Gulelat Amdie
Gulelat Amdie
Gulelat Amdie
BY
GULELAT AMDIE
A thesis submitted to
May, 2005
BY GULELAT AMDIE
Advisors
May, 2005
Addis Ababa, Ethiopia
ADDIS ABABA UNIVERSITY
BY
GULELAT AMDIE
-------------------------------------
------------------------------------------- -----------------------------------------
Examiner
------------------------------------------- ----------------------------------------
Examiner
DECLARATION
I, the under signed, declare that this is my original work has never been presented in this or
any other university and that all the source material used for the thesis have been duly
acknowledged.
Signature-----------------------------------
This thesis has been submitted for examination with my approval as a university
Advisor:
Name-----------------------------------------
Signature------------------------------------
Date------------------------------------------
ACKNOWLEDGMENT
I would like to acknowledge Action aid Ethiopia and Addis Ababa University for financing this
study and the Department of Community Health for the supports it rendered to me in
My sincere and deepest gratitude goes to my advisors Dr. Alemayehu Worku and Professor
Yemane Berhanne for their unreserved assistance, timely comments and relevant guidance
from the beginning of the research proposal to the write-up of the final paper. My
appreciation also goes to my instructor Dr. Nigussie Deyassa, who deserves great respect
and due regard for the support he rendered from the beginning to the end of this work.
I am very grateful to all the study participants and particularly those women who disclosed or
let us to know their painful violence experiences. I also thank health workers for their
responsible data collection and support and I appreciate their contribution in carrying out this
I would also like to acknowledge and thank Addis Ababa City Administration Health Bureau,
Organization for Social Services in AIDS, Zewditu Hospital, Kazanchis, Kebena, Kirkos, AND
Addis Ketema health centers staffs and the respective health institutions heads for their
permission and cooperation to conduct of this study. I also thank the Ethiopian Women
Lawyers association for the psychological and legal support it provides for those of study
iv
TABLE OF CONTENTS
ACKNOWLEDGMENT .....................................................................................................IV
TABLE OF CONTENTS ....................................................................................................V
LIST OF TABLES .............................................................................................................VI
LIST OF ABBREVIATIONS .............................................................................................VII
ABSTRACT ....................................................................................................................VIII
1. INTRODUCTION ........................................................................................................... 1
2. LITERATURE REVIEW ................................................................................................. 3
2.1 HIV/AIDS IN ETHIOPIA .................................................................................................. 3
2.2 GENDER BASED VIOLENCE IN ETHIOPIA ............................................................................ 4
2.3. Overlaps between Violence and HIV Infection among women ................................... 6
2.4. CONNECTIONS BETWEEN GENDER BASED VIOLENCE AND HIV INFECTION .......................... 7
2.5 RATIONALE .................................................................................................................. 14
3. OBJECTIVES .............................................................................................................. 15
3.1 GENERAL OBJECTIVE ................................................................................................... 15
3.2. SPECIFIC OBJECTIVES .................................................................................................. 15
4. METHODS AND MATERIALS..................................................................................... 16
4.1. STUDY DESIGN: CROSS-SECTIONAL SURVEY ................................................................. 16
4.2. STUDY SETTING .......................................................................................................... 16
4.3 STUDY POPULATION ..................................................................................................... 16
4.4 SAMPLE SIZE ............................................................................................................... 18
4.5 SAMPLING ................................................................................................................... 18
4.6 DATA COLLECTION ....................................................................................................... 19
4.6.1 Data Collection Instrument ............................................................................................................. 19
4.6.2 Data Collection Setting and Process........................................................................................... 20
4.6. 3. Measurement and Variables ....................................................................................................... 21
4.6.4 Data Collectors ...................................................................................................................................... 23
4.6.5 Data Quality Control ............................................................................................................................ 24
4.7 OPERATIONAL DEFINITIONS ........................................................................................... 25
4.8 ETHICAL CONSIDERATIONS ........................................................................................... 27
4.9 DATA ANALYSIS ........................................................................................................... 29
5. RESULTS.................................................................................................................... 31
5.1 PARTICIPATION ....................................................................................................... 31
8. CONCLUSIONS AND RECOMMENDATIONS............................................................ 58
9. REFERENCES ............................................................................................................ 60
v
LIST OF TABLES
Table 1: Basic Social and demographics of women and current/most recent partner
as reported by the women among 743 women attending VCT services in 6 32
health institutions in Addis Ababa city, 2005.
Table 3: Table 3: Prevalence of Substance use and risky sexual behavior and
Crude odds ratios for testing HIV seropositive by substance use and 40
sexual behavior among 743 women attending VCT service in 6 health
institutions in Addis Ababa city, 2005.
Table 4: Odds ratios for association between IPV and Risk behavior women 43
attending VCT services in Addis Ababa City, 2005.
Table 5 Odds ratios for testing HIV seropositive by experience of IPV after 46
adjusting for women risk behavior and socio-demographic characteristics
among women attending VCT services, 2005
: Table 6 Table 5: Crude odds ratios for testing HIV seropositive by experience of 47
various forms of sexual violence among women attending VCT services in
Addis Ababa City, 2005.
: Table 49
7: Odds ratios for associations between forms of gender-based violence and
risk behavior, among women attending VCT service in Addis Ababa city,
2005.
vi
LIST OF ABBREVIATIONS
vii
ABSTRACT
Background: HIV/AIDS and violence are among the major health problems affecting the
violence and fear of violence are emerging as important risk factors contributing to their
vulnerability to HIV infection, directly through forced sex and indirectly by constraining
women’s ability to negotiate safe sexual behavior. Few studies linked men’s use of violence
to their own high risk sexual behavior, hence, their own as well as their partner’s risk of HIV
infection. Despite the recognition of both problems of violence and HIV in Ethiopia, no study
yet assessed gender based violence as a risk factor for women’s HIV infection.
Objectives: This study was conducted: to measure the prevalence of various forms of
gender based violence, including intimate partner physical and sexual violence and to assess
associations between gender-based violence, HIV risk behaviors and HIV infection among
women attending voluntary counseling and testing service in Addis Ababa City.
Methods: We did a cross sectional study among 743 women attending VCT service at six
health institutions in Addis Ababa City. Women who ever engaged in steady heterosexual
multiple male partners, casual partners, transactional sex, condom use and alcohol/Khat use
and experience of psychological, physical and sexual violence from intimate partner,
childhood sexual assault, forced first intercourse and sexual assault by non partners were
assessed using structured questionnaire and linked with women serostatus data.
viii
RESULT: The prevalence of lifetime intimate partner physical and sexual violence was
54.6% and 41% respectively and 21.8% of women reported experiencing forced sex or rape
at their first sex. At the date of interview 35.4% of women tested HIV positive. After
adjustment for socio-demographic characteristics and women risk behavior, intimate partner
violence was associated with HIV seropositivity. Childhood sexual assault, forced first
intercourse and adult sexual assault by non partner were not associated with HIV serostatus.
