Women’s Empowerment a Determinant for Contraceptive use among women in Ethiopia: A secondary analysis of Ethiopian Demographic and
Health Survey from 2016
Name: Samira Dini Words: 10,124
Master’s degree in Global Health, 30 credits, Fall 2020
Department of Women’s and Children’s Health (IMCH), International Maternal and Child
Health
Abstract
Introduction Ethiopia has one of the largest populations in the world, an estimate of 114 million inhabitants. With more than 40% of the population below the age of 15 the country has to make further progress in meeting its family planning needs. The fertility rate has slowly declined, but the population continues to grow. Efforts to reduce gender disparities and
empower women have fallen short in many parts of the world. Evidence suggesting a link between women’s empowerment, health outcomes and health care service utilization.
Method A secondary analysis of the 2016 Ethiopian Demographic and Health Survey was conducted. The aim of this study was to determine the association between women’s empowerment, sociodemographic and reproductive factors and contraceptive use among married women and women living with partner aged 15-49 in Ethiopia. Logistic regression, bivariate, and descriptive analysis was conducted.
Results Decision-making role in regard to husband’s money was a strong predictor for contraceptive use. Women who alone or jointly made decision were more likely to use contraceptives. The state of wealth of women was a significant determinant for contraceptive use. Those with lower education were more likely to use contraceptives compared to those with higher education. Women who did not intend to have more children were more likely to use contraceptives.
Conclusion This study showed that contraceptive use is associated with women’s economic
decision-making age, and several sociodemographic and reproductive factors. Improving
women’s empowerment, through decision making power can improve contraceptive use and
therefore achieve better maternal health.
Table of Contents
ABSTRACT ... II LIST OF ABBREVIATIONS AND ACRONYMS: ... V
1. INTRODUCTION ... 1
1.1. The Programme of Action and the SDGs ... 1
1.2. The concept of sexual and reproductive health and rights ... 2
1.3. Unmet need for family planning ... 3
1.4. Risk factors for low prevalence of contraceptives ... 4
1.5. Empowerment and Contraceptive use ... 4
1.6. The Empowerment theory ... 6
1.7. The Ethiopian context ... 6
1.8. The Ethiopia governments commitments for family planning services ... 7
1.9. Rationale for the Study ... 8
1.10. Aim of the study ... 8
1.11. Research question ... 8
2. METHOD ... 9
2.1. Study design ... 9
2.2. Study setting ... 9
2.3. Study population ... 10
2.4. Sample size ... 11
2.5. Data collection ... 11
2.6. Variables measurements ... 12
2.6.1. Outcome variable ... 12
2.6.2. Exposure variables ... 13
2.7. Missing data and bias ... 14
2.7.1. Weighting of data ... 14
2.7.2. Imputation ... 15
2.8. Statistical analysis ... 15
2.9. Ethical consideration ... 16
3. RESULTS ... 16
3.1. Socio-demographic factors ... 16
3.2. Reproductive factors ... 17
3.3. Association between contraceptive use and sociodemographic and reproductive factors 17 3.4. Determinants for contraceptive use ... 19
4. DISCUSSION ... 21
4.1. Summary of main findings ... 22
4.2. Decision-making role and contraceptive use ... 22
4.3. Sociodemographic factors and contraceptive use ... 24
4.4. Reproductive factors and contraceptive use ... 26
4.5. Strength and limitation ... 27
5. RECOMMENDATION ... 28
5.1.Future research ... 28
5.1. Policy intervention ... 29
6. CONCLUSION ... 30
7. ACKNOWLEDGEMENTS ... 30
REFERENCES ... 31
APPENDIX ... 35
List of Abbreviations
CAPI – The Computer-assisted personal interviewing
CPR - Contraceptive prevalence rate
DHS - Demographic and Health Survey
EAs - Enumeration areas
EDHS - Ethiopia Demographic and Health Survey
HIC - High Income Countries
ICPD - International Conference on Population and Development
IRB - ICF institutional Review Board
LMIC- Low- and Middle-Income Countries
MDG - Millennium Development Goals
NRERC - National Research Ethic Review Committee
PHC - Population and Housing Census
SDG - Sustainable Development Goals
SRHR - Sexual Reproductive Health and Rights
SRH - Sexual reproductive health
STIs - Sexually transmitted infectious
UN - United Nations
1 1. INTRODUCTION
1.1. The Programme of Action and the SDGs
The International Conference on Population and Development (ICPD) in 1994 set out an agenda to develop inclusive, equitable and sustainable global development. This agenda has contributed to advances in equality and empowerment for women, global health and life expectancy, and education for girls. The ICPD conference gave an action (Programme of Action) towards these broad areas that involved working interdisciplinary and
recognizing that reproductive rights is fundamental to development and population concerns.
It was established that reproductive rights are human rights and a part of sexual and reproductive health (SRHR). Women’s full participation at all levels in the social,
economic and political decisions in their communities became a priority in the Programme of Action. While advances have been made in expanding access to reproductive health care and lowering birth rates, including education and economic status of women in the past, much remains to be accomplished. Areas as increased use of contraception, decreased maternal mortality, educational programs are vital in meeting the Programme of Action (1).
In 2000 the General Assembly adopted the Millennium Development Goals (MDGs) which has contributed to improvements in poverty, health, education and gender equality. The SRHR of women were not included in the agenda, although a consensus that women’s SRHR played a central role in societal development had been established, particularly in the ICPD) (2). Agenda 2030 for Sustainable Development (SDG) call on countries by 2030 to ensure access to SRHR health-care services including family planning among others. The SDG recognize SRHR as a cornerstone of reaching sustainable development globally. The aim to leave no one behind in the SDGs stems from inequalities in access to resources and ability to exercise basic rights, including SRHR. SDG 3.7 calls countries to ensure universal access to sexual and reproductive health-care services, including family planning, information and education, and the integration of reproductive health into national strategies and programmes.
