastly, children aged one year were more likely to be taken to a medical facility during ill health 
compared to children aged four years. 
With reference to fever/cough symptoms, significant predictors were ethnicity, type of place of 
residence, region, wealth index, working status and age of child. The results showed that the Mole 
Dagbani mothers were more likely to seek some treatment than no treatment compared to mothers 
from the ‘other’ ethnic group. Mothers living in urban areas were also less likely to seek medical 
84 
 
University of Ghana http://ugspace.ug.edu.gh
care compared to mothers living in rural areas. Mothers from poorest, poorer and middle are less 
likely to seek medical treatment compared to mothers from richest household. Similarly, mothers 
without work were less likely to seek medical treatment compared to working mothers. 
7.3 Conclusions 
The purpose of this research was to determine relevant factors influencing health-seeking 
behaviors of mothers for childhood illnesses using the Andersen and Newman’s (1973) Health 
Utilization Model as a theoretical framework. The findings of the study showed the different 
components that determine health-seeking behaviour of mothers. Predisposing factors such as 
mothers’ education, age, ethnicity, religion and age of child were identified to influence health-
seeking behaviour. Working status, place of residence, region of abode and wealth index were 
enabling factors affiliated with care seeking for childhood illness. The type of disease also 
influences the decision to seek treatment. The findings illustrate the relationship between 
predisposing, enabling and need factors in health-seeking behaviour of mothers. These factors help 
to explain the complex variation in predicting health-seeking behaviour of mothers for childhood 
illness. 
There is very little difference between health-seeking behaviour of mothers relative to illness 
symptoms exhibited by under-five children Ghana.  Mother’s education was found not to be 
statistically significant in predicting health-seeking behaviour for diarrhoea symptoms but was 
found significant for fever/cough symptoms. Children whose mothers had no education were 
significantly more likely to seek non-medical treatment compared to mothers with 
secondary/higher education. Mothers age, ethnicity, religion, place of residence, region, wealth 
85 
 
University of Ghana http://ugspace.ug.edu.gh
index, working status and age of child were also found as significant predictors of health-seeking 
behaviour for under five children in Ghana.  
7.4 Recommendations 
Empirical findings from this study show that maternal education was a predictor for health-
seeking for fever/cough symptoms among children. Less educated mothers are more likely to use 
other forms of treatments such as pharmacy, drug peddlers and traditional practitioners instead of 
visits to medical facilities for children exhibiting fever/cough symptoms. The study suggests that 
improving the educational status of women will improve better health-seeking practices as 
mothers will seek appropriate treatment from the right sources, including from health facilities 
where proper diagnosis could be done to increase the chances of child survival during ill health. 
Another important finding was that mothers in urban areas, not working and poor were less 
likely to use medical facilities during children illness. This implies that improving the 
socioeconomic status of women through job creation will ensure they have funds to access health 
facilities when children are sick to reduce mortality rates. However, further research should be 
done to explore why urban mothers are not using medical facilities where there are relatively 
more health facilities. 
Many mothers belonging to the other religion failed to seek medical treatment for their sick 
children. This study proposes that health planners should collaborate with other religious group 
leaders for the purpose of educating their members with health knowledge on health-seeking 
practices for children to ensure better health outcomes. Results from the study also showed that 
ethnicity significantly predict health-seeking behaviour. Based on this, there is the need for 
86 
 
University of Ghana http://ugspace.ug.edu.gh
further studies to explore the health-seeking behaviour of women among the various ethnic 
groups to identify potential cultural practices that influencing health-seeking. 
Furthermore, although quite urbanized and with more health care facilities, mothers from Greater 
Accra, Volta, Ashanti and Northern regions seek treatment from non-medical facilities. 
Qualitative enquiry is need to examine why mother preferred alternative sources for childhood 
illnesses. Finally, since mothers are more likely to seek treatment for children aged one, health 
care professions need to emphasize the need for appropriate treatment for all ages as children 
under five are vulnerable to severe consequences to diseases. 
 