In our study participants, condom use, refusal of sex and demand for monogamous
CONCLUSION: in our study participants, women partnered with violent men are at increased
risk of HIV infection. Our data support the hypotheses that abusive partners are more likely to
have HIV and place their female partners at high risk of HIV. HIV/AIDS intervention need to
target male sexual risk taking and need to work at broader and societal level to challenge
ix
1. INTRODUCTION
Victims of HIV/AIDS epidemic in sub-Saharan Africa are young and female. From an
estimated 8.6 million young victims in the region, 5.7 million are young women.
Worldwide, African women are the group most severely affected by the epidemic
accounting 66% of those infected between the ages 15 and 24 years(1). In Ethiopia the
prevalence of HIV is 5% in females and 3.8% in males and the number of females
infected by HIV/AIDS is much higher than the number of males (2). As women get
increasingly infected by HIV/AIDS, gender based violence and gender inequality have
A growing number of studies have documented the high prevalence of intimate partner
violence and sexual violence against women worldwide. A recent study conducted in
south central Ethiopia, reported that 49% and 59% of women physically and sexually
abused by their partners at some point in their life respectively (5). Other studies
conducted in different areas reported attempted rape ranging from 10-11.5% , including
rape more than once and consequences like unwanted pregnancy, suicidal attempt, and
abortion ( 6).
Violence has been linked to different health outcomes, both in short term and long term.
physical functioning, psychological well being like depression, suicidal attempts, and face
1
unsafe abortion and unwanted pregnancy. Violence during pregnancy not only affect the
women but also the growing fetus and has been associated with miscarriage, still birth,
premature labor and birth, fetal injury and low birth weight which is major cause of infant
contributing to women’s increased risk of HIV infection both directly through forced sex
and indirectly by constraining women’s ability to negotiate the circumstances in which sex
takes place and the use of condoms. Sexual abuse during childhood also seems to be
associated with high-risk behaviors in later stages of life that may also increase the risk of
HIV. Few studies also highlight that men’s use of violence is linked to their own high risk
sexual behavior, hence, their own as well as their partner’s risk of HIV infection, (9).
As the epidemic of HIV and violence continue seriously affecting women, identifying the
factors that place women at risk of HIV and violence is crucial to plan and implement
effective prevention strategies (9). However, given that intimate partner violence is
considerably more common than rape, the possibility that it could contribute to risk of HIV
infection is an important, but largely unexplored area (9). Despite the recognition of
problem of both violence and HIV in Ethiopia, yet no study examined violence against
women as a risk factor for their HIV infection and no research assesed the impact of
violence, HIV risk behaviors and risk of HIV/AIDS infection among women attending VCT
2
2. LITERATURE REVIEW
2.1 HIV/AIDS IN Ethiopia
Ethiopia has one of the worst HIV epidemics in the world. There are an estimated 1.5
million HIV infected individuals. According to the MOH 2004 report the HIV prevalence
estimate for Ethiopian adults in 2003 was 4.4%. The corresponding estimate for urban
Ethiopia is 12.6% whereas that for rural Ethiopia is 2.6%. The HIV prevalence for Addis
Ababa in 2003 was estimated at 14.6%. The 2003 estimated prevalence among female
was 5% whereas among males was 3.8%. The estimated number of new AIDS cases in
the adult population in 2003 was 98,000 of which 54% females and 46% males. This
indicates that women are the group severely affected by the epidemic (2).
On one hand, AIDS awareness in Ethiopia is generally high, yet existing public health
preventative and educational efforts have had limited success in curtailing the epidemic
(10). Ethiopia Demographic and Health Survey (CSA 2001) reported high level of
awareness of HIV/AIDS amongst the general population, specifically 96% of males 85%
females reported high level of awareness about HIV/AIDS (11). On the other hand
population did not use condom with their non commercial partners. Such lack of condom
extramarital sex in the past 12 months and 73% perceive no risk of HIV infection after
Recently in Addis Ababa, much attention has been focused on the prevention of prenatal
HIV transmission using antiretroviral drugs (12). Less attention has focused on the critical
3
issue of women’s HIV risk although protecting the child bearing women from HIV infection
Moreover, evidence suggests that women are more vulnerable than men to becoming
understanding of the circumstances of Ethiopian women’s lives which place them at risk
of HIV infection and impact their ability to implement self protective strategies such as
condom use mutual monogamy and abstinence. This information is necessary to identify
possible strategies for interventions efforts, and to inform appropriate policy development
not limited to: acts of physical, sexual, and psychological violence by intimate partners,
dating partners or family members; sexual assault and rape (including stranger rape,
acquaintance/date rape and marital rape); childhood sexual assault of girls; sexual
includes: Physical violence (e.g. slaps, punches, kicks, assaults with a weapon,
homicide); Sexual violence (e.g. rape, coercion and abuse includes use of physical force,
verbal threats, and harassment to have sex, unwanted touching or physical advances,
forced degrading acts that often persist over time and are accompanied by threats on
part of the perpetrator); Psychological violence (e.g. belittling the woman, preventing her
4
from seeing family and friends, intimidation, withholding resources, preventing her from
population are difficult to obtain. The most reliable data generally derive from surveys
conducted in Meskan and Mereko district in south central Ethiopia found 29% prevalence
their life time. Among pregnant women the history of physical violence during a
pregnancy was 77%. Forty four percent of the women experienced sexual violence
within the last 12 months and 59% experienced sexual violence by partner at some point
in their lives. When severity of violence considered, 35% of women experienced at least
one severe form of violence including hitting, kicking, choking or the use of weapon. The
study also reported considerable overlap between sexual and physical violence. Among
ever partnered women 42% experienced both physical and sexual violence over their life
time. Among the study participants 7% reported sexual violence before the age of 15
year (5).
violence among high school female students in Debark, North West Ethiopia (6), The
prevalence of performed and attempted rape was 8.8% and 11.5%, respectively. The age
range of performed rape victims was between 12 and 21 years. Of the 19 (8.8%) who
reported rape being performed on them, unwanted pregnancy, suicide attempt, vaginal
discharge and abortion were the consequences in 21%, 15.8%, 10.5% and 5.3%,
5
respectively (6). One study conducted in Addis Ababa showed 15.6% prevalence of
completed rape within three months among female street adolescents in Addis Ababa,
including experiencing rape more than once and consequences like unwanted
Today, half or more of the 40 million people infected with HIV in the world are women.