SDG 5.6 calls to ensure universal access to sexual and reproductive health and reproductive
rights in accordance with the Programme of Action of the International Conference on
Population and Development and the Beijing Platform for Action (2). Other goals that relate
to SRHR are 4, 6, 10 and 16.
2
The agenda 2030 has contributed to increased access to affordable modern contraceptive methods, but still many women report an unmet need for contraceptives (3). Among the 1.9 billion women of reproductive age (15-49 years), 1.1 billion had a need for family planning in 2019. They were either currently using contraceptives or had an unmet need for family
planning. Approximately of these women 842 million were users of modern methods of contraception and 80 million are users of traditional methods of contraception. In 2019, the number of women of reproductive age having unmet need for any contraceptive method were 190 million, an increase from 156 million in 2000. The proportion of women with unmet need for contraceptive methods in 2019 was 10%, and women who have a need for family planning has increased from 74% to 76% since 2000 (4).
1.2. The concept of sexual and reproductive health and rights
The United Nations conference of the 1990s agreed on the following concepts and definitions of SRHR. Reproductive health is a state of complete physical, mental and social well-being in relation to all reproductive systems. Sexual health relates to the ability to have safe and satisfying sex lives. Gender relation should be equal, responsible and respectful. Behaviors to counter sexually transmitted infectious (STIs) are related to sexual health. It aims at
enhancing life and personal relations, and sexual health services should cover more than counselling and care related to reproduction and sexually transmitted diseases. Reproductive rights mean for couples and individuals to decide freely and responsibly the number, spacing and timing of their children, and to have information. To make decisions concerning
reproduction without being discriminated, pressured or enduring violence. Sexual rights include the right of women to have control over and decide freely and responsibly on matters related to their sexuality, including sexual and reproductive health, free of discrimination, pressure and violence (5)
The reproductive choices of women greatly affect theirs and their children’s life prospects.
The age of mother, spacing of births, care during pregnancy and after delivery are important
for infant survival and development. Children’s wellbeing, in the first year, is dependent on
their mother’s health. Children who are first born to young mothers, born in to larger families,
and children who are unwanted are at higher risk of poorer health and mortality compared to
other children. Unwanted births are at greater risk of poor health and mortality, due to limited
resources, negligence, and discrimination. Parents may discriminate in the allocation of food,
parental time, attention, and preventative and therapeutic health care (6).
3 1.3. Unmet need for family planning
Women with unmet need for contraceptives are those who are sexually active but not using any method of contraception, and report not wanting any children or wanting to delay the next child. They have a desire to stop or delay childbearing, while not using any contraceptive method. This includes women whose pregnancies are unwanted or mistimed at the time of conception, and postpartum women who are not using family planning and whose last birth was unwanted or mistimed. Unmet need addresses women’s reproductive intentions and contraceptive behavior (7). Promoting modern contraceptive use and addressing unmet need are significant to improve the health and reproductive wellbeing of women (8). Despite progress, around 214 million women in low- and middle-income countries (LMIC) are not using a contraceptive method. Thus, less are able to decide if, how many, and when to have children. Between 2015 and 2019, almost half of all pregnancies were unintended.
Unintended pregnancies often end up in abortion, and most of abortion frequently occur in LMIC. Women living in LMIC are almost three times as likely to have unintended
pregnancies than those in high- income countries (HIC) (9).
The fertility rate in the world is decreasing, which can be linked to more access to contraceptives and a higher contraceptive prevalence rate (CPR), suggesting
increased efforts in achieving SDGs and the ICPD (10). Although, the population of women of reproductive age is growing through 2030 in countries with larger gaps in meeting the need for family planning. Many countries with low levels of satisfied demand for family planning with modern contraceptive methods are experiencing rapid growth in the number of women in reproductive age, creating additional challenges in expanding family planning services and meeting demand. The majority of these countries are located in Sub-Saharan Africa (4).
Significant disparities remain in the need for family planning satisfied with modern
contraceptive methods accords the world. In Sub-Saharan Africa more than half (55 per cent)
of the need for modern contraceptive methods are met. Variation exists across countries, and
within the same region in the number of women in reproductive age using modern methods of
contraceptive. Estwatini and Nambia (both 52 per cent) have the highest use of modern
contraceptive methods, while South Sudan (4 per cent) and Chad (6 per cent) had the lowest
in 2019 in Africa. More than 1 in every 10 women in reproductive age use a traditional
contraceptive method, the highest number of countries located in Europa (8 countries) (4).
4
1.4. Risk factors for low prevalence of contraceptives
The use of modern contraceptives varies by age and marital status. The prevalence of modern contraceptive methods is higher among women aged 30 to 39, in the middle of the
reproductive age (1). Although, older women are less likely to report unmet need for
contraception compared to younger (11). Contraceptive prevalence is low among adolescent women (15 to 19 years) and women above 40 years, although married adolescent women’s use of modern contraceptive methods has increased in Africa and Asia. Among unmarried women, the proportion of women using modern contraceptive methods varies across countries, from high levels in Malawi and Uganda and lower in Indonesia (4).
Domestic violence is negatively associated with use of contraception. Women whom accept or justify for wife beating are less likely to use contraception (11). Distance to health facility is related with contraception use. Women who live further away from health facility are less likely to be using a modern method of contraception (12). Communities with lower mean age at marriage report lower percentage of contraceptive use (13).The desire to have more
children influence contraceptive use. Women who express desire to have children more children are more likely to bear more children and have bigger family. The main reason reported by women to not use contraceptives are the desire to have more children, fear of infertility and fear of side effects (14).
Education is positively associated with using contraception and negatively associated with an unmet need for contraceptive. Women with education (primary, secondary or tertiary) are more likely to use contraceptive, compared with women with no education (11,15,16). Men’s education is also positively associated with using contraception. Higher levels of male partner education are associated with contraceptive use. There is an association between women’s status, contraception usage and unmet need. The number of women’s households’ decisions are associated with an increase in use of contraception. Women who have worked in the past 12 months reported a higher prevalence of contraception use (11). Also, women living in urban areas, wealthy households, working outside the home and with increased exposure to information on HIV/AIDS have a higher rate of contraception use (15).
