UNIVERSITY OF CAPE COAST CHILDHOOD VACCINATION IN GHANA FROM 1998 TO 2014 BY EUGENE BUDU Thesis submitted to the Department of Population and Health of the Faculty of Social Sciences, College of Humanities and Legal Studies, University of Cape Coast in partial fulfilment of the requirements for the award of Masters of Philosophy Degree in Population and Health JULY 2020 © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library ii DECLARATION Candidate’s Declaration I hereby declare that this thesis is the result of my own original research and that no part of it has been presented for another degree in this university or elsewhere. Candidate’s Signature……………….. Date……………………. Name: Eugene Budu Supervisors’ Declaration I hereby declare that the preparation and presentation of the thesis were supervised in accordance with the guidelines on supervision of the thesis laid down by the University of Cape Coast. Supervisor’s Signature…………………………. Date………………………. Name: Prof. Eugene K. M. Darteh © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library iii ABSTRACT The adverse effect of vaccine preventable diseases (VPDs) on the health of children cannot be overemphasized. These diseases are major causes of illness, disability and death in childhood, and account for 17 percent of global total under-five mortality per year. Vaccination remains one of public health’s greatest achievements. It is one of the major ways to reduce child morbidity and mortality from common VPDs. The vaccines given to children as part of the Extended Program on Immunisation in Ghana includes polio 1, 2 and 3, measles, Bacille Calmette-Guerin (BCG), Diphtheria-Pertussis-Tetanus (DPT) 1, 2 and 3 as well as yellow fever vaccines. Using the health care service utilisation model, this study assessed childhood vaccination in Ghana from 1998 to 2014. Mothers of children who are aged 0-23 months from four rounds 1998, 2003, 2008 and 2014 of the Ghana Demographic and Health Survey were used for the study. Univariate and multivariate analyses were employed in the study. Over the period, BCG was the most received vaccine while the least received vaccine was measles. Vaccination coverage for all the four vaccines received by children increased from 1998 to 2014. Thus, from 83% to 94% for BCG, 54% to 71% for DPT, 54% to 79% for polio and 43% to 50% for measles. Age of the mother, occupation, education, religion and parity were the factors that influenced the uptake of the vaccination. There is the need for MoH and the Ghana Health Service to strengthen interventions and messages targeting mothers and potential mothers on the importance of vaccination and immunisation for their children. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library iv KEY WORDS Determinants Extended Programme on Immunization (Epi) Immunization Trends Vaccination © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library v ACKNOWLEDGEMENTS I am particularly grateful to a number of people who have contributed immensely to this work from its conceptions. I am highly indebted to my supervisor, Prof. Eugene K. M. Darteh for meticulously reading through the work and for encouraging me. During the work, I received support from some colleagues and friends individually and severally. To all these colleagues and friends, especially Bright, Ivy, Joojo and Ishmael, I say a very big thank you for all your support. To my family, especially to my father, Mr. Solomon S. Budu, I say thank you very much for supporting and encouraging me throughout the work. Thank you for showing so much interest in this work and giving me the reason to complete the work. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library vi DEDICATION To my family © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library vii TABLE OF CONTENTS Page DECLARATION ii ABSTRACT iii KEY WORDS iv ACKNOWLEDGEMENTS v DEDICATION vi TABLE OF CONTENTS vii LIST OF TABLES xi LIST OF FIGURES xii LIST OF ACRONYMS xiii CHAPTER ONE: INTRODUCTION Background to the Study 1 Statement of the Problem 3 Objectives of the Study 7 Research Questions 7 Rationale of the Study 7 Organisation of the Study 8 CHAPTER TWO: LITERATURE REVIEW Introduction 9 Concepts of Vaccination and Immunisation 9 History of Vaccination 10 The Extended Program on Immunisation 14 Implementation of the EPI 16 © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library viii Evaluation of the EPI 18 Results of the Implementation of the EPI 19 The Expanded Programme on Immunisation in Ghana 19 EPI and the Health Sector Reforms 22 The Comprehensive Multi-Year Plan 23 Goals and Objectives 25 Objectives of the cMYP 26 Trends of Childhood Vaccination 27 Maternal Age as a Determinate of Childhood Vaccination 29 Maternal Educational Level as a Determinate of Childhood Vaccination 30 Wealth Status as a Determinant of Childhood Vaccination 32 Region as a Determinant of Childhood Vaccination 33 Place of Residence as a Determinate of Childhood Vaccination 34 Mother’s Occupation as a Determinate of Childhood Vaccination 36 Ethnicity as a Determinate of Childhood Vaccination 38 Parity as a Determinate of Childhood Vaccination 39 Marital Status as a Determinate of Childhood Vaccination 41 Religion as a Determinant of Childhood Vaccination 43 Theoretical Framework 45 Health Care Services Utilisation Model 45 Conceptual Framework 47 © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library ix CHAPTER THREE:METHODS OF DATA COLLECTION AND ANALYSIS Introduction 49 Study Area 49 Research Design 51 Source of Data 52 Target Population 53 Sampling Procedures 53 Acquisition of Data 54 Description and Definition of Dependent Variables 55 Independent Variable 56 Data Analysis 56 Data Limitation 58 CHAPTER FOUR: RESULTS AND DISCUSSION Introduction 59 Results 59 Socio-Demographic Characteristics of Respondents 59 Trends of Under Five Vaccination from 1998-2014 64 Prevalence of Childhood Vaccination in Ghana from 1998-2014 66 Determinants of Measles Vaccination 69 Determinants of Polio Vaccination 75 Determinants of BCG Vaccination 79 © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library x Determinants of DPT Vaccination in Ghana from 1998-2014 83 Discussion 88 Trends of Childhood Vaccination 88 Pattern of Childhood Vaccination 88 Determinates of Childhood Vaccination 91 CHAPTER FIVE: SUMMARY, CONCLUSIONS AND RECOMMENDATIONS Introduction 95 Key Findings of the Study 96 Conclusions 100 Recommendation 101 Areas for Future Studies 101 REFERENCES 103 © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library xi LIST OF TABLES Table Page 1 The EPI schedule for Ghana 21 2 Sampling Procedure 54 3 Socio-Demographic Characteristics of Respondents 60 4 Socio-demographic characteristics of mothers and prevalence of childhood vaccination in Ghana from 1998-2014 68 5 Determinants of vaccination for measles among children under five in Ghana 70 6 Determinants of vaccination for polio among children under five in Ghana 75 7 Determinants of vaccination for BCG among children under five in Ghana 80 8 Determinants of vaccination for DPT among children under five in Ghana 84 © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library xii LIST OF FIGURES Figure Page 1 Health Care Utilisation Model 46 2 Conceptual framework 48 3 Trends of childhood vaccination from 1998 to 2014 65 © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library xiii LIST OF ACRONYMS BCG CDD DHS DPT EPI GAVI GDHS GSS HepB KDHS MoH NGO OPV PHC UNICEF VPD WHA WHO ZDHS Bacillus Calmette Guḗrin Control of Diarrhoea Diseases Demographic and Health Survey Diphtheria-Tetanus-Pertusis Expanded Programme on Immunisation Ghana Demographic and Health Survey Ghana Statistical Service Global Action for Vaccination and Immunisation Hepatitis B Kenya Demographic and Health Survey Ministry of Health Non-Government Organization Oral Polio Vaccine Primary Health Care United Nations Children Fund Vaccine Preventable Disease World Health Assembly World Health Organization Zimbabwe Demographic and Health Survey © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 1 CHAPTER ONE INTRODUCTION Background to the Study The adverse effects of vaccine-preventable diseases on the health of children cannot be overemphasized (Asamoh, Agardh, Petterson & Ostergen, 2014). Vaccine-preventable diseases infectious disease for which an effective preventive vaccine exists. The most common and serious vaccine-preventable disease tracked by the World Health Organization (WHO) are diphtheria, Haemophilus influenza serotype b infection, hepatitis B, measles, meningitis, mumps, pertussis, poliomyelitis, rubella, tetanus, tuberculosis and yellow fever (WHO, 2012). These diseases are major causes of illness, disability and death in childhood. Morbidity and mortality of children under five years of age could be largely reduced through measures that are both affordable and available (United Nations International Children’s Emergency Fund [UNICEF], 2013). According to UNICEF, though two to three million childhood deaths are prevented through immunisations, 17 percent of children still die every year from vaccine-preventable diseases. Immunisation has been useful in delaying over two million deaths each year worldwide (UNICEF, 2013) and it is one of the three most successful public health initiatives (WHO & UNICEF, 2012). Globally, about three-quarters of the world’s child population is reached with the required vaccines but in sub-Saharan Africa, a little above half of the child population get access to basic immunisation (WHO, 2015). Further, in poorer remote areas of developing countries, only 1 in 20 children have access to © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 2 vaccination (WHO, 2015). A significant proportion of these deaths are attributable to vaccine-preventable infectious diseases (Liu et al., 2012). Increased attention to this problem led to the initiation of the Expanded Programme on Immunisation (EPI) in 1974 when less than five percent of children worldwide were immunised (WHO, 2012). In Africa, the EPI was launched in 1978 and by the mid-eighties, all countries had established national immunisation programmes (WHO, 2012). A mid-decade evaluation conducted in 1985 indicated that the regional immunisation coverage in Africa was still around 20% for all antigens and therefore, the year 1986 was adopted the “African Immunisation Year”, with various accelerated efforts implemented till 1990 (Bee & Barakamfitiye, 1994). Ghana, just like other African countries, accepted and launched the EPI in 1978, after a period of feasibility studies (NDPC, 2006 cited in Duah-Owusu (2010)). The objective of Ghana’s EPI was to fully immunise 80% of its children aged 0 – 11 months by 1983 (Ghana Health Services [GHS], 2012). However, after 20 years of the launch of the programme, that is, 1998, the proportion of fully vaccinated children before age one was 51% nationally (Ghana Statistical Service [GSS], 1999). The GDHS show an increase in the vaccination coverage of Ghana over the years. For instance, the 1988 GDHS reported a vaccination coverage of 47 percent, which rose to 55 percent in the 1993 GDHS. In 1998, 2003 and 2008 GDHS, the vaccination coverage of Ghana was 62 percent, 69 percent and 79 percent © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 3 respectively. In the 2014 GDHS, there was a marginal decline in the vaccination coverage to a rate of 77 percent (GSS, GHS & ICF International, 2015). From the 2014 Demographic and Health Survey (GDHS), it was realised that, although 90% of children nationally received BCG and the first dose of DPT and OPV vaccine before age one, only two-thirds of these