 
  
87 
 
University of Ghana http://ugspace.ug.edu.gh
REFERENCES 
Abdul-Aziz, A. R., Harris, E., & Munyakazi, L. (2012). Risk Factors in Malaria Mortality among 
Children in Northern Ghana: A Case Study at The Tamale Teaching Hospital. International 
Journal of Business and Social Research (IJBSR), 2(5), 34–45. 
 
Abdulraheem, I. S., & Parakoyi, D. B. (2009). Factors affecting mothers’ healthcare‐seeking 
behaviour for childhood illnesses in a rural Nigerian setting. Early Child Development and 
Care, 179(5), 671–683. https://doi.org/10.1080/03004430701500885 
Abosse, Z., Woldie, M., & Ololo, S. (2010). Factors influencing antenatal care service utilization 
in hadiya zone. Ethiopian Journal of Health Sciences, 20(2). 
Abubakar, A., Van Baar, A., Fischer, R., Bomu, G., Gona, J. K., & Newton, C. R. (2013). Socio-
cultural determinants of health-seeking behaviour on the Kenyan coast: a qualitative 
study. PloS one, 8(11), e71998. 
Acharya, D. R., Bell, J. S., Simkhada, P., Van Teijlingen, E. R., & Regmi, P. R. (2010). 
Women's autonomy in household decision-making: a demographic study in 
Nepal. Reproductive health, 7(1), 1-12. 
Ada, G. (2007). The importance of vaccination. Frontiers in Bioscience : A Journal and Virtual 
Library, 12, 1278–1290. 
Adinan, J., Damian, D. J., Mosha, N. R., Mboya, I. B., Mamseri, R., & Msuya, S. E. (2017). 
Individual and contextual factors associated with appropriate healthcare seeking behavior 
among febrile children in Tanzania. PLoS One, 12(4), e0175446. 
88 
 
University of Ghana http://ugspace.ug.edu.gh
Akeju, D. O., Oladapo, O. T., Vidler, M., Akinmade, A. A., Sawchuck, D., Qureshi, R., ... & von 
Dadelszen, P. (2016). Determinants of health care seeking behaviour during pregnancy in 
Ogun State, Nigeria. Reproductive health, 13(1), 67-74. 
Akinyemi, J. O., Banda, P., De Wet, N., Akosile, A. E., & Odimegwu, C. O. (2019). Household 
relationships and healthcare seeking behaviour for common childhood illnesses in sub-
Saharan Africa: a cross-national mixed effects analysis. BMC health services 
research, 19(1), 1-11. 
Andersen, R. & Newman, J. (1973). Societal and individual determinants of medical care 
utilization. The Milbank Memorial Fund Quarterly. Health and Society, 95-124 
Anwar, M., Green, J., & Norris, P. (2012). Health-seeking behaviour in Pakistan: A narrative 
review of the existing literature. Public health, 126(6), 507-517. 
Arifeen, S. E., Hoque, D. E., Akter, T., Rahman, M., Hoque, M. E., Begum, K., ... & Hossain, M. 
A. (2009). Effect of the Integrated Management of Childhood Illness strategy on childhood 
mortality and nutrition in a rural area in Bangladesh: a cluster randomised trial. The Lancet, 
374(9687), 393-403. 
Baker, D. P., Leon, J., & Collins, M. J. (2011). Facts, attitudes, and health reasoning about HIV 
and AIDS: explaining the education effect on condom use among adults in sub-Saharan 
Africa. AIDS and Behavior. 15(17), 1319–1327. 
Bedford, K. J. A., & Sharkey, A. B. (2014). Local barriers and solutions to improve care-seeking 
for childhood pneumonia, diarrhoea and malaria in Kenya, Nigeria and Niger: a qualitative 
study. PloS one, 9(6), e100038. 
89 
 