1.5. Empowerment and Contraceptive use
Reducing gender inequality and empowering all women is in the global agenda and efforts of
improvements are made. However, it is of importance to first understand how inequalities are
5
exhibited to make significant changes. In the past decades, efforts to reduce gender disparities and empower women have fallen short in many parts of the world (17). The growing body of evidence suggesting a link between women’s empowerment, health outcomes and health care service utilization (14,18–20).
Women’s empowerment is multidimensional, thus is challenging to measure. Studies have selected single-aspects or dimensions for empowerment on contraceptive use (19). An indicator that has been observed and linked to women’s empowerment is decision-making role (8, 9, 11). Women that take independent and joint decision with partner are more
empowered, compared to women whom others make decisions for (19). Approximately 57 % of women 15 to 49 married or in union in the world made decision about sexual relations and the use of contraceptives and health services in 2018 (21). Thus, indicating more need of efforts to increase women’s decision making-role. Higher decision-making role is associated with contraceptive use, women who make joint decisions about fertility issues are 3.7 times more likely to use modern contraceptives than those who do not make decisions (14). An increase in decision-making autonomy is also linked to women’s participation in labour force, reduction in abuse and violence and improved knowledge level (18).
Although, an association between higher levels of empowerment and positive reproductive health outcomes has been noted, social norms inhibit women from using contraceptives.
Women are defined and judged in relation to social and gender norms, in particular the ability to become mothers. A women’s pregnancy can increase her status and contribute to higher level of empowerment during her pregnancy period. The social norms and gender ideologies can pressure women to have more children as the ability to have children is a measure of
“womanhood”. The ability to have children can also increase support and the love from partner and relatives. Women are also able to obtain better health care and more likely to acquire power within household, thus the pregnancy period becomes an opportunity to exercise different elements of empowerment (20).
Although, women report that their livelihood greatly improve after initiating contraceptives.
The benefits include more spare time, energy and social engagement. Contraceptives helps
women postpone unwanted pregnancies and child births. In addition, contraceptive use can
generate income activity for family and create financial autonomy for women (16). Women
who report more consistent use of contraceptives have higher engagement of working in labor
force and receive payment (22).
6 1.6. The Empowerment theory
Empowerment suggest a different approach for developing interventions and creating social change. It directs attention towards social problems that exist due to unequal distribution and access to resources (23). Empowerment is primarily linked with changing power relations in favor of those who previously exercised little power over their lives. In this context power is control over resources (physical, human, intellectual, financial, and the self), and control over ideology (attitudes, belief and values). Empowerment is not something that can be done for or to anyone else, rather it is the process that equips people to make decisions for themselves and increases capacity to act and have influence. The concept of women’s empowerment emerged during 1980s by feminists. The new concept rejected the more top-down, paternalistic
community development approach, which ignored gender and the subordination of women in the liberation movement (24). The subordination of women and the hegemonic role in most ideologies influenced the formation of women’s empowerment. The interplay of these discourses led to a more political and transformatory idea that challenged patriarchy and incorporated gender perspective when mediating structures of class, race and ethnicity. In practice feminist introduced the absent gender dimension as fundamental category of analysis in practice of social change and development. The concept of empowerment has a long history and can be traced early back. It is embedded in in many historical struggles for social justice for instance, the Veerashaiva movement against caste and gender oppression in India, the liberation theology movement, the black power movement, feminism, popular education and other movements for equality, and democratic reforms of social change and development (24).
1.7. The Ethiopian context
Ethiopia has one of the largest populations in the world, an estimate of 114 million inhabitants. With more than 40% of the population below the age of 15 the country has to make further progress in meeting its family planning needs (25). The fertility rate has slowly declined from 5.5 in 2000 to 4.6 in 2016, but the population continues to grow. Women in rural areas have an average of 5.2 children, while women in urban areas have 2.3 children (26). Although, demand for family planning remains low in several parts of Africa, Ethiopia is among the top 10 countries with the largest increase in the proportion of women who have their need for family planning satisfied with modern methods. The number of women
demands satisfied with modern methods increased from 20% in 2000 to 63% in 2019 (4).
7
The rising age of marriage and women who remain single has reduced the fertility rate in Ethiopia. The use of modern contraceptive methods among married women has increased significantly from 6% in 2000 to 36% in 2018 (25). Despite the increase in CPR in Ethiopia since 2000 there are disparities across areas and regions (16). The disparities in CPR between urban, rural and region to region has been noted. In the capital Addis Ababa 50% among married women 15 to 49 use modern contraceptives compared to 12% in Affar region and 1%
in Somali region (26). The use of modern contraceptive among non-married women between 15 to 49 was higher 55% and 3% used traditional methods in 2016. Improvements in efforts to accelerate the reproductive health of women is necessary in Ethiopia. The country is still suffering from preventable morbidity and mortality of mothers and children, with one of the highest rates in the world. (27,28).
1.8. The Ethiopia governments commitments for family planning services The government of Ethiopia has made commitments to improve health status of women, in particular adolescent women. The country has prepared a national adolescent and youth health strategy, which is in line with the global strategy for Women’s Children’s and Adolescent’s Health (2016-2030). The commitments will be to improve the access to contraceptives, increasing CPR among married youth aged 15-24 years, increasing CPR among married women from 42% in 2014 to 55%, reduce its total fertility rate to 3.0, and lastly reach 6.2 million additional women and adolescent girls utilizing family planning services by 2020. The following activates has been undertaken in 2017-2018 to improve the reproductive health among Ethiopian adolescents and youth; expanded availability of high quality and
reproductive health services and information for adolescents and youth, increasing the availability and accessibility of contraceptives for women living in rural areas, reducing social- cultural and financial barriers, and training for health professionals to provide youth friendly services (29).