Young people aged 15-24 years account for half of all new infections worldwide. In sub-
Saharan Africa, young women account for 75 % of HIV infections and are approximately
three times more likely to be infected than young men of the same age (1). So, what
makes women, especially girls and young women so disproportionately vulnerable to HIV
and why have current AIDS control efforts largely failed to control the spread of the
epidemic in women and girls? These high rates of HIV infection in women have brought
into sharp focus the problem of violence against women and increasingly cited by
Some studies indicated that HIV infection and other STDs are among the major
consequences for women experiencing violence (4). HIV infected women reported more
experience of physical and sexual violence from intimate and non intimate partners,
including sexual assault, sexual coercion in their child hood, adolescent and adult life
than non-infected women (9). Women who are raped are also found to be at high risk of a
pre-existing sexually transmitted infection (STIs) and lower but substantial increased risk
of STIs (14). Even though such studies are limited in Africa, a small but growing number
6
of studies have begun to explore the connections between gender based violence and
risk of HIV infection among women and few studies provide valuable evidences (13). A
study conducted in Uganda showed that, women who reported being forced to have sex
against their will in the previous year had an eight fold increased risk of becoming
infected with HIV(15). Another study conducted in South Africa reported that women who
are in relationships with violent or controlling relationships were 50% more likely to
contract HIV than women not involved in abusive relationships (13). Studies conducted in
Rwanda (16), and Tanzania (also reported higher prevalence of HIV infection among
women who experienced intimate partner violence than women living in non abusive
partner violence and HIV infection after adjustment for women’s own high risk behaviors
Gender-based violence can make women vulnerable to HIV through three main
coerced sexual acts. Secondly, the trauma associated with violent experiences can
impact later sexual behavior, by increasing the women’s HIV risk taking behavior. Third,
violence or the threat of violence may limit women’s ability to adopt safer and HIV
protective sexual practices within on-going relationships and may detract from using HIV
related services such as STIs treatment, VCT and PMTC services (9, 26).
7
2.4.1 The Role of Forced Sex in HIV Transmission
Violence or forced sex can increase the risk of transmitting HIV. The biological risk of HIV
transmission will be affected by the type of sexual exposure, the presence of STDs,
exposure to vaginal excretions or blood and the degree of physiological trauma. When
sexual intercourse is forced, abrasions and cuts are more likely to appear-thus facilitating
the entry of the virus through the vaginal mucous. Adolescent girls are particularly
susceptible to HIV infection through forced sex, and even through unforced sex, because
their vaginal mucous membrane has not yet acquired the cellular density providing an
effective barrier that develops in the later teenage years. In addition, condom use in such
situations is unlikely. In such situations where all these coupled, the possibility or likely
Increasing evidence from developed and developing countries indicates the links
stages of adult life, which may increase the risk of HIV infection. Here, violence indirectly
impact women HIV risk taking behavior through long-term by affecting women’s
Many study findings reported links between violence in childhood and adolescence to
disorders, and low self esteem- these factors in turn found are associated with many of
HIV risk behavior. Early age experience of sexual violence such as forced first sexual
8
intercourse and other forms of sexual assault have been associated with traumatic
sexualisation, mental health problems, domestic violence and other problems in intimate
relationships, including engaging in behaviors associated with excess HIV risk – such as
early initiation of sexual activity, high number of sexual events within the last few months,
high number of sexual partners, high number of sexual encounters with non-primary
partners or casual partners, sex with a known risky partner, sex while intoxicated,
receptive anal sex, low condom use, recent history of STDs, and engaging in sex work
(13, 26).
determinant of high-risk sexual activity in adolescents, and forced first intercourse was
found to be associated with increased risk of teenage pregnancy fourteen-fold times after
adjustment for socio-economic characteristics and other factors. Moreover, studies also
women who were raped before age 18 years were twice as likely to be raped as adult,
compared with those who were not raped as children or adolescents (4). Women who
engaged in partnership with abusive men in their adult life more than women not
gender-based violence as an adult and having multiple sex partners, multiple sexual
9
encounters, causal sex partners, low condom use, having a male partner with other
female partners, contracting STD and, using alcohol and drugs, and engaging in
commercial sex or trading sex for money or drugs. But the evidence is not conclusive
whether having many sexual partners, engaging in a commercial sex or alcohol or drug
esteem, which then affects self-perceived ability to negotiate and implement safer sex.
Hence, both the fear of violence , treats of violence and actual violence affects women’s
While the evidence is not conclusive, women who experienced forced sex in intimate
relationships often find it difficult to negotiate condom use –either because using a
or else they fear experiencing violence from their partners. Studies have found that
women with a history of violence, fear violent retaliation for requesting condom use, and
are less likely to use condom. Likewise, attempts to use condom and access STD
services are likely to lead to abuse (4). Women in Zimbabwe reported physical violence
and forced sex as retaliation for attempts to refuse sex with their partners (24). Fearing
violence or rejection, 58% of south African girls reported avoiding discussing condom use
with their partners. Yet, in couples where only one partner is infected with HIV,
10
consistent and correct condom use provides the HIV negative person with a near zero
proportion of adolescent girls have sexual relations with men five to ten years older than
themselves. While girls are able to initially choose the older sexual partner, once in the
relationship, it is the older men who control the sexual relationship including condom use
- in some situations through the use of violence (14). A study conducted among young
women (16-23 years) in South Africa suggests that women with older partners (older than
them three or more years) have 1.6 fold higher odds of being HIV infected and young
women with older partners are 1.5 times more likely to experience physical and sexual
violence than women with partners in the peer age group. The researchers suggest that
partner violence may be a feature of relationships with older men and that age difference
between partners increases young women’s HIV risk because older men have a much
Fear of violence also seems to limits women’s ability to ask about their partners’ other
highlighted that men’s use of violence is linked to their own sexual risk taking and hence,
their own as well as their partner’s risk of STI and HIV (14). In India, a study showed that
men who had extramarital sex were six times more likely to report sexual abuse of their
wives than men who remained faithful. And men who reported an STI were 2.5 times
more likely to report abuse of wives than men who did not report an STI. In another study
from South Africa, men who reported use of sexual violence against intimate partners
11
were nearly twice as likely to have multiple partners compared to those who did not use
sexual violence. These evidences indicated that abusive men were more likely to engage
in extramarital sex, acquire HIV or other STI, and impose more risky sexual practices on
their partners and place their female partners at higher risk for HIV or other STIs possibly
which may increase women's HIV risk. Economically vulnerable women often depend on
men’s financial contributions and are less likely to successfully negotiate safer sex and
less likely to leave relationships perceived as risky. Economic vulnerability also increases
the likelihood that women will exchange sex for money or favors, and to engage in the
sex industry and the economic vulnerability limits their ability to enter this exchange on
Even though the evidence is not conclusive, violence or fear of violence has been
implicated as a barrier to women seeking HIV testing. In Uganda, research indicated that
women were afraid to ask for money or permission from their husbands to attend
HIV/AIDS facilities or seek information and in some cases explicitly forbidden from taking
HIV tests (16). Violence or fear of violence was also implicated as a barrier to disclosure
of HIV status among those women tested for HIV. Between 16 - 86 % of women in
developing countries choose not to disclose their HIV status to their partners. On the
other hand, disclosure of HIV status is considered to be important for ensuring that HIV
positive individuals are able to access a range of services including prevention of mother
12
to child transmission (MTCT), anti-retroviral treatment (ART), and psychosocial support
(20). For example, disclosure by HIV positive women to their sexual partners could
enable couples to make informed reproductive health choices such as seeking family
risk behaviors. For discordant couples, disclosure of HIV status also provides an
opportunity to the male partner to take additional measures not to be infected from his
Studies on disclosure suggest that for a majority of women, their partners’ reaction was
violence or negative outcomes on women tested for HIV and disclosed their HIV sero-
blame, abandonment, anger, and violence. In studies from Tanzania, South Africa, and
Kenya, among those who do not disclose their status, between 16 - 51 % women
reported fear of violence as one of the major barriers to disclose their HIV status.