children had received the third dose of DPT and OPV by this age (GSS, GHS & ICF International, 2015). This shows a nation-wide high dropout rate. Despite the increase, these figures are short of the target of 80% coverage which the programme aimed at. The vaccines given to children as part of the EPI in Ghana include polio 1, 2 and 3; measles; BCG; DPT1, 2 and 3 as well as yellow fever vaccines. Studies on child immunization have focused extensively on maternal characteristics. Particularly, educational attainment and literacy (Ataguba et al., 2016; Barata et al., 2012), employment status, age and age at birth, media exposure, marital status, religion and place of residence have been found to have strong relationships with the immunization status of a child. The presence of such strong maternal effects on child immunization coverage is a reflection of the traditional childcare responsibility of mothers in most developing countries, which dovetails to the policy recommendation that empowering women in the household decision- making process, such as child health, may be crucial to achieving universal immunization coverage among underserved populations (Chan, 2014). Statement of the Problem Over the years, Vaccine-Preventable Diseases (VPDs) have caused more deaths among children under five (WHO & UNICEF 2012). One major way to © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 4 reduce child morbidity and mortality from common VPDs is immunisation (Brenzel, 2016). Vaccination remains one of the public health greatest achievements (Asamoh, Agardh, Petterson & Ostergen, 2017). Vaccines have been able to stop the spread of diseases and, in some cases, eradicate those diseases completely. A very good example is the smallpox disease which claimed a lot of lives before a vaccine was discovered (UNICEF, 2014). VPDs bring a lot of cost, both monetary and non-monetary (Cynthia, Whitney, Zhou, Singleton, & Schauchat, 2014), to the individual, the society, the nation and the world as a whole. (Cynthia et. al, 2014). National immunisation survey of the United States of America observed that immunisation has the potential to prevent an estimated 322 million illnesses, 21 million hospitalisations, 732,000 deaths and a net savings of $295 billion direct cost as well as a net saving $1.38 trillion in total societal cost between the periods of 1994 to 2013 (Cynthia et al., 2014). The World Health Organization estimated that, in 2014, a total of about 5,427,000 individuals died from a vaccine-preventable disease, with more than half of the deaths (about 57 percent) occurring in Africa. These deaths generally occur mostly among pre-school children usually less than five years old (WHO, 2016). To improve vaccine coverage in the world, particularly in Africa, the WHO instituted the EPI in 1974, with the aim of improving immunisation coverage in the world by improving availability and accessibility to vaccines (Cynthia et al., 2014). Today, some areas in the world have been able to reach 80% and over coverage of their populace eligible to receive various vaccines (WHO, 2014). It is worth noting © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 5 that, before the inception of the EPI, Africa was recording less than five percent of total immunisation coverage (WHO, 2014). Refusal and obstacles to immunisations, therefore, serve as a major contributory factor to the spread and devastating effects of vaccine-preventable diseases (Schwartz, 2013). Vaccine-preventable diseases are usually cause-specific; that is, they are either caused by viruses or bacteria (WHO, 2009). These diseases become more serious in populations where people are vulnerable or not immunised against the disease-causing organism; however, if the majority of the population are protected, they provide herd immunity which turns to protect those not immunised (Fine, Eames & Heymann, 2011). When childhood vaccines are delivered at the right time, they increase the protection of children from vaccine-preventable diseases, minimise the risk of children getting infected and decrease the chance of outbreaks of the disease (McNair, 2014) WHO estimates that 17 percent of global annual childhood mortality could be prevented through increasing routine vaccination coverage of which 2.2 percent could be prevented through pertussis vaccination, 2.3% through Hib vaccination, 1.3% through measles vaccination, 0.7% through tetanus vaccination, 5.2% through rotavirus vaccination and 5.4% through pneumococcal vaccination (WHO 2013). Immunisation and vaccination are a key indicator of the Sustainable Development Goal (SDG) 3. Target 3.1.1, which is a proportion of the target population covered by all vaccines included in their national programme and target 3.8.1 which is coverage of essential health services or “universal health coverage (UHC)- index of which immunisation is 1 of 16 tracer indicators aggregated © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 6 together (UNDP, 2016). This makes immunisation and vaccination essential since it has the potential for contributing to the achievement of SDG 3. A set target for vaccination coverage by WHO is 80%. There has been an increase in the vaccination coverage in Ghana over the years. From 47 percent in 1988 to 77 percent in 2014 which is very close to the national target of 80 percent (GSS, GHS and ICF Marco, 2015; WHO, 2015). There exist urban-rural differences and regional disparities with regards to vaccination in Ghana (WHO, 2015). There are also some children who do not receive any of the vaccines (Armah et al., 2016). The question one may ask is: why is it that some children do not receive some of the vaccines? why are some children not receiving any of the childhood vaccines? Few studies have been conducted in Ghana over the years about vaccination and immunisation in Ghana. Some of the studies concentrated on issues relating to a particular type of vaccine against a specific health outcome (Armah et al., 2016; Binka et al., 2003; Coleman, Levison, Sangi-Haghpeykar, 2011; Enweronu-Laryea, et al., 2014; Woods & Shattuck, 2000). Also, there is evidence suggesting that most of the studies on childhood vaccination have been conducted in some selected places in Ghana (Adokiya, Baguune & Ndago, 2017; Bruce, 2017). Studies that have been conducted nationwide considered the effects of childhood immunisation (Matthews and Diamond, 1997) and the inequalities in child immunisation (Asuman et al., 2018). However, there has been no studies that consider a nationwide investigation into the determinates of childhood vaccination and it is for these reasons that there is a need for a nationally representative study in Ghana to assess childhood vaccination © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 7 in Ghana from 1998 to 2014 with emphasis on children under 24 months to fill the gap and contribute to the discourse of childhood vaccination. The focus is on children under 24 months because according to Ghana’s vaccination schedule, by age 23 months, a child is expected to receive all the needed vaccines. Objectives of the Study The main objective of the study is to assess childhood vaccinations in Ghana from 1998 to 2014. Specifically, the study seeks to: 1. Analyse the trend of childhood vaccination coverage from 1998 to 2014 2. Examine the patterns of childhood vaccination from 1998 to 2014 3. Assess the determinants of childhood vaccination from 1998 to 2014 Research Questions 1. What is the trend of childhood vaccination coverage in Ghana from 1998 to 2014? 2. What are the patterns of childhood vaccination in Ghana from 1998 to 2014? 3. What are the determinants of childhood vaccination in Ghana from 1998 to 2014? Rationale of the Study The health of infants and children is of great concern to the international community and in recent times, has gained much attention. As a result of this, both the Millennium Development Goals (MDGs) and the Sustainable Development Goals (SDGs) attest to this fact. It is, therefore, necessary to understand the trends © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 8 and determinants of vaccination among children in Ghana because it affects their health. Secondly, the study will be relevant to key institutions such as the Ministry of Health (MoH), GHS and Non-Governmental Organisations (NGOs) to understand the trends and determinants of vaccination in Ghana. The study will inform health policies to improve child survivorship and reduce child mortality and morbidity. The study will further provide an understanding of childhood vaccination in Ghana and add to existing discourse on the subject. It will also serve as the basis for further research on issues of child health. Organisation of the Study The study is organized in five chapters. Chapter One includes the background to the study, statement of problem, objectives of the study, hypothesis of the study, rational of the study and the organization of the study. Chapter Two presents the literature on coverage and determinates of childhood vaccination and related issues as well as the theoretical and conceptual frameworks of the study. The third chapter also focuses on the methods of the study. Among these issues are the study area, target population, research design, sources of data, sampling design and sample size, acquisition of data and analysis of the data, and limitations of the study. The analysis and discussion of results are presented in Chapter Four. The last chapter is devoted to summary of the main findings, conclusions, recommendations, and suggestions for further research. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 9 CHAPTER TWO LITERATURE REVIEW Introduction Vaccination of children has become a worldwide concern and has attracted much attention; hence, the introduction of the EPI in 1974 as a means of providing the needed vaccines, especially for children under five years of age (WHO, 2012). This chapter comprises the empirical and theoretical literature as well as the conceptual framework. The empirical literature explores the history of EPI and specific issues on childhood vaccination. The theoretical literature reviews the Theory of Planned Behaviour, the Health Belief Model and the Health Care Service Utilisation Model. Concepts of Vaccination and Immunisation Vaccination is the administration of antigenic material (a vaccine) to stimulate an individual’s immune system to develop adaptive immunity to a pathogen (Chang, Brewer, Rinas, Schmitt & Smith, 2009). Vaccines can prevent or ameliorate infectious diseases. The effectiveness of vaccination has been widely studied and verified (Chang et. al., 2009; Fiore, Bridges & Cox, 2009). It is the most effective method of preventing infectious diseases (United States Centres for Disease Control and Prevention, 2013). Vaccination is largely responsible for the worldwide eradication of smallpox and the elimination of diseases such as polio, measles, and tetanus from much of the world (WHO, 2015). © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 10 Immunisation is the process of stimulating an active immunology defence in preparation for meeting the challenge of future exposure to disease (Elder & Bullough, 1990). Thus, it is the introduction of weakened, live or dead micro- organism called vaccines to the body system to stimulate the production of the antibodies to confer immunity. In other words, immunisation consists of the administration of a modified or attenuated product of a micro-organism to mimic natural infection and evoking an immunologic response that presents little or no risk to the recipient. A vaccine is