Although, less investment is made on family planning services compared to maternal and
newborn care. Investing in the need for modern contraception is more cost effective than
focusing in maternal and newborn health care among adolescent in Ethiopia. However,
maternal and newborn health care is essential to improving health outcomes. But meeting the
needs for modern contraception would lower pregnancy related cost. The cost of preventing
pregnancy through modern contraceptives is lower than the cost of care for an unintended
pregnancy, each additional dollar spent on contraceptive services for adolescents would
8
reduce the cost of maternal and newborn health care for adolescents in Ethiopia by 2.34 dollars (29).
1.9. Rationale for the Study
This research is necessary to better understand determinants that influence women’s contraceptive use, more specifically how women’s empowerment factors influence
contraceptive use. Also, sociodemographic and reproductive factors that influence women’s contraceptive utilization have previously been important (14,18,19,30). A better
understanding of this determinants could accelerate the Ethiopian governments efforts in improving the overall health status of women. Despite the high fertility and high unmet need for contraceptives, the country CPR is very low (36%) (25). This study provides an
opportunity to study empowerment, economic, socio-cultural, and interpersonal factors that influence the reproductive health of women. In order to further guide policies for
governmental and organizational entities within the country this type of study is needed.
Although much literature has examined the relationship between women’s decision-making role and reproductive health outcomes, limited research has been conducted in Ethiopia. Also, these results could be contextualized to similar contexts, as contraceptive use is generally low in Sub-Saharan Africa (18).
1.10. Aim of the study
- To determine the association between women’s empowerment factors (decision- making role) and contraceptive use among married women and women living with partner in Ethiopia.
- To determine the association between sociodemographic and reproductive factors and contraceptive use among married women and women living with partner in Ethiopia.
1.11. Research question
Is there an association between women’s empowerment, sociodemographic and reproductive
factors and contraceptive use among married women and women living with partner aged 15-
49 in Ethiopia?
9 2. METHOD
2.1. Study design
This study was conducted using the 2016 Ethiopia Demographic and Health Survey (EDHS) (26). The 2016 EDHS is a descriptive cross-sectional study design. The intention of this study was to determine the association between women’s empowerment, sociodemographic and reproductive factors and contraceptive use among married women and women living with partner aged 15-49 in Ethiopia. Logistic regression analysis was conducted to determine the association between these variables. Additionally, bivariate analysis and descriptive analysis was conducted to find correlation as well as get an overview of the data.
2.2. Study setting
The geographical area was Sub-Saharan Africa and the country Ethiopia. The country of
Ethiopia is located in the eastern part of Africa and its neighboring countries are Eritrea,
Djibouti, Somalia, Kenya, and Sudan (Figure 1). Ethiopia is divided in nine geographical
regions (Tigray, Afar, Amhara, Oromia, Somali, Benishangul-Gumuz, Southern Nations
Nationalities and People Region, Gamnella and Harari) and two administrative cities (Addis
Ababa and Dire Dawa). It’s the second most populous country in Africa with an estimate of
114 million people. The country is considered to be a low-income country but has the lowest
level of income inequality in Africa and one of the lowest in the world. More than 80% of the
population lives in rural areas. Infant, child and maternal mortality has fallen over the last
decades and the total fertility rate has declined slowly. The rising age of marriage and the
increasing proportion of women remaining single is contributing to fertility decline, but still
the population continues to grow. Ethiopia remains one of the poorest countries in the world,
due to rapid population growth and food insecurity (25).
10
Figure 1. Ethiopia map. (25).
2.3. Study population
Participants of the study were all women age 15-49. A total of 15,683 women were selected and consented to be interviewed about various of health topics. Women who were currently married or living with partner were included in this study. Formerly married including widowed, divorced, separated women and women who have lived with partner but are not now were excluded (5,859 participants). Thus, leaving a final sample population of 9,824.
Figure 2 displays the participants who are included in this study.
Figure 2. Study participants
Participants of the Ethiopian 2016 DHS women’s
questionnaire:
(N=15,683)
Excluded: Participants who were never married, widowed, divorced or not living with partner (N=5,859)
Included: Women who
were married, or living
with partner (N=9,824)
11 2.4. Sample size
A total of 18,008 households were selected for the 2016 EDHS, 16,659 were successfully interviewed, yielding a response rate of 98%. Out of the interviewed households, 15,683 women (95% response rate) and 12,688 men (86% response rate) were individually interviewed. The sampling frame used for the 2016 EDHS is the Ethiopian Population and Housing Census (PHC), from 2007 by CSA. The PHC is a list of 84,915 geographic areas, called enumeration areas (EAs). Each EA covers approximately 181 households. The purpose of a sampling frame is to collect a representative sample of the population in a country or setting. The sampling frame includes information about EA location, type of residence (urban or rural), and number or residential households (26).
The 2016 EDHS was designed to provide estimates for the whole country, urban and rural areas, and for each region and administrative city. The sample was stratified at two stages.
Each region was stratified into urban and rural, a total of 21 sampling strata. Samples of EAs were selected independently in each stratum in two stages. Implicit stratification and
proportional division through the sampling frame (PHC) within each sampling stratum (26).
In the first stage of EA sampling, a total of 645 EAs (202 urban and 443 rural) were selected with probability proportional to EA size and with independent sampling stratum in each EA.
From September to December 2015 a household investigation was carried out in all of the selected EAs. The founding’s from the households served as a basis for the selection of the households in the second stage. As the selected EA were overwhelmingly large (more than 300 households), each large EA was segmented. With probability proportion one segment was selected, contributing to a household listing. In the second stage, a number of 28 households per cluster were selected with an equal probability systematic selection from the household listing. All women age 15-49 and all men age 15-59 who were permanent residents of the households or visitors who stayed in the household the night before the survey were eligible to participate (26).