Generally, violence seems to challenge all available HIV prevention measures such as
abstinence, monogamy and condom use and intervention strategies such as VCT,
PMTCT, ART, and psychosocial support. In other words these major intervention
13
2.5 Rationale
Most of the studies to explore the link between violence and HIV were conducted in
developed countries and due to low prevalence of HIV and violence, these studies had
drawn samples from very high-risk groups extremely exposed to violence. Studies
conducted in Africa were limited in number and limited in measuring the different forms,
frequencies and severity of violence and measured experiences of violence after women
knowing their serostatus, likely to introduce either recall or reporting bias and it is
impossible to distinguish whether partners who learned of the women’s HIV serostatus
become violent as a result. In addition, these studies used self reported HIV risk
behaviors and history of STDs as proxy indicators of HIV risk, potentially unreliable
assumption that there is simple relation between risk behaviors and history of STDs and
HIV risk.
Above all much of the studies to explore the link between violence and HIV have been
conducted in developed countries and their application in different cultural and social
context of Ethiopia is doubtful. (13, 17) Therefore, to fully understand the association
between HIV infection and gender based violence in Ethiopian context this cross
sectional survey was conducted among women attending VCT services in Addis Ababa
city, specifically to explore the following questions. Did HIV positive women experience
more intimate partner violence, than HIV negative women? Does self-reported HIV risk
condom and alcohol use associated with experience of gender-based violence? Finally,
does the association between violence and HIV sero-status sustain after adjustment for
14
3. OBJECTIVES
To assess the relationship between gender-based violence, HIV risk behaviors and HIV
1. To measure the prevalence of various forms of gender based violence among women
4. To assess the relationships between gender-based violence and HIV risk behaviors
5. To assess the association between gender based violence and HIV infection among
15
4. METHODS AND MATERIALS
In Addis Ababa City, there are more than 30 governmental and non governmental health
institutions provide VCT service. From these health institutions providing VCT service, six
sites were selected to conduct the study based on their willingness, cooperation,
availability of adequate clients and counselors (to conduct the study with out constraining
the routine service). The selected VCT centers were Zewditu Hospital, Organization for
Social Services in Aids Lideta VCT center, Kazanchis Health Center, Addis Ketama
Health Center, Kirkos Health Center, and Kebena Health Center. Zewditu hospital and
Organization for Social Services in Aids Lideta VCT center have free standing VCT
clinics and the rest provide the service in integrated manner. The selected health
institutions are used by the general population and they had the personnel, equipments
and supplies and room to conduct the study. Therefore, the study participants were
recruited from women attending voluntary testing and counseling services in these sites.
The Study population was women attending VCT services in six health institutions in
Addis Ababa city. From women who presented for VCT from December, 2004 to March
2005, women who were eligible to the study were enrolled in the study. In order to assess
16
socio-demographic characteristics of partners and their risk behaviors as perceived by
the women and violence inflicted by intimate partners, women who were not engaged in
regular sexual relationship with a male partner in their life time were excluded from the
study.
Previous studies indicated that women’s reporting experience of violence and risk
result. It may also affect women’s disclosure of risk behavior or introduce interviewee and
interviewer bias (13). Therefore, to ensure that all participants would be unaware of their
HIV status at the time of the interview, those women who had previously received their
Prior research demonstrated that being commercial sex worker is a high risk factor for
HIV infection and experiencing violence from intimate and non intimate partners . Besides
commercial sex workers may not have regular partners (4). Therefore, our primary
objective of assessing violence mainly from regular intimate partners may be affected by
high prevalence of both violence and HIV infection among these groups. As result,
women had ever been engaged in commercial sex work were excluded from
participation.
Generally, eligible women were 18 years or older (VCT Policy), ever engaged in regular
sexual relationship with a male partner in life time, never engaged in commercial sex
work (in bars or as a street girl), never tested for HIV and received HIV test result, not
17
presented for VCT with their male partners, not present for VCT with partners, have
common language with interviewer, mentally or physically capable to answer the study
We use EPI6 statat calculator for cross sectional study design to calculate the sample
size. To calculate the required sample size we used 25% of prevalence of physical and
variable and we assume 14% prevalence of HIV among Unexposed group (no physical
and sexual violence). We calculated the sample size to identify 2 times odds of HIV
among women exposed to physical and sexual violence. Confidence level of 95% and
power 80% considered. The calculated sample size was 740 and by adding 10% for non
4.5 Sampling
From the total women presented for VCT from December, 2004 to March 2005, women
who were eligible to the study were enrolled in the study. All eligible women during the
study period were included when the counselors had adequate time to interview them
18
4.6 Data Collection
The instrument used to collect data is structured questionnaires, which was first drafted in
English and then translated to Amharic. Our primary tool for assessing the experience of
intimate partner violence (IPV) was the WHO violence against women instrument,
developed for a multi country study on women’s health and domestic violence (19). After
translating WHO questionnaire from English to Amharic we compared it with the WHO
Amharic version particularly used by the Butajira Reproductive Health Project and further
commented by lay persons, professionals in the field and further pre-tested and
discussed with data collectors to ensure group consensus on meanings and appropriate
The questionnaire contained sections for assessing demographics of women and current
or most recent regular partner, marital status and duration of relationship, risk behavior
of women and current or most recent regular partner as perceived by the women,
violence, women substance use, transactional sex and section to end the interview on
positive note. More sensitive issues, including the nature and extent of partner and non
partner violence and women’s use of substance were explored in latter sections once
rapport between interviewers and respondents had been established. To the extent
possible, the questionnaire employed standard measures which had been tested and
19
4.6.2 Data Collection Setting and Process
Data collection was done from December 2004 to March 2005 in six health institutions
VCT centers found in Addis Ababa City Administration. Almost all data on socio-
demographic characteristics of the respondent and eligibility criteria were part of the
routine VCT risk assessment and counseling form. After completing the pretest
counseling and establishing a good rapport with the potential participant and assessing
the respondent for eligibility, the interviewers asked the women to participate in the study.