a suspension of antigens in a base (Fatriregun & Etukiren, 2014). Vaccines are either made from the organisms which are dead or alive but are harmless. They may also be made from toxoid which relates to the toxins or poisons produced by the organism which do no damage but protect the body. There are three main types of vaccines: live vaccines, dead vaccines, and antitoxins (CDC, 2013). A live vaccine consists of living but weakened micro-organisms (attenuated) grown in the laboratory. Examples of these vaccines are the Poliomyelitis, Measles and BCG vaccines. A dead vaccine contains dead micro-organisms but when introduced into the body, enables it to produce antibodies. An example of such vaccines is DPT/HibHepB vaccine. Antitoxins are the altered harmless toxins of the organism which, when injected into the body, cause it to make antibodies. An example is the tetanol toxoid vaccine (CDC, 2013). History of Vaccination According to Gross and Sepkowitz (1998), the process of inoculation was first used by Chinese physicians in the 10th century. According to Ole Lund, the © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 11 earliest documented examples of vaccination are from India and China in the 17th century, where vaccination with powdered scabs from people infected with smallpox was used to protect against the disease (Mehmed & Basri, 2016). The tradition of vaccination may have originated in India in AD 1000 (Mehmed & Basri, 2016). The mention of inoculation in the Sact'eya Grantham, an Ayurvedic text, was noted by Henri Marie Husson in the journal Dictionaire des sciences médicales (Iseman, 2002). However, according to Wujastyk (1995), the idea that inoculation originated in India has been challenged, as few of the ancient Sanskrit medical texts described the process of inoculation. Accounts of inoculation against smallpox in China can be found as early as the late 10th century and it was reported to be widely practiced in China in the reign of the Longqing Emperor during the Ming Dynasty (Needham et. al, 2013). Evidence from Silverstein (2009) indicates that two reports on the Chinese practice of inoculation were received by the Royal Society in London in 1700; one by Dr. Martin Lister who received a report by an employee of the East India Company stationed in China and another by Clopton Havers. According to Voltaire (1742), the Turks derived their use of inoculation from neighbouring Circassia. Voltaire does not speculate on where the Circassians derived their technique from, though he reports that the Chinese have practiced it “these hundred years”. The Greek physicians Emmanuel Timonis (1669–1720) from the island of Chios and Jacob Pylarinos (1659–1718) from Cephalonia practised smallpox inoculation at Constantinople in the beginning of 18th century (Karaberopoulos, © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 12 2006) and published their work in Philosophical Transactions of the Royal Society in 1714 (Pylarinum, 1714; Timonius & Woodward, 1716). This kind of inoculation and other forms of variolation were introduced into England by Lady Montagu, a famous English letter-writer and wife of the English ambassador at Istanbul between 1716 and 1718, who almost died from smallpox as a young adult and was physically scarred from it. Inoculation was adopted both in England and in America nearly half a century before Jenner's famous smallpox vaccine of 1796 (Henricy, 1796) but according to Gross and Sepkowitz (1998), the death rate of about 2% from this method meant that it was mainly used during dangerous outbreaks of the disease and remained controversial. It was noticed during the 18th century that people who had suffered from the less virulent cowpox were immune to smallpox, and the first recorded use of this idea was by a farmer, Benjamin Jesty at Yetminster in Dorset, who had suffered the disease and transmitted it to his own family in 1774, his sons subsequently not getting the mild version of smallpox when later inoculated in 1789. It was Edward Jenner in Berkeley, who established the procedure by introducing material from a cowpox vesicle on Sarah Nelmes, a milkmaid, into the arm of a boy named James Phipps. Two months later, he inoculated the boy with smallpox and the disease did not develop. In 1798, Jenner published An Inquiry into the Causes and Effects of the Variolae Vacciniae, which coined the term vaccination and created widespread interest. He distinguished ‘true’ and ‘spurious’ cowpox and developed an “arm-to-arm” method of propagating the vaccine from the vaccinated individual’s pustule. Early attempts at confirmation © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 13 were confounded by contamination with smallpox, but despite controversy within the medical profession and religious opposition to the use of animal material, by 1801, his report was translated into six languages and over 100,000 people were vaccinated (Gross & Sepkowitz, 1998). Since then, vaccination campaigns have spread throughout the globe, sometimes prescribed by law or regulations. Vaccines are now used against a wide variety of diseases. Louis Pasteur, further, developed the technique during the 19th century, extending its use to killed agents protecting against anthrax and rabies. The method Pasteur used entailed treating the agents for those diseases so they lost the ability to infect, whereas inoculation was the hopeful selection of a less virulent form of the disease, and Jenner’s vaccination entailed the substitution of a different and less dangerous disease. Pasteur adopted the name vaccine as a generic term in honour of Jenner’s discovery. Maurice Hileman, who was an American microbiologist and specialized in vaccinology, was the most prolific vaccine inventor, developing successful vaccines for measles, mumps, hepatitis A, hepatitis B, chickenpox, meningitis, pneumonia and ‘Haemophilus influenza’ (Offit, 2007). In modern times, the first vaccine-preventable disease targeted for eradication was smallpox (WHO, 2014). WHO coordinated this global eradication effort. The last naturally occurring case of smallpox occurred in Somalia in 1977. In 1988, the governing body of WHO targeted polio for eradication by 2000 (WHO, 2014). Although the target was missed, eradication is very close (UNICEF, 2015). In 2000, the Global Alliance for Vaccines and Immunisation was established to strengthen routine vaccinations and © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 14 introduce new and under-used vaccines in countries with a per capita GDP of under US $1000. The Extended Program on Immunisation The WHO initiated the EPI in May 1974, with the goal of ensuring universal availability of key life-saving vaccines to children in the developing world. In the recommendation of the World Health Assembly (WHA), the programme was to ensure that member states develop or maintain immunisation and surveillance programme against measles, poliomyelitis, tuberculosis, diphtheria, pertussis and tetanus (WHO, 1996). In furtherance of this recommendation, a special fund was established to support EPI and a Global Advisory Group (GAG), involving experts in immunisation and disease control, was convened (Sebastian & Weiser, 1979). In 1984, WHO established a standard schedule for these vaccines targeting optimal points of early childhood in order to allow children to develop immunity safely, effectively and as early as possible (WHO, 2012). According to Plotkin and Plotkin, (2008), this schedule takes into account age-specific risks of disease, age-specific risks of complications and age-specific patterns of immunity. Immunisation against VPDs through EPI is one of the most economical public health interventions available (UNICEF, 2002) that contributed extensively to achieving the MDG to reduce the mortality rate of children under five by two thirds between 1990 and 2015 (UNICEF, 2015). Immunisation against major diseases was one of the eight elements of the PHC, thus making the EPI an integral part of the PHC system. In May 1982, the World Health Assembly reviewed the EPI and urged member states of WHO to immunise all their children by 1990 (UNICEF, 1983). © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 15 At this Assembly, member states were called on to promote EPI within the context of Primary Health Care (PHC), invest adequate human and financial resources in EPI, ensure that programmes are continuously evaluated and pursue research efforts as part of the programme operations. UNICEF also developed the Universal Child Immunisation Programme and set a global target of 80% immunisation coverage among infants worldwide by 1990. Ten years after the initiation of the EPI, in 1984, the WHO established a standardized vaccination schedule for the original EPI vaccines: Bacillus Calmette Guḗrin (BCG), diphtheria-tetanus-pertussis (DTP), oral polio and measles. Increased knowledge of the immunologic factors of disease led to new vaccines being developed and added to the EPI’s list of recommended vaccines: Hepatitis B (HepB), yellow fever in countries endemic for the disease, and Haemophilus influenza meningitis (Hib) conjugate vaccine in countries with high burden of disease (Jamison et. al., 2013). In 1999, the Global Alliance for Vaccines and Immunisation (GAVI) was created with the sole purpose of improving child health in the poorest countries by extending the reach of the EPI. The GAVI brought together a grand coalition, including the UN agencies and institutions (WHO, UNICEF, the World Bank), public health institutes, donor and implementing countries, the Bill and Melinda Gates Foundation and The Rockefeller Foundation, the vaccine industry, non- governmental organisations (NGOs) and many more (McCoy et. al.,2009). The creation of the GAVI has helped to renew interest and maintain the importance of © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 16 immunisations in battling the world’s large burden of infectious diseases (McCoy et. al.,2009.). The current goals of the EPI are: • to ensure full immunisation of children under one year of age in every district, • to globally eradicate poliomyelitis, • to reduce maternal and neonatal tetanus to an incidence rate of less than one case per 1,000 births by 2005, • to cut in half the number of measles-related deaths that occurred in 1999, and • to extend all new vaccine and preventative health interventions to children in all districts in the world. In addition, the GAVI has set up specific milestones to achieve the EPI goals: that by 2010 all countries have routine immunisation coverage of 90% of their child population, that HepB be introduced in 80% of all countries by 2007, and that 50% of the poorest countries have Hib vaccine by 2005 (Halder, Cochi, Bilous & Cutts, 2004). Implementation of the EPI In each of the United Nations’ member states, the national governments create and implement their policies for vaccination programs following the guidelines set by the EPI (WHO, 2012). Setting up an immunisation programme was multifaceted and contained many complex components including a reliable cold chain system, transport for the delivery of the vaccines, maintenance of vaccine stocks, training and monitoring of health workers, outreach educational © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 17 programs to inform the public, and a means of documenting and recording which child receives which vaccines. Each region has slightly varying ways of setting up and implementing its immunisation programs based on their level of health infrastructure (WHO, 2012). Some areas will have fixed sites for vaccination (for