2.5. Data collection
The data collection took place took place from January 18, 2016, to June 27, 2016. It was
carried out by 33 field teams, each team consisted of one supervisor, one editor, three female
interviewers, one male interviewer, two biomarker technicians, and one driver. In addition, 28
quality controllers were dispatched to support and monitor the data collection.
12
The 2016 EDHS consisted of five questionnaires: The Household Questionnaire, the Woman’s Questionnaire, the Man’s Questionnaire, the Biomarker Questionnaire, and the Health Facility Questionnaire. The questionnaires were based on the DHS Program’s standard Demographic and Health Survey questionnaires, however were adapted to reflect the
population and health needs of Ethiopia. In order to do so, inputs were gathered from government ministries and agencies, nongovernmental organizations, and international donors. The questionnaires were finalized in English, then translated into Amarigna, Tigrigna and Oromiffa. The Man’s Questionnaire collected less information, as it did not contain detailed reproductive history, maternal and child health, or questions on domestic violence (26).
A total of 294 people was recruited and trained as supervisors, field editors, secondary editors, and reserve interviewers. The interviewers used tablet computers to record responses of participants. The computers were equipped with Bluetooth technology, thus enabling transfer of files from interviewers to editors (26). The Computer-assisted personal
interviewing (CAPI) is an interview technique in which an electronic device is used to answer questions (31). All data files from the 2016 EDHS was transferred to the CSA central office in Addis Ababa, where they were stored on a protected computer (26).
2.6. Variables measurements
The IBM SPSS Statistical Software was used to conduct the analysis in this study. The
variables of this study were; age, type of place of residence, education level, religion, literacy, wealth index, use of internet, own a mobile, knowledge of ovulation, knowledge of method, used anything to avoid pregnancy, unmet need for contraceptives, contraceptive use and intention, person who decides health care, person who decides household purchases, person who decides visit to family, person who decides husband’s, and current contraceptive use
2.6.1. Outcome variable
The outcome variable of this study was a categorical variable representing current contraceptive use (at the time of data collection). This variable allowed the percentage
distribution of any contraceptive method (modern or traditional) to be collected. The category for this variable consisted of: no method, traditional method, and modern method. It was possible to categorize this variable in to two categories; “No” (no method) and “Yes”
(traditional or modern method).
13 2.6.2. Exposure variables
The exposure variable variables were categorized in to three main categories (decision- making role factors, socio-demographic factors, and reproductive factors), each containing a number of variables. These variables were then categorized into different groups.
Sociodemographic factors
• Respondent’s ages. Categorized into three groups; 15-24 as youth, 25-34 young adults, and 35-49 adults.
• Type of place of residence. Categorized as; urban and rural.
• Education level. Categorized as: no education, primary, secondary and higher.
• Religion. Categorized as: Christian, Muslim, Traditional belief and other.
• Literacy. Categorized as: cannot read and can read.
• Wealth index. Categorized as: poor, middle and rich.
• Use of internet. Categorized as: yes and no.
• Own a mobile. Categorized as: yes and no.
• Own a house. Categorized as: yes and no.
Respondent age categorization was made based on United Nations (UN) definition of “youth”
as those aged 15-24 (32). Participants aged 25-34 were categorized as “young adults” and ages 35-49 as “adults”. Education was categorized into no education, primary, secondary and higher. Wealth index was coded by DHS as: poorest, poorer, middle, richer, and richest. This was re-coded as: poor, middle and rich. For religion, Orthodox Christian, Protestant and Catholic were re-coded as Christian. Literacy was coded by DHS as: cannot read, able to read parts, able to read, no card, blind. No participants answered no card. The other values were re- coded as: cannot read (cannot read, blind), and can read (able to read parts, able to read). Use of internet, own a mobile and own a house were coded as: yes and no.
Decision-making role factors
•
Person who decides health care. Categorized as: husband, partner or someone else and alone, with husband or partner.
• Person who decides household purchases. Categorized as: husband, partner or someone else and alone, with husband or partner.
• Person who decides visit to family. Categorized as: husband, partner or someone else and alone, with husband or partner.
• Person who decides what to do with husband’s money. Categorized as: husband, partner or someone else and alone, with husband or partner.
There were four variables measuring women’s empowerment: decision making in health care,
major household purchases, visits to family and relatives, and husband’s earnings. DHS coded
the responses as: alone, with partner, husband or partner alone, someone else and other. For
14
analysis, respondents were grouped in two categories, those who made decisions alone or jointly – indicating a higher level of empowerment, and those whom husbands or another person made decisions.
Reproductive factors
• Knowledge of ovulation. Categorized as: knowledge, wrong knowledge and no knowledge.
• Knowledge of method. Categorized as: no method and knowledge.
• Has had an abortion Categorized as: yes and no.
• Used anything to avoid pregnancy. Categorized as: yes and no.
• Unmet need for contraceptives. Categorized as: unmet need and no unmet need.
• Contraceptive use and intention. Categorized as: using method, non-user intends to use and does not intend to use.
• Age at first sex. Categorized as: legal age and illegal age.
• Fertility preference. Categorized as: no more children, have more children and undecided.
Knowledge of ovulation categorization was based on 1177 Vardguiden information about ovulation period (33). The original coding was: during period, after period, middle of cycle, before period, at any time, other and don’t know. These were re-coded as: wrong knowledge (during period, after period), right knowledge (middle of cycle) and no knowledge (other and don’t know). Knowledge of method was re-coded as: no knowledge and knowledge
(traditional or modern method). Unmet need for contraceptives was re-coded: unmet need (never had sex, unmet need for spacing, unmet need for limiting, spacing failure, limiting failure, menopausal, not married) and no unmet need (using for spacing, using for limiting, no unmet need). Contraceptive use and intention were re-coded as: using method (modern or traditional), non-user intends to use and does not intend to use. Age at first sex was coded as legal age and illegal age at first intercourse. Legal ages at first intercourse were respondents aged 18 and upward, and illegal age as those who were under 18. Based on the United Nations Convention on the Rights of the Child everyone under 18 years is a child (34).