Women who expressed interest in participation, and signed the informed consent
To obtain serostatus data we used the routine VCT testing procedures. In all health
institutions after the pre test counselling, the counsellors were expected to collect
specimens, provide code and submit to the laboratory. Women were interviewed after
completing the pre test counselling and then the interviewers (counsellors) collected
specimen, coded, submit specimens to the laboratory and finally collect the test results
and communicate to clients. In our study the counsellors recorded the code - they
provided for the specimen on the space provided on the questioner and when they obtain
the test result they recorded or merge the serostatus data with the interview data. All
specimens were tested using rapid test kits (Vironstika ® ETA) and positive results were
confirmed by ELISA.
20
4.6. 3. Measurement and Variables
respondent and her current or most recent partner; marital status, living arrangement and
duration of relationship. We also asked women reason for testing and partner testing
status.
HIV Risk behaviours or risk factors were assessed using indictors such as number of
regular and casual male partners, condom use, ever engaging in transactional sex,
alcohol or drug (Khat use). To assess the risk behavior of current or most recent male
partner all women were asked about their current or most recent regular partner risk
behaviors, i.e. whether their current or most recent regular partner had sexual
relationship with other women (other than the respondent) while they were in relationship
with the respondent. Women were also asked whether their partners had sex with
commercial sex workers prior to and during their relationship and partner use of alcohol
and Khat and her perceived frequency of using alcohol and Khat.
We assessed experience of violence from any intimate male partner using the WHO
violence against women instrument and our version contains 6 questions for emotional
abuse, 4 questions for physical abuse and 4 questions for sexual abuse. To assess
experience of emotional abuse the interviewer asked the participant whether a current (or
most recent partner) or any other partner ever prevented her from seeing family or
21
friends; speaking with other men or gets angry if speak; ever controlled her movements;
ever insulted; belittled or humiliated in front of other people; scare or threatened to hurt
(19).
To assess the experience of physical abuse, women were asked whether they were ever
slapped; pushed or something thrown at them could hurt ; Kicked , dragged, beaten, hit
with fist ; Choked, strangled or burnt on purpose; threatened or hurt by gun, knife, or
other weapon. Four items were inquired about sexual violence practices against women;
being physically forced to have sex; having sex because of being afraid to refuse; forced
to do something sexual which is degrading or humiliating and being forced to have sex
after beaten by partner. All women who experienced the item in question were asked the
perceived frequency (once, few times or many times) in the past 12 months and prior to
the past one year and finally we asked the reasons for partner violence (19).
To asses women’s experience of rape by non intimate partners we asked all women if
they were ever physically forced by some one who has no sexual relationship with them
to have sex when they do not want to. (4). To assess childhood sexual assault (before
age 15), we asked all participants if they were forced to have any sexual contact or to do
any other sexual contact or to have made sexual intercourse before age 15. For women
who answered affirmatively we asked age of the women and the perpetrator at the time
of the incidents and the frequencies of such events. To assess violence during women’s
experience of first sexual intercourse (first coitus), we asked women age at first
intercourse and then asked how first sex happened and made to choose the statement
22
have’, 2) ‘I Didn’t want but tricked, persuaded, forced to marry or happened any way or 3)
‘physically forced/raped’.
The interviewers were female health workers (Nurses by profession) and previously they
at least three years of experience in providing clinical care for patients and they had more
than 2 years of experience in counseling VCT clients. Data collectors were recruited in
good relationship with their clients, and their ability to record responses on questionnaire
accurately. For this study they were trained for two days by the principal investigator and
The first part of the training was devoted to sensitizing trainee interviewers on gender
issues, helping them to develop basic understanding of gender based violence, its
dynamics and causes and its impact on the health and wellbeing of women and children.
During the sessions the subject of violence was openly discussed and participants were
given the opportunities to discuss their own experience and feelings with others to help
them identify women who may need support during the study and to help interviewers to
reflect on the challenges and personal emotions associated with disclosing experience of
violence and its implication for the study. Another focus of the training was developing
skills to minimize any possible distress caused to respondents during the interview and to
23
learn skills for interviewing, taking into account safety and ethical guidelines. Interviewers
received training on how best to ask questions about violence and how to respond if
respondent did become distressed including how to become empathetic and supportive.
At the end of the interview in which a woman had disclosed violence, the interviewer was
trained to identify and reinforce the respondent’s own copping strategies and to remind
her that the information she had shared was important and would help other women.
Different sources were used for the training (18, 21, and 27).
The other part of the training focused on familiarizing interviewers with the questionnaire
and giving them the opportunity to practice using it. This includes holding discussion
the questionnaire and using a guide developed for this purpose. Data collectors also
conducted practice interviews to identify any possible future problems and to take
remedial measures. Then the interviewers conducted pilot test the questionnaire and
problems encountered summarized and discussed with each interviewer on the study
site. The interviewers also suggested reducing the number of questions due to lack of
In addition to the training of data collectors, quality control monitoring was also conducted
by the supervisor and principal investigator to maintain data quality and to avoid wrongly
higher or lower violence reports, missing values and appropriate remedies were taken in
case of deviations. During the data collection, the principal investigator also held regular
24
meeting with interviewers for emotional debriefing and these meetings provided
Intimate Partner Violence (Broad IPV): For this study defined was as women’s
experience of any item from physical or sexual violence at medium frequency (reported
by women ‘few times’) and high frequency (reported by women ‘many times’) or women
sexual, physical and sexual or psychological, physical and sexual). Intimate partner
partner violence or only low frequency of physical or sexual violence or only one type of
violence.
Childhood sexual assault: Defined as a sexual abuse experience that occurred before
they reported rape, forced first intercourse, forced marriage, or made to have sexual
penetration at first sexual intercourse (first coitus). Those women who reported use of
physical force in their first coitus were considered to have experienced forced first
25
intercourse if their first sex was unwanted or unexpected and It includes women made to
have sex when they doesn’t want, forced or had coerced penetration or tricked,
Adult sexual assault (Rape) by Non Intimate Partner (FSNIP): Defined as physically
forced or otherwise coerced penetrative sex by any one apart from a husband or boy
friend or some one who has no sexual relationship with the women before, since age of
15 years.
Current or most recent partner: is the last or the most recent regular male partner of a
women.
Number of Male Partners: For this study the number of male partners was defined as
the number of males with whom the women had penetrative sexual intercourse with or
Number of regular partners: Defined as the number of males with whom the woman
had penetrative sexual intercourse, lived with them in the same house or males with
whom the woman had regular sexual relationship with them and perceived by woman as
Number of casual partners: was defined as the number of males with whom the
woman had penetrative sexual intercourse with or with out consent. These include those
males with whom she had once or few times sexual intercourse, other than regular
partners.