University of Ghana http://ugspace.ug.edu.gh
Buor, D. (2003). Mothers’ education and childhood mortality in Ghana. Health Policy, 64(3), 
297–309. https://doi.org/10.1016/S0168-8510(02)00178-1 
Buor, D. (2004). Determinants of utilisation of health services by women in rural and urban areas 
in Ghana. GeoJournal, 61(1), 89-102. 
Caldwell, J. C. (1979). Education as a Factor in Mortality Decline An Examination of Nigerian 
Data. Population Studies, 33(3), 395–413. http://www.jstor.com/stable/2173888 
Caldwell, J., & McDonald, P. (1982). Influence of maternal education on infant and child 
mortality: levels and causes. Health policy and education, 2(3-4), 251-267. 
Charles, J. O., Udonwa, N. E., Ikoh, M. U., & Ikpeme, B. I. (2008). The role of mothers in 
household health-seeking behavior and decision-making in childhood febrile illness in 
Okurikang/Ikot Effiong Otop Community, Cross River State, Nigeria. Health care for 
women international, 29(8-9), 906-925. 
Ching, P. (1995). User fees, demand for children's health care and access across income groups: 
the Philippine case. Social Science & Medicine, 41(1), 37-46. 
Das Gupta, M. (1990). Death clustering, Mothers’ education and the determinants of child 
mortality in rural Punjab, India. Population Studies, 44(3), 489–505. 
https://doi.org/10.1080/0032472031000144866 
de Silva, M. W. A., Wijekoon, A., Hornik, R., & Martines, J. (2001). Care seeking in Sri Lanka : 
one possible explanation for low childhood mortality. Social Science & Medicine, 53(10), 
1363-1372. 
 
90 
 
University of Ghana http://ugspace.ug.edu.gh
Diaz, T., George, A. S., Rao, S. R., Bangura, P. S., Baimba, J. B., McMahon, S. A., & Kabano, 
A. (2013). Healthcare seeking for diarrhoea, malaria and pneumonia among children in four poor 
rural districts in Sierra Leone in the context of free health care: Results of a cross-sectional 
survey. BMC public health, 13(1), 157. 
 
Dimbuene, Z. T., Amo-Adjei, J., Amugsi, D., Mumah, J., Izugbara, C. O., & Beguy, D. (2018). 
Women’s education and utilization of maternal health services in Africa: a multi-country and 
socioeconomic status analysis. Journal of biosocial science, 50(6), 725-748. 
D'Souza, R. M. (1999). Care-seeking behavior. Clinical infectious diseases, 28(2), 234-234. 
Retrieved from https://www.jstor.org/stable/4460693. 
D’Souza, R. M. (2003). Role of health-seeking behaviour in child mortality in the slums of 
Karachi, Pakistan. Journal of Biosocial Science, 35(1), 131–144. 
https://doi.org/10.1017/S0021932003001317 
Dudgeon, M. R., & Inhorn, M. C. (2004). Men's influences on women's reproductive health: 
medical anthropological perspectives. Social science & medicine, 59(7), 1379-1395. 
Ellis, A. A., Doumbia, S., Traoré, S., Dalglish, S. L., & Winch, P. J. (2013). Household roles and 
care-seeking behaviours in response to severe childhood illness in Mali. Journal of 
biosocial science, 45(6), 743-759. 
Elo, I. T. (1992). Utilization of maternal health-care services in Peru: the role of women's 
education. Health transition review, 49-69. 
Fischetti, M. (2011). Female education reduces infant and childhood deaths. Scientific American. 
Retrieved August 20, 2020 from  https://www.scientificamerican.com/article/graphic-
91 
 
University of Ghana http://ugspace.ug.edu.gh
science-female-education-reduces-infant-childhood-deaths/ 
Frempong-Ainguah, F., Bailey, C. E., & Hill, A. G. (2018). Women’s health status in urban 
Ghana: dimensions and differentials using short form 36. Health and quality of life 
outcomes, 16(1), 1-13. 
Funk, T., Källander, K., Abebe, A., Alfvén, T., & Alvesson, H. M. (2020). ‘I also take part in 
caring for the sick child’: a qualitative study on fathers’ roles and responsibilities in seeking 
care for children in Southwest Ethiopia. BMJ open, 10(8), e038932. 
 