Fertility preference was re-coded as: no more children (no more, sterilized, infecund) have more children and undecided.
2.7. Missing data and bias 2.7.1. Weighting of data
After the selection of variables, the data was weighted. Weighting adjustments are used to
compensate for noncoverage and total nonresponse. Respondent are assigned greater weight
15
in the analysis to represent non-respondents, which occurs due to refusals to participate in the survey, noncontact among other reasons. When some respondents are not included in the sampling frame the missing respondents go unrepresented, then compensation is made by weighting adjustments. The main objective of weighting is to avoid biases by making each respondent representative of the target population. In appendix 1, a table demonstrating weighted data versus unweighted data is presented to compare the difference.
2.7.2. Imputation
Imputation method is assigning values for missing responses, it enables relevant records to be retained in analysis. The purpose of imputation is to reduce and eliminate bias. Missing data arising from nonresponse can lead to biases in survey if they are not dealt with (35).
Incomplete and inconsistent data is common in DHS, due to large surveys and collection of retrospective data. Thus, surveys of this kind are prone to poor reporting (36). If the
nonresponse rate is low, most likely the amount of bias in univariate analysis will be small (35). However, since multivariable analysis was performed the nonresponse for several variables may had affected the results. First, an analysis for missing values was performed to obtain an overall summary of the missing values. A pattern of more missing responses of women from rural areas and women aged 15-24 was discovered. Thus, imputation was appropriate and necessary to adjust for the nonresponses. Imputation was assigned for 5 variables with 12.8% to 0.2% missing data to compensate for missing responses. The variables with missing data were; literacy (12.8%), education level (10.3%), person who decides what to do with husbands’ money (0.6%), age at first sex (0.6%), and unmet need (0.2%). Variables with <51% missing data were removed, due to substantial proportion of missing responses.
2.8. Statistical analysis
For this thesis, data was obtained from DHS after writing a motivation letter. IBM SPSS
Statistics software was used to perform all statistical analysis. A null hypothesis was formed
which claimed that women’s empowerment, sociodemographic factors and reproductive
factors does not interfere with the usage of contraceptive methods. This study examined the
alternative hypothesis, that women’s decision making-role, sociodemographic factors and
reproductive factors does influence the usage of contraceptive methods, in particular that
women who have a high decision-making role tend to use contraceptives.
16
Firstly, descriptive statistics was performed and a frequency table to showcase the difference between weighted and unweighted data between the sample population and Ethiopia’s population. A weight sample variable was divided by 1000000 computing a new variable named weight, this variable was used to weight the sample. Then the researcher employed both descriptive and inferential statistics to represent the results of the data. For
sociodemographic factors, reproductive factors, and decision-making factors frequency (percentages) for all variables were made. A chi-square test was carried out for all exposure variables to determine if there was a correlation with contraceptive use. Next univariate logistic regression was performed with all exposure variables against the outcome variable.
For the multivariate analysis all exposure variables with a p-value of <0.05 during the univariate analysis were included in the final analysis. The significance level was set at p<0.05 with confidence interval set at 95%.
2.9. Ethical consideration
The DHS program ensures privacy and protection to participants by acquiring ethical approval by the ICF institutional Review Board (IRB). Furthermore, the DHS program has acquired ethical approval for data collection by the National Research Ethic Review
Committee (NRERC) in Ethiopia. DHS program is consistent with the standard of ensuring the protection of study participants privacy. Before each interview, an informed consent is read to respondents, who can accept or decline participation. A parent or guardian must provide consent for a child or adolescent. ICF ensures that the survey complies with the U.S.
Department of Health and Human Services regulation of respect for humans (37). Approval to use the EDHS data for this study was obtained from DHS.
3. RESULTS
This section focuses on analyzation conducted with the 2016 EDHS to support the aims and research question earlier presented.
3.1. Socio-demographic factors
The ages of respondents were not evenly distributed as a large proportion (41.6%) of the respondents were young adults and were in the age range of 25-34 years. The result shows that (24.9%) were youth between ages 15-24 and (33.5%) were adults in the age range 35-49.
The result further shows that the majority (58.2%) of the respondents had no education,
17
(9.1%) secondary education, with no respondent having a higher level of education.
Furthermore, most (54.9%) of the respondents were Christian, (76.3%) were illiterate, (95.7%) of the respondent did not use internet, (44.8%) were poor, (13.8%) were of middle class and (41.3%) were rich. This result also indicates that most (67.5%) of the respondent were not working, (70.1%) owned a mobile and (63.5%) owned a house alone, jointly or both.
3.2. Reproductive factors
The majority of the respondents had the wrong knowledge of ovulation (59%). The results also show that almost all participants (95.7%) had knowledge of contraceptive method (traditional or modern), while (50.2%) had used anything to avoid pregnancy, (89.4%) had not had an abortion, (81.1%) had no unmet need for contraceptives, and (43.2%) did not intend to use contraceptives.
3.3. Association between contraceptive use and sociodemographic and reproductive factors
Table 1 depicts the association between contraceptive method used and findings from the crosstabulation using the chi-square test. The analysis shows a statistically significant association (P<0.05) between all socio-demographic factors; age, type of place of residence, education level, religion, literacy, wealth index, use of internet own a mobile and own a house. For reproductive factors; knowledge of ovulation, knowledge of method, used anything to avoid pregnancy, unmet need for contraceptives and contraceptive use and intention were statistically significantly associated with contraceptive use. Lastly, decision- making role factors; health care, household purchases, visit to family, and husband’s money were all statistically associated with contraceptive use.