Transactional Sex: is defined as exchange of sex with men for material gains and basic
survival needs. It includes women who exchange sex for money, though they may not
necessarily identify themselves as sex workers. We asked all women whether they had
26
ever engaged in sexual relationship with any one to support themselves or their family for
exchange of cash, or food, cloth, cosmetics, fee for transportation, house rent, or school
or for other items for the women or women’s families. Women who responded yes for this
Substance Use: is lifetime use of alcohol or Khat. All women who reported use of
alcohol were asked about things which might happen to them after they had been
drinking: got in to fight, got arrested, forced by anyone to have sex against their wishes,
had sex with a men just they meet. Women who responded yes to any of these questions
To conduct the study in an ethically sensitive and appropriate manner the following
Careful Recruitment and Training of Data collectors: all interviewers were women
health workers with an educational level of a diploma or above and trained as counselors
and had adequate experience in providing clinical services and counseling. The
interviewers were trained for two days and conducted practice on how to interview for
Informed Consent: Only those women who were interested to participate and provided
written informed consent to access their HIV test result and link with the interview data
with out using personal identifiers (names) by the counselors had participated in the
study.
Interviewing: During the training of the interviewers, the importance of conducting the
interview carefully was addressed with appropriate emphasis. The interviewers training
27
included how to introduce sections enquiring about violence carefully, forewarning the
respondent about the nature of the questions, and how to help the participants to feel at
ease as well as to give the opportunity to either stop the interview, or not to answer some
of the questions. The interviewers also informed to tell all the participants that they were
selected randomly and the interview had no connection with the testing and had no effect
on the service they deserve to get from the health institution. They were also informed to
tell the respondents that they could stop the interview at any time if the respondent they
wish and need not answer all questions if they were not comfortable.
Privacy and Confidentiality: The informed consent and interview procedures were
confidentiality was addressed with appropriate emphasis during training. Data collectors
were told not to interview women they know previously. Translators were not used to
maintain confidentiality. In addition, this study was introduced to other staffs as a study
Referral and additional information: For the study purpose, health institutions with
trained nurse in psychiatry were identified and study participants or victims of violence
referred to these health institutions. We have also identified and got the cooperation of
Organization for Social Services in AIDS (OSSA), which provide psychological and social
support and victims of violence and women requiring such services referred to the
institution. We have also contacted Ethiopian women Lawyers Association and they
agreed to provide legal support including representing in court if they were asked to do so
association. In addition, a small information sheet, which lists the options and services
28
available for women experiencing violence, including address and other health referral
Ethical Approval: Ethical approval for the study was obtained from Addis Ababa
University, Department of Community Health and written permission to conduct the study
was also obtained from Addis Ababa City Administration Health Bureau and from the
health institutions. Generally, every effort was made to follow all WHO ethical and safety
recommendations for research on VAW and on research using HIV test results.
First all completed questionnaires were checked for inconsistencies and missed values.
Questionnaires with significant problems were excluded from the analysis. Before data
entry appropriate coding and editing was performed. Data were entered into Epinfo
6.04d (Center for Disease Control and Prevention 2001). After data entry random
counter checking of already entered data with the hard copy was performed and data
cleaning like correcting outlier entries was also performed and the analysis was
violence, various risk behaviors and HIV infection using simple frequencies, proportions
and ratios. We then analyzed patterns of overlap between different types of gender
characteristics, violence, risk behaviors and HIV infection using logistic regression model.
Previous studies indicated that the more severe the abuse, the grater its impact on
women’s physical and mental health and the impact over time of different types of abuse
29
and of multiple episodes of abuse also appears to be cumulative (4). Because the type of
violence (psychological, physical and sexual) and the frequency of violence are
women HIV risk, WHO recommended research to look not just the presence of violence
and HIV/AIDS but how frequency and severity of violence related to HIV/AIDS. However,
yet there is no agreement on measuring the severity of violence and previously only
physical acts that are more severe than slapping, pushing throwing an object are
generally defined as ‘severe violence’, however, some observers suggest that a focus on
acts alone hide the atmosphere of terror that sometimes permeates violent relationship
and qualitative studies reported that some women find the psychological abuse and
degradation even more intolerable than the physical violence (4).. Therefore, to explore
the presence of a possible dose response relationship between violence and women’s
serostatus and to develop a summary measure for analysis, we assessed the number of
different types and frequency of intimate partner violence with respect to women’s
serostatus against the baseline category of no intimate partner violence. And in our
analysis we also noted that overlaps of different number of types and frequencies of IPV
experience of intimate partner violence (two or more types of abuse or mid to high
(Psychological abuse only, or low frequency physical or sexual abuse only). This
30
5. RESULTS
5.1 Participation
For this study 814 eligible women were approached regarding potential participation. Of
these 17(2.1%) women refused to stay and participate in the study and 18 (2.2%)
reported not having regular intimate partner at the middle of the interview and 36 (4.4%)
participants do not provide usable data. Overall, from 814 eligible women were
approached for the study and 743 (91.3%) women were interviewed.
From the study participants 292 (39.3%) women were between age group 18 and 24 and
319(42.9%) were between the age group 25 to 34 (Table 1). The levels of completed
education reported showed that more than 83% had formal education and most of them
(54.6%) attended high school. About 104 (14%) of our study participants reported their
first sex before age 15 and the majority 314 (42.3%) reported age 15 to 17 for their first
sex. From all participants 253 (34%) women were currently married or living with a male
partner, 282 (38%) had a steady boy friend with whom they were neither married nor
living together, 153 (20.6%) were separated, divorced or widowed and the rest 55 (7.4%)
women had no sexual relationship with a male partner at the time of the study. Overall,
61.2% of women were ever married and 38.8% were never married in their life time but
31
Table 1: Social- demographic characteristics of women and partner, among 743 women
Education of Illiterate/ Elementary (1-6) 268 (36.1) 108 (40.3) 2.7 (1.4, 5.0)
Women High school (7-12) 406 (54.6) 141 (34.7) 2.1 (1.1, 3.9)
12 + 69 (9.3) 14 (20.3) 1.00
Age at First Sex < 15 years 104 (14.0) 45 (43.3) 2.0 (1.1, 3.8)
15- 17 years 314 (42.3) 101 (32.2) 1.2 (0.7, 2.2)
18-20 218 (29.3) 78 (35.8) 1.5 (0.8, 2.6)
Don’t Know 31 (4.2) 18 (58.1) 3.6 (1.5, 8.7)
21-33 YEARS 76 (10.2) 21 (27.6) 1.00
Current Marital Married/live together 253 (34) 89 (35.2) 1.5 (1.1, 2.2)
Status Boy friend/not live together 282 (38) 74 (26.2) 1.00
Separated/divorced 83 (11.2) 27 (32.5) 1.4 (0.8, 2.3)
Widow 70 (9.4) 55 (78.6) 10 (5.5, 19)
Currently no partner/single 55 (7.4) 18 (32.7) 1.3 (0.7, 2.6)
32
Women were asked the duration of their relationship with their current or most recent
partner and 39.7 % women reported relationship of five or more years. Only 11.6% of
women were employed or had professional job and currently, 44.8% of women earn
money and the rest 55.2% have no job in which they can get money.