Gakidou, E., Cowling, K., Lozano, R., & Murray, C. J. (2010). Increased educational attainment 
and its effect on child mortality in 175 countries between 1970 and 2009: A systematic 
analysis. The Lancet, 376(9745), 959-974. https://doi.org/10.1016/s0140-6736(10)61257-
3 
Ghana Statistical Service, Ghana Health Service & ICF International. (2009). Ghana 
Demographic and Health Survey 2008 
 
Ghana Statistical Service. (2013). 2010 Population & Housing Census: National analytical 
report. Ghana Statistical Service. 
Ghana Statistical Service, Ghana Health Service, and ICF International. (2015). Ghana 
Demographic Health Survey 2014.Rockville Marylamd, USA: GSS, GHS and ICF 
International 
Glewwe, P. (1999). Why does mother's schooling raise child health in developing countries? 
Evidence from Morocco. Journal of human resources, 124-159. 
 
92 
 
University of Ghana http://ugspace.ug.edu.gh
Goldman, N., & Heuveline, P. (2000). Health-seeking behaviour for child illness in Guatemala. 
Tropical Medicine & International Health, 5(2), 145–155. 
Godana, W., & Mengiste, B. (2013). Environmental factors associated with acute diarrhea 
among children under five years of age in derashe district , Southern Ethiopia. Sci J Public 
Health, 1(3), 119–124. https://doi.org/10.11648/j.sjph.20130103.12 
Greenaway, E. S., Leon, J., & Baker, D. P. (2012). Understanding the association b e t w e e n 
maternal education and use of health services in Ghana : exploring the role of health 
knowledge. Journal of biosocial science, 44(6), 733. 
https://doi.org/10.1017/S0021932012000041 
Thomas Hakansson, N. (1985). Why do Gusii women get married? A study of cultural 
constraints and women's strategies in a rural community in Kenya. Folk, 27, 89-114. 
Hill, M. A., & King, E. (1995). Women's education and economic well-being. Feminist 
Economics, 1(2), 21-46. 
Hill, Z., Kendall, C., Arthur, P., Kirkwood, B., & Adjei, E. (2003). Recognizing childhood 
illnesses and their traditional explanations : exploring options for care-seeking interventions 
in the context of the IMCI strategy in rural Ghana. Tropical Medicine & International 
Health, 8(7), 668-676. 
Hobcraft, J. (1993). Women's education, child welfare and child survival: a review of the 
evidence. Health transition review, 159-175. 
Igun, U. A. (1979). Stages in health-seeking: a descriptive model. Social Science & Medicine. 
Part A: Medical Psychology & Medical Sociology, 13, 445-456. 
93 
 
University of Ghana http://ugspace.ug.edu.gh
 
Kayode, G. A., Grobbee, D. E., Koduah, A., Amoakoh-Coleman, M., Agyepong, I. A., Ansah, 
E., van Dijk, H., & Klipstein-Grobusch, K. (2016). Temporal trends in childhood mortality 
in Ghana: impacts and challenges of health policies and programs. Global Health Action, 
9(1), 31907. https://doi.org/10.3402/gha.v9.31907 
Kifle, D., Azale, T., Gelaw, Y. A., & Melsew, Y. A. (2017). Maternal health care service seeking 
behaviors and associated factors among women in rural Haramaya District, Eastern 
Ethiopia: a triangulated community-based cross-sectional study. Reproductive health, 14(1), 
1-11. 
Kolola, T., Gezahegn, T., & Addisie, M. (2016). Health care seeking behavior for common 
childhood illnesses in jeldu district, oromia regional state, Ethiopia. PLoS ONE, 11(10), 1–
11. https://doi.org/10.1371/journal.pone.0164534 
Kutty, V. R. (1989). Women’s education and its influence on attitudes to aspects of child-care in 
a village community in Kerala. Social Science and Medicine, 29(11), 1299–1303. 
https://doi.org/10.1016/0277-9536(89)90070-1 
Leowski, J. (1986). Mortality from acute respiratory infections in children under 5 years of age: 
global estimates. PAHO. Epidemiological Bulletin; 7 (5/6), 1986. 
Levine, M. P., Smolak, L., & Hayden, H. (1994). The relation of sociocultural factors to eating 
attitudes and behaviors among middle school girls. The Journal of Early 
Adolescence, 14(4), 471-490. 
LeVine, R. A., LeVine, S. E., Rowe, M. L., & Schnell-Anzola, B. (2004). Maternal literacy and 
94 
 