Table 1
illustrates the Bivariate analysis of the association between sociodemographic factors, reproductive factors, decision-making role and contraceptive use among married women and women living with partner aged 15-49 in Ethiopia (total n=9824)Variables Total sample
N=9.824 N= 2,980 (30.3%)
Contraceptive use Yes
N= 6,844 (69.7%) Contraceptive use No
P-Value
Socio-demographic factors Age
15-24 Youth 25-34 Young adults 35-49 Adults
2447 (24.9) 4090 (41.6) 3287 (33.5)
762 (31.1) 1366 (33.4) 852 (25.9)
1685 (68.9) 2724 (66.6) 2435 (74.1)
P<0.01*
Type of place of residence Urban
Rural 2491 (25.4)
7333 (74.6) 1128 (45.3)
1852 (25.3) 1363 (54.7) 5481 (74.7)
P<0.01*
18
Education level No education Primary Secondary
5720 (58.2) 4318 (32.7) 890 (9.1)
1278 (22.4) 1304 (40.6) 398 (44.6)
4440 (77.6) 1909 (59.4) 495 (55.4)
P<0.01*
Husbands education level No education
Primary Secondary Higher
4524 (46.1) 3054 (31.1) 1226 (12.5) 1020 (10.4)
922 (20.4) 1141 (37.4) 494 (40.3) 423 (41.5)
3602 (79.6) 1913 (62.6) 732 (59.7) 597 (58.5)
P<0.01*
Religion Christian Muslim
Traditional belief Other
5391 (54.9) 4318 (44) 60 (.6) 55 (.6)
2228 (41.3) 720 (16.7) 22 (36.7) 10 (18.2)
3163 (58.7) 3598 (83.3) 38 (63.3) 45 (81.8)
P<0.01*
Literacy Cannot read
Can read 7499 (76.3)
2325 (23.7) 1880 (25.1)
1100 (47.3) 5619 (74.9) 1225 (52.7)
P<0.01*
Wealth index Poor
Middle Rich
4406 (44.8%) 1359 (13.8%) 4059 (41.3%)
714 (16.2) 467 (34.4) 1799 (44.3)
3692 (83.8) 892 (65.6) 2260 (55.7)
P<0.01*
Use of internet Never
Yes 4406 (44.8)
1359 (13.8) 4059 (41.3)
2767 (29.4)
213 (50) 6631 (70.6)
213 (50)
P<0.01*
Own mobile No
Yes 6636 (67.5)
3188 (32.5) 1759 (25.5)
1221 (41.5) 5126 (74.5) 1718 (58.5)
P<0.01*
Own a house No
Alone, jointly or both 6885 (70.1)
2939 (29.9) 1279 (35.7)
1701 (27.3) 2306 (64.3) 4538 (72.7)
P<0.01*
Reproductive factors Knowledge of ovulation
Knowledge Wrong knowledge No knowledge
2231 (22.7) 5795 (59) 1798 (18.3)
906 (40.6) 1664 (28.7) 410 (22.8)
1325 (59.4) 4131 (71.3) 1388 (77.2)
P<0.01*
Knowledge of method No method
Knowledge 420 (4.3)
9404 (95.7) 0 (0)
2980 (31.7) 420 (100)
6424 (68.3)
P<0.01*
Used anything to avoid pregnancy No
Yes 4895 (49.8)
4929 (50.2) 0
2980 (60.5) 4895 (100)
1949 (39.5)
P<0.01*
Has had an abortion No
Yes 8778 (89.4)
1046 (10.6) 2701 (30.8)
279 (26.7) 6077 (69.2)
767 (73.3)
P<0.05*
Unmet need for contraceptives Unmet need
No unmet need 1855 (18.9)
7969 (81.1) 1855 (100)
4989 (62.6) 0 (0)
2980 (37.4)
P<0.01*
Contraceptive use and intention Using method
Non-user intends to use Does not intend to use
2980 (30.3) 2597 (26.4) 4247 (43.2)
2980 (100) 0 (0) 0 (0)
0 (0) 2597 (100) 4247 (100)
P<0.01*
Decision making role Person who decides health care
Husband, partner or someone else Alone, with husband or partner
7993 (81.4) 1831 (18.6)
2583 (32.3) 397 (21.7)
1434 (78.3) 5410 (67.7)
P<0.01*
Person who decides household purchases
Husband, partner or someone else Alone, with husband or partner
8189 (83.4) 1635 (16.6)
366 (22.4)
2614 (31.9) 1269 (77.6) 5575 (68.1)
P<0.01*
Person who decides visit to family Husband, partner or someone else Alone, with husband or partner
8189 (93.4) 1635 (16.6)
366 (22.4)
2614 (31.9) 1269 (77.6) 5575 (68.1)
P<0.01*
Person who decides what to do with money husbands earns Husband, partner or someone else Alone, with husband or partner
7340 (74.7) 2484 (25.3)
561 (22.6) 2419 (33)
1923 (77.4) 4921 (67)
P<0.01*
19 3.4. Determinants for contraceptive use
The binary logistic regression was structured to determine the association between
contraceptive use and decision-making role, socio-demographic and reproductive factors.
Univariate Analysis
The univariate analysis found that decision-making role in regard to health care, household purchases, visit to family and friends, and husbands’ earnings were associated with
contraceptive use. Socio-demographic factors such as age, type of place of residence, education level, literacy, use of internet, wealth index, respondent currently working, ownership of house, and ownership of mobile also had a significant association with
contraceptive use. However, religion had one insignificant value. For the Muslim religion the OR (0.9) showed a lower likelihood of using contraceptives but was not significant.
Christians and Traditional believers had a high likelihood to use contraceptives.
Reproductive factors such as age at first sex, knowledge of ovulation, and fertility preference had an association with contraceptive use. The significance level was set at p<0.05 and CI at 95%. Table 3 illustrates the association between contraceptive use and other factors.
Multivariable Analysis
The multivariable analysis found that decision-making in regard to husbands’ earnings, age, type of place of residence, education level, literacy, wealth index, respondent currently working, age at first sex, knowledge of ovulation, and fertility preference had a significant value (p<0.05) on contraceptive use. All other variables no longer held any association. The predictors that still held significant results and others are further explained below and referenced in Table 3.