The distribution of the men’s age was generally higher than that of women’s and more
than 90% of women had relationship with older partner and of which more than half of the
women were in regular relationship with partner older than them by five or more
years(Table 1). Only 7% of women had regular relationship with a partner less or equal to
their age. The levels of completed education reported for partners showed that more than
93% had formal education and most of them (59.5%) attended high school. The
distribution of the men’s education was generally comparable to the women’s, but the
men were far more likely to be earning money. Thirty percent of women were reported
that their current or most recent partner has been a seasonal or daily worker, in the
From our study participants 576 (77.5%) women reported being emotionally, physically or
sexually assaulted by intimate male partner in their life time with 524 (70.5%) of the
overall sample reporting more than one incident (Table 2). Likewise, 427 (57.5%)
at least once during the past 12 months, with 406(35.7%) reporting more than one
33
incident. About 167 (22.5%) participants reported no abuse in their lifetimes and 316
Four hundred six (54.6 %) participants reported being physically assaulted by a male
partner at least once in their life time, with 337 (45.4%) reporting more than one incident
(Table 2). About 295 (39.3%) participants reported being physically abused by a male
partner in the past 12 months with 236 (31.8 %) of the overall sample reporting more than
one incident. The most common form of life time physical violence reported (52.9%) was
moderate form of violence which is the male partner pushing, slapping or throwing some
thing that could hurt the women. About 295 (37.7%) women were experienced at least
one severe form of physical violence in their lifetime and 231 (31.1%) reported more than
About 370 (49.8%) of participants reported that they experienced at least one form of
sexual violence from their male partners in their life time, with 324 (43.6%) women
experiencing more than one incident. Likewise, 265 (35.7%) women experienced one or
more acts of sexual violence in the past 12 months, with 237 (31.4%) reporting more than
one incident. The most common form of sexual violence reported was the male partner
forcing the respondent to have sex and 188 (25.3%) of women reported being forced to
have sex just after they were physically assaulted or beaten by male partner.
34
Table 2-: Prevalence of Intimate partner violence - Psychological, physical and Sexual
violence in Lifetime and in the past 12 months among 743 women attending VCT
35
5.4 Overlaps between Different Types of Intimate Partner Violence
From all study participants 99(13.3%) reported lifetime psychological violence. Only 21
(2.8%) reported sexual violence only, 13 (1.7%) physical violence only and 94 (12.7%)
experiencing psychological abuse with sexual abuse, and the great majority 127 (38.6%)
reported experiencing psychological, physical and sexual abuse(Table 2). Other patterns
From different types of intimate partner violence (psychological, physical or sexual) 156
(21%) women experienced two of the three types and 287(38.6%) experienced all types
intimate partner violence, 461 (62%) of women experienced two or more types of
violence or medium to high frequency of physical or sexual violence (Broad IPV) in their
life time.
About 81 (10.9%) participants reported that they were physically forced in their first
participants reported childhood sexual assault prior to age 15 years (either unwanted
sexual contact, rape, forced first intercourse , or coerced to marriage) (Table 2). From
the study participants 112 (14.9%) women reported experience of rape by non intimate
partner.
36
A total of 153 (21.8%) women reported that they were physically forced or raped at their
first sex. About 114 (15.3%) women were raped or forced at their first sex before they
were 18 years old and 81 (10.9%) were raped or forced at their first sex before they were
15 years old(Table 2). Generally, 276 (37.1%) reported that their first sex was unwanted
(i.e. physically forced, raped or tricked or coerced in to sex). Forced sex in adulthood
someone at work place (17%) (Data not shown in tables). Family members, friends of
family members and neighborhoods, each account for 13.8% of the total perpetrators.
Neighbors were sited as perpetrators for 35.7% of childhood sexual abuse (sexual
contact).
With regarded to violence perpetrated by intimate partners, the most commonly cited
reason for assault by intimate male partner was the jealousy of the partner or suspecting
women’s infidelity; this was cited by 48.4% of women reporting intimate partner violence.
Other commonly cited reasons were the woman's refusal to have sex (44.3%), alcohol
use by men (41.6%), suspected infidelity of partner by the woman (mistrusting partner)
(35%), arguments over money (23.3%). Other reasons were less frequently cited as a
5.6 Reason for Seeking VCT Service and Prevalence of HIV Infection
During the pre test counseling, women were asked their main reason for seeking VCT
service and 12.7% of participants reported that they seek VCT because they perceived
having higher risk of HIV and 34.9% perceive that their partner had higher risk or risk
37
behavior. Of all participants only 6.6% , 4.2% and 3.4%, come to VCT for premarital or
reunion, ANC and visa purpose respectively. Twenty participants (2.7%) reported that
they come to VCT as a result of sexual violence. Of all participants 65.4% reported that
their current or most recent partner was not tested for HIV. During the pre test counseling
women were asked about their intention or plan to share their test result to current
regular partner. Among women who were in regular relationship with a male partner at
the time of the study, 380 (72%) has a plan to share their test result to their partner and
27% refused or were not sure about sharing their test result to their partner.
From the study participants 263 (35.4%) women were tested HIV-positive and 480
(64.6%) were tested HIV-negative on the date of the interview. The prevalence of HIV
among women below age 25 was 19.9% and for age group 25 to 34 was 43% and for
age group 35 to 45 was 53.4%. The prevalence of HIV among currently married women
or women living with a partner was 35.2% and among widow was 78.6%. The prevalence
of HIV among ever married and never married was 73.8% and 26.2% respectively.
Study participants reported from 1 to 29 life time male partners with a mean of 3.54, and
a median of 2. About 228 (37.6%) participants reported having only one male partner in
their lifetime and 578 (62.4%) of participants reported less than five male partners in their
life time (Table 3).Of particular note 335 (45.1%) of participants reported having at least
one casual partner in their life time and 141 (19%) of participants reported casual
38
partners in the past 12 months. About 203 (28%) women reported that they had ever
used condom with partner to avoid STDs. From those women who had casual partners
only 28.1% reported as they used condom with at least one casual partner. From those
women who had never used condom with current or most recent regular partner, 119
(19.4%) women reported requesting current or most recent regular partner to use
About 144(19.4%) of the women were engaged in transactional sex (sex for material or
money gain or to support themselves and family) (Table 3). About 349 (47%) women
reported lifetime (ever) use of alcohol and only 19% of women ever used Khat. From
those who ever used alcohol 91(26.1%) women reported problems associated with use of
alcohol like engaging in fight, having accident, injury, arrest, or having sex with some one
We assessed women’s perceptions about their current or most recent regular partners
HIV risk behavior and substance use. About 418 (56.2%) women reported as they know
or believe that their current partner had have sexual relationship with other women (other
than the respondent) while they were in steady or regular sexual relationship. Likewise,
200 (27%) women reported that they know, believe or think that that their current or most
recent partner ever had sex with commercial sex workers. Overall, 545 (74.1%) women
reported that their current or most recent partner use alcohol, of which 180 (33%) women
reported that their partners use alcohol four or more days per week. Besides, 270
(35.3%) women reported that their partners use Khat, of which 118 (43.7%) reported use
39
Table 3: Prevalence of Substance use and risky sexual behavior and Crude odds ratios
for testing HIV seropositive by substance use and sexual behavior among 743 women
attending VCT service in 6 health institutions in Addis Ababa city, 2005.