example, health care facilities such as hospitals or health posts that include vaccination with many other health care activities). But in areas where the number of structured health facilities is small, mobile vaccination teams consisting of staff members from a health facility can deliver vaccines straight to individual towns and villages (McKee, 2016). These ‘outreach’ services are often scheduled throughout the year, however, in especially under-developed countries where proper communication and infrastructure are absent, cancellation of the planned immunisation visits leads to deterioration of the program. A better strategy in such countries is the ‘pulse immunisation’ technique, where ‘pulses’ of vaccines are given to children in annual vaccination campaigns (Jamison et. al., 2013; McKee, 2016) Additional strategies are needed if the area of the program consists of poor urban communities, because such areas tend to have low uptake of vaccination programs. Door-to-door canvassing, also referred to as channeling, is used to increase uptake in such hard-to-reach groups. Finally, periodic national-level mass vaccination campaigns are being increasingly included in the programs (McKee, 2016). © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 18 Evaluation of the EPI In each country, immunisation programs are monitored using two methods: an administrative method and through community-based surveys. The administrative method uses immunisation data from public, private, and NGO clinics. Thus, the accuracy of the administrative method is limited by the availability and accuracy of reports from these facilities. This method is easily performed in areas where government services deliver the immunisations directly or where the government supplies the vaccines to the clinics. In countries without the infrastructure to do this, community-based surveys are used to estimate immunisation coverage (McKee, 2016) Community-based surveys are applied using a modified cluster sampling survey method developed by the WHO. Vaccine coverage is evaluated using a two-stage sampling approach in which 30 clusters and seven children in each cluster are selected. Health care workers with no or limited background in statistics and sampling are able to carry out data collection with minimal training (McKee, 2016). This provides a way to get information from areas where there is no reliable data source. It is also used to validate reported vaccine coverage (for example, from administrative reports) and is expected to estimate vaccine coverage within 10 percent (Jamison et. al., 2013). Surveys or questionnaires, though frequently considered inaccurate due to self-reporting, can provide more detailed information than administrative reports alone. If home-based records are available, vaccination status is determined and dates of vaccination can be reviewed to determine if they were given at an ideal age © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 19 and in appropriate intervals. Missed immunisations can be identified and further qualified. Systems of vaccine delivery besides clinics used for administrative evaluation can be identified and included in the analysis (McKee, 2016). Results of the Implementation of the EPI Before the initiation of the EPI, child vaccination coverage for tuberculosis, diphtheria, pertussis, tetanus, polio, and measles was estimated to be fewer than 5 percent. Now, not only has coverage increased to 79 percent (CDC, 2006), it has been expanded to include vaccinations for hepatitis B, Haemophilus influenzae type B, rubella, tetanus, and yellow fever. The impact of increased vaccination is clear from the decreasing incidence of many diseases. For example, measles deaths decreased by 60% worldwide between 1999 and 2005, and polio, although missing the goal of eradication by 2005, has decreased significantly as there were fewer than 2,000 cases in 2006 (CDC, 2006). The Expanded Programme on Immunisation in Ghana The Expanded Programme on Immunisation was adopted by Ghana in 1978. According to MOH/WHO/UNICEF in 1989, the EPI policy states that, “The government of Ghana has adopted the policy of reducing to manageable levels the morbidity and mortality of children under 5 years against the following common childhood preventable diseases, tuberculosis, measles, diphtheria, poliomyelitis, pertussis, neonatal tetanus, and yellow fever, through the Expanded Programme on Immunisation (EPI).” © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 20 The Specific Objectives of the Programme were: • to promote continued immunisation of infants through the existing health delivery system. • to combat diseases for which there are effective vaccines and immunisation methods. • to ensure that by 1990, all children in the first year of life will have ready access to immunisation and 80% of them will be fully immunised. • to ensure that a course of tetanus toxoid immunisation is started in at least 80% of women aged between 12-44 years in any particular year. • to reduce the incidence and mortality due to measles, whooping cough, polio and tetanus by 80%, and tuberculosis by 50% of the present levels. • to improve data collection and to establish better disease surveillance through regular reporting from all the health facilities. • to achieve immunisation in all health facilities. • to promote EPI with Control of Diarrhoea Diseases (CDD) within the framework of Primary Health Care (PHC) delivery system. • to improve EPI services and coverage through research. • to strengthen intersectoral co-operation as well as teamwork amongst health workers and the community in order to improve acceptance and coverage of EPI and CDD services. • to improve and expand the coverage of health education. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 21 The EPI schedule of Ghana specifies the age at which a particular vaccine is to be given and the disease that is to be protected against. Table 1 below shows the present EPI schedule for Ghana. Table 1: The EPI schedule for Ghana Vaccine Age Given Disease Protected Against Bacille Calmette-Guerin (BCG) At Birtth Tuberculosis Oral Polio Vaccine (OPVo) At Birth Poliomyelitis OPV1 6 week Poliomyelitis OPV2 10 weeks Poliomyelitis OPV3 14 weeks Poliomyelitis DPT (Diphtheria-Pertussis -Tetanus) 6 week Diphtheria -Pertussis – Tetanus DPT 10 weeks Diphtheria -Pertussis – Tetanus DPT 14 weeks Diphtheria -Pertussis – Tetanus Measles 9 months Measles Yellow Fever 9 months Yellow Fever Source: Ministry 0f Health, Ghana (1989) There was an update in 1997 of the EPI policy guidelines and the following goals were set: • Control of Measles; • Elimination of Neonatal Tetanus; • Eradication of Poliomyelitis; • Control of Hepatitis B; and • Control of yellow fever all by the year 2000. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 22 EPI and the Health Sector Reforms The MoH began implementing institutional health reforms in 1997. The thrust of the reforms is decentralization to the district (the operational) level and the integration of services at the district level. EPI services have, therefore, been decentralized to the district level where its operational activities are a component of an integrated package of health interventions provided by the District Health Management Teams (DHMT), the health providing institutions (public, private, and NGO), and the Sub District Health Teams. Though some targeted funds are still received for some EPI activities (RED approach, Surveillance, Training of Midwives in EPI, and Injection Safety), most of the funding comes through the “common basket”. The capacity of the district level to operate as a viable, efficient and technically competent unit was enhanced by several initiatives and policies. The Strengthening of District Health Systems (SDHS) initiative (between the late 80s and early 90s) was geared towards strengthening the capacity of the District Health Management Teams to undertake situational analysis, problem identification and analysis, planning, implementing and evaluation of their work. The District and sub-districts were, therefore, trained to strengthen the planning process of district operations. The District Health System Operation (DISHOP) training in the late 90s followed this intervention. The District Director of Health Services (DDHS) is ultimately responsible for the Management of EPI in the District. This responsibility is often delegated to a DHMT member who is either a Technical Officer (DC) or a public health nurse. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 23 The creation of Budget Management Centres (BMCs) envisaged that funds are made available to meet the operational cost of implementing the operational plans of such BMCs. The operational planning and funding of EPI activities within the district and sub-districts are very much dependent on the availability of funds to the DHMT and the Sub District Health Teams (SDHT). It is also very much dependent on the discretion and priorities of the leadership of especially the DHMTs. EPI service delivery is actually done by the sub-district health teams, consisting largely of community health nurses, field technicians, midwives, and public health nurses. Inadequate or non-release of BMC funds to the sub-districts by the DHMT remains a real threat to the effective EPI service delivery in the country. The Comprehensive Multi-Year Plan The comprehensive multi‐year plan (cMYP) is the medium-term plan for the Expanded Programme on Immunization (EPI) in Ghana. As expected, it provides the strategic direction of the immunization programme for the period. The cMYP is always prepared in consonance with the 5‐year strategic plan of the health sector. The current cMYP (2007‐2011) had to be revised to 2014 to accommodate the plans to introduce three new vaccines – pneumococcal, rotavirus and second dose measles. Ghana launched the Expanded Programme on Immunization (EPI) in June 1978 with six antigens – BCG, measles, diphtheria‐pertussis‐tetanus (DPT) and oral polio for children under one year of age together with tetanus toxoid (TT) vaccination for pregnant women. The launch was in response to the national health policy to reduce morbidity and mortality of vaccine preventable diseases which © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 24 then contributed significantly to both infant and child mortality in the country. It was also in consonance with the immunization policy of the government which sought to ensure that all children receive these vaccines before their first birthday of life. Three strategies are implored for the delivery of the immunization services in the country‐ static, outreach and campaigns to reach out to most of the unreached populations. The goal is to reduce burden of vaccine preventable diseases with the view of contributing to the overall poverty reduction and health care strengthening in the country. The Expanded Programme on Immunization (EPI) which is responsible for immunization in Ghana is located within the Diseases Control Department (DCD) of the Public Health Division of the Ghana Health Service. It is headed by a Public Health Specialist and assisted by trained personnel who are specialists in areas that include logistics management, data management, cold chain management, injection safety, social mobilization and communication. Ghana’s first major challenge and experience in introducing new vaccines into the routine immunization programme was at the time of introducing the haemophilus influenza type b (Hib) and the hepatitis B vaccines in the combination of DPT‐HepB+Hib (also referred to as pentavalent) in 