University of Ghana http://ugspace.ug.edu.gh
health behavior: A Nepalese case study. Social Science and Medicine, 58(4), 863–877. 
https://doi.org/10.1016/S0277-9536(03)00261-2 
LeVine, R. A., & Rowe, M. L. (2009). Maternal literacy and child health in less-developed 
countries: Evidence, processes, and limitations. Journal of Developmental & Behavioral 
Pediatrics, 30(4), 340-349. https://doi.org/10.1097/dbp.0b013e3181b0eeff 
Lleras-Muney, A. (2005). The relationship between education and adult mortality in the United 
States. Review of Economic Studies, 72(1), 189–221. https://doi.org/10.1111/0034-
6527.00329 
Livingstone, A. M. (1994). Women’s participation in primary health care program in Ghana. 
CIDA. 
 
MacKian, S. (2003). A review of health seeking behaviour: Problems and prospects. Health 
Systems Development Programme. Health Policy and Planning, 19(3), 137-146. 
Mandela, N. (2003). Lighting your way to a better future. Speech delivered at the launch of 
Mindset Network. Planetarium, University of Witwatersrand, Johannesburg, South Africa, 
16. 
Martin, T. C. (1995). Women’s education and fertility: results from 26 Demographic and Health 
Surveys. Studies in family planning, 26(4), 187–202. 
Mashal, T., Nakamura, K., Kizuki, M., Seino, K., & Takano, T. (2007). Impact of conflict on 
infant immunisation coverage in Afghanistan: a countrywide study 2000-2003. 
International Journal of Health Geographics, 6, 23. https://doi.org/10.1186/1476-072X-623 
95 
 
University of Ghana http://ugspace.ug.edu.gh
Mukunya, D., Kizito, S., Orach, T., Ndagire, R., Tumwakire, E., Rukundo, G. Z., Mupere, E., & 
Kiguli, S. (2014). Knowledge of integrated management of childhood illnesses community 
and family practices ( C-IMCI ) and association with child undernutrition in Northern 
Uganda : a cross-sectional study. BMC public health, 14(1), 1–7. 
Nair, H., Simõ, E. A. F., Rudan, I., Gessner, B. D., Azziz-baumgartner, E., Zhang, J. S. F., 
Feikin, D. R., & Mackenzie, G. A. (2013). Global and regional burden of hospital 
admissions for severe acute lower respiratory infections in young children in 2010 : a 
systematic analysis. The Lancet, 381(9875), 1380-1390.  
Niraula, B. B. (1994). Use of health services in hill villages in central Nepal. Health Transition 
Review, 151-166. 
Noordam, A. C., Carvajal-Velez, L., Sharkey, A. B., Young, M., & Cals, J. W. (2015). Care 
seeking behaviour for children with suspected pneumonia in countries in sub-Saharan 
Africa with high pneumonia mortality. PloS one, 10(2), e0117919. 
Obermeyer, C. M. (1993). Culture, maternal health care, and women's status: a comparison of 
Morocco and Tunisia. Studies in Family Planning, 24(6), 354-365. 
https://doi.org/10.2307/2939245 
Ojanuja, D.N. and Gilbert, C. 1992: Women's access to health care in developing countries. 
Social Science & Medicine, 35(4): 613-617. 
Okoko, B. J., & Yamuah, L. K. (2006). Household decision-making process and childhood 
cerebral malaria in The Gambia. Archives of medical research, 37(3), 399-402. 
Olenja, J. (2003). Editorial: Health seeking behaviour in context. East African Medical Journal, 
80(2), 61–62. https://doi.org/10.4314/eamj.v80i2.8689 
96 
 