Decision making role
For decision making role in regard to health care, household purchases, and visit family a
positive association was found between contraceptive use and females whom make decision
alone or jointly (with husband, partner or someone else) in the unadjusted analysis (univariate
analysis). However, in the adjusted analysis (multivariable analysis) these results no longer
were statistically significant. Decision making in regard to husbands´ earnings the results
were significant in the unadjusted and adjusted analysis. These women were more likely to
20
use contraceptives, compared to women whom husband, partner or someone else make decisions.
Socio-demographic factors
The results reveal that youth (15-24) and young adults (25-34) had a higher likelihood for using contraceptives compared to adults (35-49). For type of place of residence urban residents had higher likelihood for using contraceptives compared to rural residents having OR of 1.16. Respondents and husband’s education level were significantly associated with contraceptive use. Those with lower education level were more likely to use contraceptives compared to those with secondary or higher education. Respondents who were illiterate had a lower likelihood (OR 0.68) for utilizing contraceptives. A negative association was found between contraceptive use and wealth index and respondent currently working. However, for religion, use of internet, own a house, and own a mobile the results found in the unadjusted analysis were not held at the adjusted.
Reproductive factors
Underaged females (<18) were more likely to use contraceptives compared to women of legal age. Also, females whom fertility preference was to not have more children were more likely to use contraceptives, compared to females who want to have children or had not decided.
Those with right knowledge of ovulation had a higher likelihood (OR 1.77) for using contraceptives compared to those with no knowledge and wrong knowledge.
Table 2.
Unadjusted Logistic and Adjusted logistic regression. Showing the association between contraception use and decision-making role, socio-demographic factors, and reproductive factors (variables included in the Adjusted Analysis model were those who had a p-value of <0.05). *Bold text indicates values that are statistically significant at p<0.05.Variable
Unadjusted
OR (95%CI) P-value
Adjusted
OR (95%CI) P-value
Decision making role Person who decides health care
Husband, partner or someone else Alone, with husband or partner
Ref
1.73 (1.529-1.946) P<0.01*
Ref
1.08 (0.918-1.279) P=0.344 Person who decides household
purchases
Husband, partner or someone else Alone, with husband or partner
Ref
1.63 (1.626-1.434) P<0.01*
Ref
1.07 (0.907-1.264) P=0.421 Person who decides visit to family
Husband, partner or someone else Alone, with husband or partner
Ref
1.63 (1.434-1.843) P<0.01*
Ref
1.042 (0.853-1.272) P=0.688 Person who decides what to do with
money husbands earns
Husband, partner or someone else Alone, with husband or partner
Ref
1.67 (1.516-1.873) P<0.01*
Ref
1.26 (1.100-1.438) P<0.01*
21
Socio-demographic factors Age
35-49 Adults 25-34 Young adults 15-24 Youth
Ref
1.43 (1.295-1.586) 1.30 (1.151-1.451)
P<0.01*
P<0.01*
Ref
1.73 (1.53-1.951) 1.80 (1.500-2.100)
P<0.01*
P<0.01*
Type of place of residence Rural
Urban
Ref
2.50 (2.228-2.693) P<0.01*
Ref
1.16 (0.997-1.345) P<0.05*
Education level Secondary Primary No education
Ref
0.85 (0.732-0.987) 0.36 (0.309-0.414)
P<0.05*
P<0.01*
Ref
1.27 (1.054-1.496) 1.07 (0.857-1.330)
P<0.01*
P<0.05*
Husbands education level Higher
Secondary Primary No education
Ref
0.57 (0.952-0.804) 0.84 (0.729-0.973) 0.36 (0.313-0.417)
P=0.417 P<0.05*
P<0.01*
Ref
1.22 (1.017-1.473) 1.72 (1.43-2.066) 1.28 (1.054-1.562)
P<0.05*
P<0.01*
P<0.01*
Religion Other Christian Muslim
Traditional belief
Ref
3.17 (1.594-6.303) 0.90 (0.452-1.795) 2.60 (1.099-6.177)
P<0.01*
P=0.766 P<0.05*
Ref
1.90 (0.922-3.895) 0.70 (0.334-1.420) 1.366 (0.550-3.392)
P=0.082 P=0.313 P=0.502 Literacy
Can read Cannot read
Ref
0.37 (0.338-0.410) P<0.01*
Ref
0.68 (0.581-0.794) P<0.01*
Wealth index Rich
Middle Poor
Ref
0.66 (0.579-0.747) 0.24 (0.220-0.269)
P<0.01*
P<0.01*
Ref
0.38 (0.337-0.440) 0.852 (0.730-0.993)
P<0.01*
P<0.05*
Use of internet Yes
No
Ref
0.42 (0.343-0.507) P<0.01*
Ref
0.98 (0.793-1.255) P=0.984 Respondent currently working
Yes No
Ref
0.55 (0.507-0.607)
P<0.01* Ref
0.811 (0.733-0.898) P<0.01*
Owns a house Yes
No
Ref
1.49 (1.355-1.616)
P<0.01* Ref
1.02 (0.921-1.138) P=0.664 Own a mobile
Yes No
Ref
0.48 (0.441-0.529)
P<0.01* Ref
1.01 (0.882-1.170) P=0.831 Reproductive factors
Age at first sex Legal age Illegal age
Ref
0.87 (0.800-0.956) P<0.01*
Ref
1.106 (0.997-1.227) P<0.05*
Fertility preference No more children Have more children Undecided
Ref
0.75 (0.681-0.818) 0.61 (0.487-0.777)
P<0.01*
P<0.01*
Ref
0.626 (0.558-0.702) 0.615 (0.476-0.794)
P<0.01*
P<0.01*
Knowledge of Ovulation No knowledge
Right knowledge Wrong knowledge
Ref
2.31 (2.015-2.660) 1.37 (1.205-1.544)
P<0.01*
P<0.05*
Ref
1.77 (1.016-3.068) 1.20 (0.700-2.066)
P<0.05*
P=0.501