2002 with GAVI support. The National Immunization Programme is one of the most successful and cost effective programmes implemented in Ghana. The uniqueness of the NIP has been the innovativeness and adaptation it has gone through with the support of national and international partners. As a vertical programme the NIP has successfully transformed itself and integrated within the country’s decentralized © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 25 health system. EPI in Ghana aims at protecting every child in Ghana from nine common childhood diseases; namely, tuberculosis, poliomyelitis, diphtheria, neonatal tetanus, whooping cough, hepatitis B, haemophilus influenza type b, measles and yellow fever. The development of immunization in Ghana was very slow in the early 1990s. In the early 2000s, the delivery was accelerated with the global initiative and support from GAVI. Subsequently with the introduction of new and underutilized vaccines and sustained NIDs, more children have been immunized. Greater attention has also been drawn to the importance of the national immunization programme and its integration with the health systems, gaining overall policy support. Immunization continues to be one of the most cost‐effective of all child survival health interventions. Traditionally, immunization has had children and women as the main foci of attention but recently this is being broadened to include the wider population in line with the current global immunization vision and strategy (GIVS). Ghana intends to introduce two new vaccines (Rotavirus and Pnneumocco) and a second dose for measles. Goals and Objectives This comprehensive Multi‐Year Plan (cMYP) of Action covers the fiscal years 20010‐2014. The objectives and activities set forth in this Multi‐Year Plan provide the framework required to meet the previously stated goal of reducing infant and child mortality and morbidity associated with vaccine ‐preventable diseases (VPD). Further, this Plan addresses new challenges and expands the previous plan by providing guidelines for the introduction of new © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 26 vaccines. The 2010‐2014 cMYP of Ghana is aimed at achieving five major goals which are within the context of the goals of GIVS with strategic components as described below: 1. Reach out and protect more people than being done currently 2. Rapidly increase immunity to selected VPDs in order to accelerate Reduction of morbidity and mortality from VPDs 3. Introduce New & Under‐utilized Vaccines 4. Strengthen surveillance 5. Integrate EPI with other Interventions in the context of Health System Development Objectives of the cMYP Ghana immunization objectives are linked to the sector objectives (objective 3) and the efforts for achieving the MDGs. The specific objectives of the National Immunisation Programme are 1. Reach everyone targeted for immunization to achieve and sustain by 2014, 94% coverage in all childhood immunisations and 85% for Tetanoid Toxins for Pregnant women 2. Improve communication, advocacy and information dissemination 3. Strengthen surveillance system 4. Improve programme management and integration with health systems The cMYP provides a comprehensive overview of the NIP and is the document to provide direction guidance to national and sub‐national levels for © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 27 incorporation into their annual plans. It is also the documents that will advise national policies in setting national targets for all immunization indicators. National performance will therefore be monitored based on the indicators set in the cMYP. The cMYP contain a set of programme and financial indicators. These indicators will be monitored and feedback provided to policy and programme managers. Monitoring of the cMYP will be done within the existing health sector monitoring and reporting framework. Trends of Childhood Vaccination Rates of vaccine-preventable diseases have decreased in many parts of the world in the past few decades, but many children remain unvaccinated (de Figueiredo et al., 2016). In 2012, WHO reported that 21·8 million children younger than 1 year had not completed diphtheria, tetanus, and pertussis (DTP) immunisation series, with a similar number not receiving a single measles vaccination (WHO, 2012). This brought a reason for researchers across the globe to have carried out studies on trends of childhood vaccination Semali (2010) carried out a study that aimed at determining the trend in disparities in completion of immunisation using Tanzania Demographic and Health Surveys (DHS). The last three data sets (1990, 1996 and 2004) were downloaded and analysed, of which all children between 12 -23 months would have completed all vaccinations required by the 12th month. Results from the study showed that, across the time periods 1990, 1996 to 2004/05, the percentage of children completing vaccination was similar (71.0% in 1990, 72.7% in 1996 and 72.3% in 2005). © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 28 Similarly, Singh (2013) assessed the urban-rural and gender difference in child immunisation coverage during 1992–2006 across six major geographical regions in India. Three rounds of the National Family Health Survey (NFHS) conducted during 1992–93, 1998–99 and 2005–06 were analysed. Bivariate analyses, urban-rural and gender inequality ratios, and the multivariate-pooled logistic regression model were applied to examine the trends and patterns of inequalities over time. The analysis of change over one and half decades (1992– 2006) showed considerable variations in child immunisation coverage across six geographical regions in India. Full immunisation coverage had increased from 35% in 1992 to 44% in 2006. Again, Tao, Petzold and Forsberg (2013) evaluated the change in routine vaccination coverage over time based on survey data and compared it to estimations by the WHO and United Nations Children’s Fund (UNICEF). Data of vaccination coverage of children less than 5 years of age was extracted from Demographic and Health Surveys (DHS) conducted in 71 low- and middle-income countries from 1986 to 2009. Overall, trends for vaccination coverage of tuberculosis, diphtheria, tetanus, pertussis, polio and measles were analysed and compared to WHO and UNICEF estimates. Results from the study showed that from 1986 to 2009, the annual average increase in vaccination coverage of the studied diseases ranged between 1.53 and 1.96% units according to DHS data. Vaccination coverage of diphtheria, tetanus, pertussis, polio and measles was all under 80% in 2009. Ushie, Fayehun and Ugal (2014) carried out a study to examine the trends in vaccination coverage for the assessment of successes or failures of the © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 29 immunisation efforts. The study used four NDH surveys datasets between 1990 and 2008, which generated child health information including the proportion that had had any or all basic childhood vaccines. A combined total of 44,071 (weighted) children were involved in the study. Results of the study showed that more complete vaccination coverage was reported in the 1990 survey, followed by 2008, 1999 and 2003. Maternal Age as a Determinate of Childhood Vaccination In relation to maternal age as a determinant of childhood vaccination, Kamau and Esamai (2001) conducted a study to establish the factors that determine the levels of immunisation among children under five years in Mathare Valley. A cross-sectional study describing the situation at a point in time was employed in the study. The study population was mothers with children under five years in Mathare Valley who had been resident there for a period not less than five years prior to the study. Results from the study showed that age is a significant determinant of immunisation coverage. Russo et al. (2015) carried out a study to identify determinants for incomplete vaccination status and to assess the risk of poliovirus spread in the study population. A cross-sectional household survey was conducted in November- December 2013, using the WHO two-stage sampling design. An interviewer- administered questionnaire was used to obtain information from consenting parents of children aged 12–23 months. Vaccination coverage was assessed by vaccination card and parents’ recall. Results from the study showed that incomplete immunisation status was associated with younger mothers. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 30 Lastly, Adokiya, Baguune and Ndago (2017) evaluated the immunisation coverage and its associated factors among children aged 12–23 months in Techiman Municipality, Ghana. A cross-sectional cluster survey was conducted among 600 children. Data was collected using a semi-structured questionnaire through face-to- face interviews. Descriptive statistics such as percentages, frequencies and cross- tabulations were performed to estimate the odds ratio of not being fully immunised. Results showed that children of older mothers were more likely to be fully immunised. For example, the AOR was 0.15 (95%CI 0.05–0.87) when the age was 40–49 years, compared to less than 20 years. Maternal Educational Level as a Determinate of Childhood Vaccination On maternal education as a determinant of childhood vaccination, Kamau and Esamai (2001) conducted a study to establish the factors that determine the levels of immunisation among children under five years in Mathare Valley. A cross- sectional study describing the situation at a point it time was employed in the study. The study population was mothers with children under five years in Mathare Valley who had been resident there for a period not less than five years prior to the study. Results from the study showed that level of education is a significant determinant of immunisation coverage. Hu, Li, Chen, Chen and Qi (2013) carried out a study to determine the coverage of childhood immunisation appropriate for age among socio- economically disadvantaged recent migrants living in East China and to identify the determinants of full immunisation uptake among these migrant children. The study was a cross-sectional survey of 1,426 migrant mothers with a child aged ≤24 © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 31 months, who were interviewed with a pretested questionnaire. Results of the study showed that the likelihood of a child receiving full immunisation rise with parents’ educational level. Mukungwa (2015) utilized the Zimbabwe Demographic and Health Survey (ZDHS) data to analyse the variables of immunisation status of children aged 12- 23 months in Zimbabwe. A multivariate binary logistic regression analysis of the data was performed. The data consisted of 978 children aged 12-23 months from the selected households. Results showed that children of mothers with secondary education and above were more likely to be vaccinated than children of mothers with no formal education. Finally, Mbengue et al., (2017) assessed routine immunisation uptake and factors associated with full immunisation status among Senegalese children aged 12–23 months. The study used the 2010–2011 Senegalese Demographic and Health Survey data. The DHS was a two stages cross-sectional survey carried out in 2010– 2011. The analysis included 2199 children aged 12–23 months. The interviewers collected information on vaccine uptake based on information from vaccination cards or maternal recall Univariate and multivariable logistic regression models were used to identify the determinants of full childhood immunisation. Results from the study showed that mothers who attended at least secondary education level had full childhood immunisation compared to those with less than secondary level education. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 32 Wealth Status as a Determinant of Childhood Vaccination Hu, Li, Chen, Chen and Qi (2013) carried out a study to determine the coverage of childhood immunisation appropriate for age among socio- economically disadvantaged recent migrants living in East China and to identify the determinants of full immunisation uptake among these migrant children. The study was a cross-sectional survey of 1,426 migrant mothers with a child aged ≤24 months, who were interviewed with a pretested questionnaire. Results of the study showed that higher household income significantly increases the likelihood of full immunisation. Mukungwa (2015) utilized the Zimbabwe Demographic and Health Survey (ZDHS) data to analyse the variables of immunisation status of children aged 12- 23 months in Zimbabwe. A multivariate binary logistic regression analysis of the data was performed. The data consisted of 978 children aged 12-23 months from the selected households. Results showed that the likelihood of vaccination increased with wealth status. Lakew, Bekele and Biadgilign (2015) carried out a study to identify factors associated with full immunisation coverage among children aged 12-23 months in Ethiopia. This study used the 2011 Ethiopian Demographic and Health Survey data. The survey was cross sectional by design and used a multistage cluster sampling procedure. A total of 1,927 mothers with children of 12-23 months of age were extracted from the children's dataset. Mothers' self-reported data and observations of vaccination cards were used to determine vaccine coverage. In terms of wealth status, there was a 40 % (AOR 95 % CI; 1.4 (1.06-1.94)) more likelihood in © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 33 receiving full vaccination among children born to mothers of rich wealth index group compared with children from women of poor wealth index group. Finally, Ekouevi et al., (2018) carried out a study to estimate the immunisation coverage among children aged 12–23 months and to identify factors associated with incomplete immunisation status in Togo. A cross-sectional survey was conducted in the six health regions of Togo. Children aged 12 to 23 months who were living with one of their parents or guardians from selected households were recruited for the study. Data was collected using a pre-tested questionnaire through face-to-face interviews. Multilevel logistic regression analyses were performed to assess factors associated with incomplete immunisation coverage. Results from the study showed that the likelihood of incomplete immunisation in children decreased with the increase in household’s income (a OR = 0.73, 95% CI (0.58–0.93)). Region as a Determinant of Childhood Vaccination In terms of region as a determinant of childhood vaccination, Mukungwa (2015) utilized the Zimbabwe Demographic and Health Survey (ZDHS) data to analyse the variables of immunisation status of children aged 12-23 months in Zimbabwe. A multivariate binary logistic regression analysis of the data was performed. The data consisted of 978 children aged 12-23 months from the selected households. Results showed a significant association between region and variations in immunisation. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 34 Lakew, Bekele and Biadgilign (2015) carried out a study to identify factors associated with full immunisation coverage among children aged 12-23 months in Ethiopia. This study used the 2011 Ethiopian Demographic and Health Survey data. The survey was cross-sectional by design and used a multistage cluster sampling procedure. A total of 1,927 mothers with children of 12-23 months of age were extracted from the children’s dataset. Mothers’ self-reported data and observations of vaccination cards were used to determine vaccine coverage. With region, women from Afar, Amhara, Oromiya, Somali and Southern Nation and Nationalities People administrative regions were less likely to fully vaccinate their children compared to women who reside in Addis Ababa. Finally, Mbengue et al., (2017) assessed routine immunisation uptake and factors associated with full immunisation status among Senegalese children aged 12–23 months. The study used the 2010–2011 Senegalese Demographic and Health Survey data. The DHS was a two stage cross-sectional survey carried out in 2010– 2011. The analysis included 2199 children aged 12–23 months. The interviewers collected information on vaccine uptake based on information from vaccination cards or maternal recall Univariate and multivariate logistic regressions models were used to identify the determinants of full childhood immunisation. Results from the study showed that region has a statistically significant effect on immunisation. Place of Residence as a Determinate of Childhood Vaccination In terms of place of residence, Fernandez, Awofeso and Rammohan (2011) investigated rural-urban differentials in measles vaccination coverage among young Indonesian children and sought to identify key factors influencing the © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 35 probability of a child receiving the first dose of measles vaccination in Indonesia. Data used in the analyses were sourced from the nationally representative Indonesia Demographic and Health Survey, 2007. The influence of location of residence, household wealth, maternal and paternal education, total children ever born and use of skilled birth attendants on measles vaccination coverage was investigated using bivariate analysis and chi-square tests. The independent effects of these variables were established using binomial logistic regression analysis. Results from the study showed significantly higher rates of first-dose measles vaccination in urban as compared with rural areas. Again, Ushie, Fayehun and Ugal (2014) carried out a study to examine the trends in vaccination coverage for the assessment of successes or failures of the immunisation efforts. The study used four NDH Surveys datasets between 1990 and 2008, which generated child health information including the proportion that had had any or all basic childhood vaccines. A combined total of 44,071 (weighted) children were involved in the study. Results from the study showed that children in urban areas have consistently higher immunisation rates than those in rural areas. Abshoko (2016) carried out a study to statistically identify and analyse the various possible determinants of full immunisation among children in rural and urban households of SNNPRS, Ethiopia. The sampling technique employed was multistage stratified cluster sampling. Analysis of the study revealed that only 18.3% of children under 5 years of age were fully immunised in the region. Results of the multiple binary logistic regression showed that place of residence was an important determinant factor affecting full child immunisation in the region. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 36 Finally, Murtaza, Mustafa and Awan (2016) investigated the determinants and reasons for not vaccinating children in Pakistan. The study used the Pakistan Integrated Household Survey/Household Integrated Economic Survey 2001–2002 data. Demographic, distance to health facility, poverty status, literacy and education, and location of residence were used as determinants of non- immunisation of children. Descriptive statistics, including frequency distribution, proportions for categorical variables and mean for continuous variables, and logistic regression analysis were done using the Stata 11.0. Results indicated that a child living in the rural area had almost 1.7 times higher odds of not receiving vaccination as compared to a child living in the urban area. Mother’s Occupation as a Determinate of Childhood Vaccination In relation to occupation and childhood vaccination, Tagbo et al. (2014) assessed the vaccination coverage and its associated factors in children aged 11 -23 months in Enugu Metropolis. A cross-sectional study in which caregivers and their children pair, aged 11 -23 months attending children's outpatient clinics in Enugu Metropolis was undertaken. Respondents were selected consecutively while data were collected using a pretested interviewer-administered semi-structured questionnaire. Data were analysed using SPSS version 20.0 while level of significance was set at p < 0.05. Logistic regression analysis was used to identify independent predictors of full vaccination. Awasthi et al, (2015) conducted a study to identify the determinants of complete immunisation status among children aged 12–23 months in urban slums of Varanasi in India. A modified WHO EPI cluster sampling method was used for © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 37 sample selection. Data on 384 children were collected using a pretested questionnaire through house to house visit. Chi-square test, bivariate and multivariate logistic regression were used to assess the factors associated with complete immunisation status in the urban slums of Varanasi. Results from the study showed that employment status of mother is a strong determinant of complete immunisation. Again, Russo et al. (2015) carried out a study to identify determinants for incomplete vaccination status and to assess the risk of poliovirus spread in the study population. A cross-sectional household survey was conducted in November- December 2013, using the WHO two-stage sampling design. An interviewer- administered questionnaire was used to obtain information from consenting parents of children aged 12–23 months. Vaccination coverage was assessed by vaccination card and parents’ recall. Results from the study showed that immunisation status was associated with mother’s employment status. Finally, Adenike, Adejumoke, Olufunmi and Ridwan (2017) worked on maternal characteristics and immunisation status of children in North-central part of Nigeria. The objective of the study was to compare the mother’s characteristics and the child’s immunisation status in some selected rural and urban communities in the North-central part of Nigeria. A descriptive cross-sectional study, using a multistage sampling technique was used to select 600 respondent women with an index child between 0-12 months. Results from the study showed that there was a statistically significant association between respondents’ employment status and the child’s immunisation status (P<0.05). © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 38 Ethnicity as a Determinate of Childhood Vaccination In relation to ethnicity and childhood vaccination, Muiru (2009) investigated the factors influencing the utilisation of immunisation services in the low and high childhood mortality regions in Kenya. The study used secondary data obtained from KDHS 2003 and focused on 1,206 children aged 12 to 24 months. Cross tabulation was used to examine the hypothesized association between immunisation status and various independent variables. Chi-square test was carried out to show the level of significance of the association. Logistic regression was applied to show the effect of the selected variables on the dependent variable. The factors that were found to influence utilisation of immunisation services included ethnicity. Duah-Owusu (2010) investigated the relationship between social determinants of health and immunisation using data from the 2003 Ghana DHS (Demographic and Health Survey). The analyses utilised data from 5691 women who took part in the survey. It compared a national sample with the three northern- most regions rural sub-sample. These samples were women who were permanent residents of a household with a firstborn child below five years. The weighted national and three northern-most rural regions usual resident extracted sample numbered 2460 and 462, respectively. Bivariate correlation analyses and logistic regression analyses were carried out in both samples. The analyses showed that being fully immunised was associated with ethnicity. According to Duah-Owusu (2010), full immunisation was seen to be high with children born by women who are Akans. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 39 Morever, Duru et al. (2016) conducted a study to determine the immunisation coverage, status and the determinants in under 5-year-old children in Owerri Municipal, Imo State. The study was a community-based cross-sectional study involving 420 women and 743 under 5-year-old children. A multistage sampling technique was employed and data were collected using a pretested, semi structured interviewer administered questionnaire. The bivariate analysis showed statistically significant associations between the immunisation status of the children and ethnicity (p < 0.01). Lastly, Adokiya, Baguune and Ndago (2017) evaluated the immunisation coverage and its associated factors among children aged 12–23 months in Techiman Municipality, Ghana. A cross-sectional cluster survey was conducted among 600 children. Data was collected using a semi-structured questionnaire through face-to- face interviews. Descriptive statistics such as percentages, frequencies and cross- tabulations were performed to estimate the odds ratio of not being fully immunised. Results showed that, in comparing other ethnic groups to Akan (the main ethnic group), Frafra (AOR = 4.71, 95%CI = 146–15.18) have a higher likelihood of not being fully immunised whereas Kusaasi (AOR = 0.09, 95%CI = 0.02–0.51) have a higher likelihood of being fully immunised. Parity as a Determinate of Childhood Vaccination On parity as a determinant of childhood vaccination, Mukungwa (2015) utilized the Zimbabwe Demographic and Health Survey (ZDHS) data to analyse the variables of immunisation status of children aged 12-23 months in Zimbabwe. A multivariate binary logistic regression analysis of the data was performed. The © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 40 data consisted of 978 children aged 12-23 months from selected households. Results showed that children of the 1st birth order were more likely to be vaccinated than children of birth order 6+. Again, Awasthi et al. (2015) conducted a study to identify the determinants of complete immunisation status among children aged 12–23 months in urban slums of Varanasi in India. A modified WHO EPI cluster sampling method was used for sample selection. Data on 384 children were collected using a pretested questionnaire through house to house visit. Chi-square test, bivariate and multivariate logistic regression were used to assess the factors associated with complete immunisation status in the urban slums of Varanasi. Results from the study showed that parity less than three (OR = 2.84, 95% CI 1.98–3.73) is a strong determinant of complete immunisation Russo et al. (2015) conducted a study to identify determinants for incomplete vaccination status and to assess the risk of poliovirus spread in the study population. A cross-sectional household survey was conducted in November- December 2013, using the WHO two-stage sampling design. An interviewer- administered questionnaire was used to obtain information from consenting parents of children aged 12–23 months. Vaccination coverage was assessed by vaccination card and parents’ recall. Results from the study showed that incomplete immunisation status was associated with being the ≥3rd born child in the family. Similarly, Yenit, Assegid and Abrha (2015) also conducted a study to identify factors associated with incomplete childhood vaccinations among children 12-23 months of age in Machakel Woreda, East Gojjam Zone, and Northeast © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 41 Ethiopia. Community-based unmatched case-control study design with quantitative and qualitative methods of data collection was conducted in Machakel district, Northeast Ethiopia from March 20 to April 30, 2014. Quantitative data were collected from mothers who had 12-23 months aged children. Census was done to identify all cases and controls. For quantitative method, a sample of 154 cases and 154 controls were selected using a stratified multistage random sampling technique. Data were collected using a pretested structured questionnaire using interview data collection technique and data were entered into Epi Info software and analysed using Statistical Package for Social Sciences for windows version 16 and logistic regression methods. Results from the study showed that parity is a predictor of defaulting from completion of childhood vaccination. Lakew, Bekele and Biadgilign (2015) carried out a study to identify factors associated with full immunisation coverage among children aged 12-23 months in Ethiopia. This study used the 2011 Ethiopian Demographic and Health Survey data. The survey was cross-sectional by design and used a multistage cluster sampling procedure. A total of 1,927 mothers with children of 12-23 months of age were extracted from the children's dataset. Mothers' self-reported data and observations of vaccination cards were used to determine vaccine coverage. Vaccination coverage was high among single parity mothers. Marital Status as a Determinate of Childhood Vaccination On how marital status determines childhood vaccination, Weiss, Choudhary and Solomon (2013) explored determinants of routine immunisation performance to strengthen routine immunisation services. The researchers conducted a © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 42 secondary data analysis of the latest project household immunisation survey in 2011 and compared the findings to reports of past surveys in the Core Group Polio Project (CGPP program area) and at the Uttar Pradesh state level (as measured by children’s receipt of DPT vaccinations). This was done to judge if there is any evidence that routine immunisation services are being disrupted. They also model characteristics of survey respondents and respondents’ exposure to CGPP, communication activities against their children’s receipt of key vaccinations in order to identify determinants of routine immunisation coverage. Results from the study showed that marital status influences immunisation coverage. Similarly, Lakew, Bekele and Biadgilign (2015) carried out a study to identify factors associated with full immunisation coverage among children aged 12-23 months in Ethiopia. This study used the 2011 Ethiopian Demographic and Health Survey data. The survey was cross-sectional by design and used a multistage cluster sampling procedure. A total of 1,927 mothers with children of 12-23 months of age were extracted from the children’s dataset. Mothers’ self-reported data and observations of vaccination cards were used to determine vaccine coverage. Results from the study showed that children of never married women were more likely to have full vaccination coverage compared to those who were married. Finally, Adokiya, Baguune and Ndago (2017) evaluated the immunisation coverage and its associated factors among children aged 12–23 months in Techiman Municipality, Ghana. A cross-sectional cluster survey was conducted among 600 children. Data was collected using a semi-structured questionnaire through face-to- face interviews. Descriptive statistics such as percentages, frequencies and cross- © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 43 tabulations were performed to estimate the odds ratio of not being fully immunised. Results showed that children of married couples (AOR = 0.29, 95%CI = 0.13–0.68) were more likely to be fully immunised compared to children of never married/singles mothers. Religion as a Determinant of Childhood Vaccination With religion as a determinant of childhood vaccination, Weiss, Choudhary and Solomon (2013 explored determinants of routine immunisation performance to strengthen routine immunisation services. The researchers conducted a secondary data analysis of the latest project household immunisation survey in 2011 and compared the findings to reports of past surveys in the Core Group Polio Project (CGPP program area) and at the Uttar Pradesh state level (as measured by children’s receipt of DPT vaccinations). This was done to judge whether there is any evidence that routine immunisation services are being disrupted. They also modelled characteristics of survey respondents and respondents’ exposure to CGPP, communication activities against their children’s receipt of key vaccinations in order to identify determinants of routine immunisation coverage. Results from the study showed that religion influences immunisation coverage. Lakew, Bekele and Biadgilign (2015) carried out a study to identify factors associated with full immunisation coverage among children aged 12-23 months in Ethiopia. The study used the 2011 Ethiopian demographic and health survey data. The survey was cross-sectional by design and used a multistage cluster sampling procedure. A total of 1,927 mothers with children of 12-23 months of age were extracted from the children’s dataset. Mothers’ self-reported data and observations © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 44 of vaccination cards were used to determine vaccine coverage. It was found from the study that the coverage of vaccination differed by religion. Adokiya, Baguune and Ndago (2017) evaluated the immunisation coverage and its associated factors among children aged 12–23 months in Techiman Municipality, Ghana. A cross-sectional cluster survey was conducted among 600 children. Data was collected using a semi-structured questionnaire through face-to- face interviews. Descriptive statistics such as percentages, frequencies and cross- tabulations were performed using SPSS, with bivariate and multivariate logistic regression analysis conducted using Stata 12.1 version to estimate the odds ratio of not being fully immunised. Results showed that children from the Christian religion (AOR = 0.17, 95%CI = 0.06–0.50) were more likely to be fully immunised compared to children from Islam. Lastly, Asuman, Ackah and Enemark (2018) examined rural-urban inequalities in child immunisations in Ghana. Using data from the recent two waves of the Ghana Demographic and Health Survey, the authors examine the probability that a child between 12 and 59 months receives the required vaccinations and proceed to decompose the sources of inequalities in the probability of full immunisation between rural and urban areas. In terms of religion and vaccination, the study revealed that, compared to children of mothers with no or other religious backgrounds, children with Christian or Moslem mothers are more likely to be fully immunised. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 45 Theoretical Framework This section reviews the model used to study vaccination. This is the Health Care Services Utilisation Model propounded by Andersen in 1995.. Health Care Services Utilisation Model The Health Care Services Utilisation Model is a behavioural model which was propounded by Anderson in the 1960s (Andersen, 1995). It has been used widely to explain factors that promote or impede health care utilisation. The core postulation of this framework is that some persons have a higher tendency to utilize health service than others. The propensity to use the services are varied and influenced by three sets of individual characteristics: predisposing factors, enabling factors and need-for-care factors. Predisposing factors denote the demographic, social structure and health belief characteristics, the individual’s demographic characteristics that affect his/her decision not to use or to use a health care service. The characteristics may include the individual’s sex, age and marital status. Social structure consists of education, ethnicity and occupation and the health belief factors consist of values, attitudes of health care services providers and knowledge about health care. Enabling factors are the resources or means which are accessible to an individual to pursue healthcare services. Enabling factors are measured at the household level, thus, the availability of income and the community level, the availability and location of healthcare facilities in the community. Need-for-care factors refer to how individuals perceives their own general health and functional condition, as well as their familiarity with the signs and © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 46 symptoms of ill health, agony and concerns about their health (Andersen, 1995). The need-for-care factors are affected by the predisposing factors and the enabling factors of an individual. The health care services utilisation model embraces both structural and material factors which are barely taken into consideration in the social- psychological model (Hausamann-Muela, Ribera, & Nyamongo, 2003). Figure 1: Health Care Utilisation Model Source: Andersen, 1995 Predisposing Factors ➢ Age ➢ Marital status ➢ Education ➢ Occupation ➢ Ethnicity ➢ Values ➢ Attitudes ➢ Knowledge about health care Enabling Factors ➢ Availability of income ➢ Location Need for care Factors ➢ Chronic conditions ➢ Self-reported health status ➢ Cognitive impairment Health care Service Utilisation © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 47 Conceptual Framework The Health Care Service Utilisation Model by Andersen (1995) was adapted as the conceptual framework for this study. This is because the model best fit considering the objectives, the data availability and focus of the study. The model identifies the interactions among predisposing factors, enabling factors and the need-for-care factors in the utilisation of health care services. In order to precisely replicate the variables of interest of the study in the adapted model, the framework uses age, occupation, ethnicity, education, region and marital status. The enabling factors were wealth status and place of residence whilst the need-for-care factor was parity. Some variables were excluded because they were not captured in the GDHS data set for which the study was grounded. The independent variables for the study are predisposing factors which are age, occupation, ethnicity, education, region and marital status; enabling factors are wealth status and place of residence and the need-for-care factor is, parity. The dependent variable for the study is childhood vaccination. It is expected that the interaction of predisposing factors, enabling factions and need for care factor would influence vaccination of children under five years of age. From the framework, each of the three groups of variables individually affect childhood vaccination and collectively they have an influence the outcome variable. That is only the predisposing factors affect the outcome variable, only the enabling factors affect the outcome variable and the need for care only also affect the outcome variable. The combination of the predisposing, enabling and need for care factors also affect the outcome variable. © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 48 The strength of the model lies in the fact that it looks at the direct factors that lead to the utilisation of health care services. Specifically, it looks at the individual characteristics and the resources or means available to pursue health care services. The limitation of the model is that it is biased towards the utilisation of healthcare services and does not look at the satisfaction of clients in their utilisation of healthcare services. The second limitation is that it is silent about the healthcare systems and the social network and the influence of culture on the utilisation of health care service. Figure 2: Conceptual framework Source: Adapted from Andersen (1995). Childhood vaccines Enabling Factors ➢ Wealth ➢ Residence Need for care Factors ➢ Parity Predisposing Factors ➢ Age ➢ Marital status ➢ Education ➢ Occupation ➢ Ethnicity ➢ region © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 49 CHAPTER THREE METHODS OF DATA COLLECTION AND ANALYSIS Introduction This chapter discusses the study area, sources of data, methods of data collection, sampling procedure, data acquisition, definition and description of variables, data processing and management, data analysis as well as the data limitations. Study Area The Republic of Ghana, the study area, is located on the West African coast and it has a total land area of 238,533 square kilometers (GSS, 2013). It is bounded by three French-speaking countries (Burkina Faso on the north, Togo in the east and Cote d’Ivory on the west. The south of Ghana lies the 560-kilometers-long Gulf of Guinea. Ghana lies between latitudes 4º and 12ºN and longitudes 4ºW and 2ºE, and the Greenwich Meridian line passes through the sea point of Tema about 24 kilometers to Accra, the capital of the country. Ghana is a low-lying country with a few series of hills on the eastern border and Mountain Afaddjato, the maximum point of 883 metres above sea level which is west of Volta Region. The population of Ghana, according to the National Population Census conducted in 1960, 1970, 1984, 2000 and 2010 stood at 6,726,815; 8,559,313; 12,296,081; 18,912,079 and 24,658,823 respectively (GSS, 2013). The annual growth rate from 1960 to 1970 was 2.7 percent, 2.6 percent from 1970 to 1984, 2.7 percent from 1984 to 2000 and 2.5 percent from 2000 to 2010 accordingly. The population density based on the 2010 population and housing census was 103 © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 50 people per square kilometre, with Greater Accra having the highest of 1235.8 people per square kilometer and Northern, the lowest, with 35.2 people per square kilometer. Ghana as at 2000 had a sex ratio of 95 males per 100 females. The country has been divided into 10 regions. These are Western Region, Central Region, Greater Accra Region, Volta Region, Eastern Region, Ashanti Region, Brong Ahafo Region, Northern Region, Upper East Region and Upper West Region. With these 10 administrative regions, Ashanti Region has the highest population of 19.4%. Greater Accra has the second highest with 16.3% followed by Eastern Region with 10.7%. Northern Region has 10.1%, Western region has 9.6%, Brong Ahafo has 9.4%, Central Region has 8.9% and Volta Region has 8.6%. The Upper East Region and the Upper West Region has the least population of 4.2% and 2.8% respectively (GSS, 2013). Ghana has about 51 percent of its population living in urban centres and 49 percent in the rural areas. About 3,405,406, representing 13.8 percent of the total population of the country are below age 5 (Ghana Statistical Service, 2010) In relation to economic activities, Ghana has most of the population in the economically active group. About 42 percent of these economically active people are skilled in agriculture, 21 percent are service and sales workers and forestry workers whereas 9.5 percent are professionals, managers and technicians whiles 12.6% are into other occupation. (GSS, 2013). In terms of education, Ghana has about 26 percent of its population 6 years or older having no education (Never attended school, Nursery, and Kindergarten), with 56.3 percent in primary education (Primary, JSS/JHS, Middle). Secondary © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 51 education (SSS/SHS, Vocational/Technical/Commercial, Post Middle/Post- Secondary Certificate and Post-Secondary Diploma) represents 15.6 percent of the population 6 years and older. Higher education (Bachelor Degree and Post Graduate) denotes 2.1 percent of the population 6 years and older (GSS, 2013) The dominant ethnic groups in Ghana are Akan (47.5%), Mole Dagbani (16.6%), Ga-Adangbe (7.4%), Gruma (5.7%), Guan (3.7%), Grusi (2.5%) with the rest being others which consist of Mande, Hausa and other ethnic groups. With religious affiliation, the majority of Ghanaians (71.2%) are Christians (Catholic, Protestant, Pentecostal/Charismatic and other Christian) followed by Islam (17.6%), Traditionalist (5.2%), No Religion (5.3%) and 0.8% of the population with Other Religion (GSS, 2013). Research Design The GDHS uses a repeated cross-sectional research design. Cross-sectional research design helps to study the current behavior, practices, belief and attitudes of a specific group of respondents at a given point in time Creswell (2014). The cross-sectional research design was used because complete coverage of the population was impossible. It addresses the survey population in a short period of time and also produces comparable and equal valid results. This study employed a cross-sectional study design to assess childhood vaccination in Ghana. This study design was considered suitable because of the nature of the data, the objectives and the hypotheses guiding the study. The philosophical basis for the study was positivist, which asserts that knowledge is attainable and can be quantified. According to Creswell (2014), a respondent’s © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 52 belief and attitude would be a measure of his/her perception of the problem at hand while the respondent’s practices would reflect his/her behaviour in response to the problem at hand. It has an epistemological assumption since human behaviour can be studied through numeric and hard data aimed at measuring and describing the social phenomenon, and these attributes entirely coincide with the focus of this study. Source of Data The study used four datasets from the GDHS which were conducted in 1998, 2003, 2008 and 2014. This is because the first and second datasets of the GDHS (those conducted in 1988 and 1993) had a number of inconsistencies. For instance, DPT was not present in the 1988 and 1993. The GDHS is a nationwide survey that is carried out every five years since it began. The survey gathers information on fertility, family planning infant and child mortality, maternal and child health as well as nutrition. The GDHS focuses on child and maternal health and is designed to provide adequate data to monitor the population and health situation in Ghana. Typically, with the nationwide survey, the GDHS uses sample weights to regulate for effects of under-and-oversampling which have the potential to affect the generalizability of the result. The sample weights are alternation factors used to account for differences in the probability of selection and interview of respondents as a result of the design for the survey or as a result of chance (GSS, GHS & ICF Marco, 2015). © University of Cape Coast https://erl.ucc.edu.gh/jspui Digitized by Sam Jonah Library 53 The GDHS was carried out by the GSS with ICF Macro an International company, providing technical support for the survey through Measure DHS. GDHS is funded by United States Agency for International Development (USAID) and the Government of Ghana with support from other donor agencies such as UNICEF, the Global Fund, the International Labor Organization (ILO), the Danish International Devel