University of Ghana http://ugspace.ug.edu.gh
Osamor, P. E., & Grady, C. (2016). Women’s autonomy in health care decision-making in 
developing countries: a synthesis of the literature. International journal of women's 
health, 8, 191. 
Osumanu, I. K. (2014). Household environmental and behavioural determinants of childhood 
diarrhoea morbidity in the Tamale Metropolitan Area ( TMA ), Ghana. Geografisk 
Tidsskrift-Danish journal of geography, 107(1), 59-68 
https://doi.org/10.1080/00167223.2007.10801375 
Pillai, R. K., Williams, S. V., Glick, H. A., Polsky, D., Berlin, J. A., & Lowe, R. A. (2003). 
Factors affecting decisions to seek treatment for sick children in Kerala, India. Social 
Science and Medicine, 57(5), 783–790. https://doi.org/10.1016/S0277-9536(02)00448-3 
Plantin, L., Olykoya, A., & Ny, P. (2011). Positive health outcomes of fathers’ involvment in 
pregnancy and childbirth paternal support: a scope study literature review. Fathering: A 
Journal of Theory, Research, and Practice about Men as Fathers, 9(1), 87-102. 
Pokhrel, S., & Sauerborn, R. (2004). Household decision-making on child health care in 
developing countries: the case of Nepal. Health Policy and Planning, 19(4), 218–233. 
https://doi.org/10.1093/heapol/czh027 
Pokhrel, S., Snow, R., Dong, H., Hidayat, B., Flessa, S., & Sauerborn, R. (2005). Gender role 
and child health care utilization in Nepal. Health policy, 74(1), 100–109. 
https://doi.org/10.1016/j.healthpol.2004.12.013 
Ross, C. E. ., & Mirowsky, J. (1999). Refining the Assocation between Education and Health : 
The Effects of Quantity , Credential , and Selectivity. Demography, 36(4), 445–460.   
97 
 
University of Ghana http://ugspace.ug.edu.gh
Rowe, M. L., Thapa, B. K., Levine, R., Levine, S., & Tuladhar, S. K. (2005). How does 
schooling influence maternal health practices? Evidence from Nepal. Comparative 
Education Review, 49(4), 512-533 
Rueda-Clausen, C. F., Campbell, J., & Baker, P. N. (2011). Current challenges in pregnancy-
related mortality. Obstetrics, Gynaecology & Reproductive Medicine, 21(8), 229-235. 
Schellenberg, J. R. A., Adam, T., Mshinda, H., Masanja, H., Kabadi, G., Mukasa, O., ... & 
Mgalula, L. (2004). Effectiveness and cost of facility-based Integrated Management of 
Childhood Illness (IMCI) in Tanzania. The Lancet,  364(9445), 1583-1594. 
https://doi.org/10.1016/S01406736(04)17311-X 
 
Shaikh, B. T., & Hatcher, J. (2005). Health seeking behaviour and health service utilization in 
Pakistan: challenging the policy makers. Journal of Public Health, 27(1), 49–54. 
https://doi.org/10.1093/pubmed/fdh207 
Shapiro, D., & Tenikue, M. (2017). Women’s education, infant and child mortality, and fertility 
decline in urban and rural sub-Saharan Africa. Demographic Research, 37, 669-708. 
Shewade, H. D., Aggarwal, A. K., & Bharti, B. (2013). Integrated Management of Neonatal and 
Childhood Illness (IMNCI): skill assessment of health and Integrated Child Development 
Scheme (ICDS) workers to classify sick under-five children. Indian Journal of Pediatrics, 
80(6), 448–454. https://doi.org/10.1007/s12098-012-0835-4 
Som, C. K. (2014). Maternal Education and Child Survival in Ghana. University of Ghana, 
Legon. 
98 
 
University of Ghana http://ugspace.ug.edu.gh
Sreeramareddy, C. T., Shankar, R. P., Sreekumaran, B. V., Subba, S. H., Joshi, H. S., & 
Ramachandran, U. (2006). Care seeking behaviour for childhood illness- a questionnaire 
survey in western Nepal. BMC International Health and Human Rights, 6(1), 1–10. 
https://doi.org/10.1186/1472-698X-6-7 
Taffa, N., & Chepngeno, G. (2005). Determinants of health care seeking for childhood illnesses 
in Nairobi slums. Tropical Medicine and International Health, 10(3), 240–245. 
https://doi.org/10.1111/j.1365-3156.2004.01381.x 
Tazinya, A. A., Halle-Ekane, G. E., Mbuagbaw, L. T., Abanda, M., Atashili, J., & Obama, M. T. 
(2018). Risk factors for acute respiratory infections in children under five years attending 
the Bamenda Regional Hospital in Cameroon. BMC Pulmonary Medicine, 18(1), 7. 
Thind, A., & Andersen, R. (2003). Respiratory illness in the Dominican Republic : what are the 
predictors for health services utilization of young children ?. Social science & medicine, 
56(6), 1173–1182. 
Thind, A., & Cruz, A. M. (2003). Determinants of children’s health services utilization in the 
Philippines. Journal of Tropical Pediatrics, 49(5), 269–273. 
https://doi.org/10.1093/tropej/49.5.269 
Thomas, D. (1999). Fertility, education and resources in South Africa. Critical perspectives on 
schooling and fertility in the developing world, 138-180. 
Ustrup, M., Ngwira, B., Stockman, L. J., Deming, M., Nyasulu, P., Bowie, C., ... & Fischer, T. 
K. (2014). Potential barriers to healthcare in Malawi for under-five children with cough 
99 
 
University of Ghana http://ugspace.ug.edu.gh
and fever: a national household survey. Journal of health, population, and nutrition, 
32(1), 68. 
Ujunwa, F. A., & Ezeonu, C. T. (2014). Risk factors for acute respiratory tract infections in 
under-five children in enugu Southeast Nigeria. Annals of medical and health sciences 
research, 4(1), 95-99. 
UNICEF. (2014). State of the World’s Children 2014 in Numbers: Every Child Counts. 
UNICEF. (2019). Under-five mortality. Retrieved August 24, 2020, from 
https://data.unicef.org/topic/child-survival/under-five-mortality/ 
UNICEF. (n.d.). Immunization. Retrieved August 26, 2020, from 
https://www.unicef.org/ghana/immunization 
United Nations. (2016). The Sustainable Development Goals Report 2016. eSocialSciences 
van Lunzen, J., & Altfeld, M. (2014). Sex differences in infectious diseases–common but 
neglected. The Journal of infectious diseases, 209(suppl_3), S79-S80. 
Vikram, K., & Vanneman, R. (2020). Maternal education and the multidimensionality of child 
health outcomes in India. Journal of biosocial science, 52(1), 57–77. 
https://doi.org/10.1017/S0021932019000245 
Wado, Y. D. (2013). Women’s autonomy and reproductive health-care-seeking behavior in 
Ethiopia. Women and Health, 58(7), 729–743. 
https://doi.org/10.1080/03630242.2017.1353573 
100 
 
University of Ghana http://ugspace.ug.edu.gh
 World Health Organization. (2016). Global Strategy for Women’s, Children’s and Adolescents’ 
Health, 2016–2030. 
World Health Organization. (2017). Diarrhoeal Diseases. Retrieved August 26, 2020 from 
https://www.who.int/news-room/fac-sheets/detail/diarrhoeal-disease 
World Health Organization. (2019). World Malaria Report,2019. 
 
 
 
 
101