40
5.8 Relationships between Violence and HIV risk behaviors
Table 4 provides odds ratios from multiple logistic regression models for the association
between intimate partner violence and each risk behaviors; To explore the presence of a
possible ‘dose-response’ relationship between violence and risk, we modeled the number
experience of IPV (two or more types of abuse or mid to high frequency physical or
sexual abuse) versus "limited to no" experience of IPV (psychological abuse only, or low
frequency physical or sexual abuse only) to address the observation that both frequency
of abuse and number of types were associated with risk behaviors and HIV infection.
After adjustment for age, education and occupational status, childhood sexual abuse was
associated with having three or more male sex partners in one’s lifetime (because of its
sex, and alcohol or Khat use, but not with condom use. Women experience of FFI or rape
(before age 18) was associated with all risk behaviors except never use of condom and
alcohol or Khat use. Women’s experience of rape as adult also associated with all risk
behaviors except transactional sex and never use of condom. After accounting for
women risk of violence such as having multiple partners, casual partners, engaging in
transactional sex and substance use, women experienced childhood sexual abuse or had
forced first intercourse were more likely to be abused in adult life by intimate partners
41
(tables4). Women’s experience of rape by non intimate partner as adult was associated
with a 2 fold elevation in risk for lifetime IPV and significantly associated with increased
risk of IPV in the past year than women not raped by non intimate partner.
After adjusting for socio-demographic characteristics and current or most recent regular
partner risk behavior, having four or more male partners in life time was associated with
women life time experience of intimate partner violence and having five or more male
partners was associated with women’s experience of IPV in the past 12 months. Use of
Khat was also associated with partner violence. Having casual partners, engaging in
transactional sex, life time use of alcohol and condom use were not associated with IPV.
Likewise, after adjustment for socio-demographic characteristics and for women risk
behavior, women whose partner consumed alcohol four or more days per week faced
risks of lifetime IPV 5 times higher than those whose partners never drank (Table 4).
Women whose partners consumed alcohol one to three days per week was also
significantly reported partner violence than those with partners in the non–drinking
reference group. Women who knew or believe that their partner might have had sex with
other women while they were in relationship were 4 times likely to report intimate partner
violence in. Likewise women who believed or suspected that their partners had sex with
commercial sex workers were also 2 times more likely to report intimate partner violence.
42
Table 4: Odds ratios for associations between IPV and risky behaviors, among women
attending VCT service in Addis Ababa city, 2005.
3 Male Partner 2.6 (1.6, 4.1) 2.0 (0.9, 4.1) 1.8 (0.9, 3.7)
4 Male Partners in life time 4.4 (2.3, 8.2) 3.1 (1.1, 8.4) 2.4 (0.9, 6.2)
≥5 Male in life time 5.7 (3.3, 9.9) 4.7 (1.7, 12.9) 3.5 (1.4, 8.6)
Casual partner, lifetime 2.7 (2.0, 3.7) 1.1 (0.6, 2.1) 1.4 (0.8, 2.4)
Ever Transactional Sex 1.8 (1.2, 2.6) 0.4 (0.2, 0.9) 0.7 (0.4, 1.2)
Never used condom 1.4 (1.0, 1.9) 1.2 (0.7, 2.0) 1.7 (1.0, 2.8)
Women Alcohol Use 4.5 (2.6, 8.0) 1.2 (0.6, 2.8) 1.1 (0.6, 2.2)
Women use of Khat 3.3 (2.0, 5.5) 2.6 (1.2, 5.3) 1.5 (0.8, 2.8)
Partner has sex with other women 8.5 (5.8, 12.) 4.2 (2.4, 7.4) 3.1 (1.7, 5.6)
Partner has sex with CSWs 10.9 (6.7, 18) 2.0 (1.0, 4.1) 2.1 (1.1, 4.1)
Partner Alcohol Use 1-3days/week 3.3 (2.3, 4.8) 1.8 (1.1, 3.1) 1.5 (0.9, 2.5)
Partner Alcohol Use ≥ 4days/week 14.1 (8.0, 24.9) 5.0 (2.3, 11) 2.4 (1.3, 4.7)
Partner Khat use ≤ 3days/week 1.8 (1.2, 2.7) 1.3 (0.7, 2.1) 1.2 (0.7, 1.9)
Partner Khat use ≥4 days/week 6.9 (3.8, 12.) 1.4 (0.6, 3.1) 1.7 (0.7, 3.4)
Childhood Sexual Assaults 1.9 (1.4, 2.7) 1.7 (1.1, 2.9) 1.3 (0.8, 2.1)
Forced First Intercourse 2.3 (1.6, 3.5) 2.0 (1.1, 3.6) 1.3 (0.7, 2.2)
Adult Rape by NIP 2.5 1.7, 3.5) 2.4 (1.4, 4.1) 1.7 (1.1, 2.8)
NB; The logistic regression models is adjusted for women age, education,
43
After adjusting for women’s age, education, current marital status, age difference and
duration of relationship with current or most recent partner, partner age and education,
women in violent relationship were significantly more likely to report not using condom
with current or most recent partner (OR=1.79, 95% CI: 1.18, 2.73) than women living in
non violent relationship, and significantly more likely to report to have had requested
current or most recent partner to use condom to avoid STDs (OR=1.98, 95% CI: 1.25,
3.12) and reported partner refusal to use condom (OR=2.56, 95% CI: 1.53, 4.28) (Data
not shown in tables). From women who reported medium or high frequency of physical or
sexual violence or two or more types of abuse, 238 (51.6%) women mentioned refusal of
We have shown that different forms of gender-based violence are associated with
between various types of violence and risk of HIV infection (Table 5 ) presents crude and
adjusted odds ratios for prevalent HIV infection associated with various types of intimate
partner violence.
The occurrence of a single type or only two types of intimate partner violence
(psychological, physical or sexual) was not associated with increased odds of being HIV
positive. But the co-occurrence of all three types of violence and increasing frequency of
physical and sexual violence associated with increased odds of HIV. Only small
proportion of women reported single or two types of violence and these women were less
44
likely to report higher frequency. When the number of types of partner violence
summarized nearly identical effect was observed for two or three types versus none or
one. Therefore, to explore the presence of possible dose response relationship between
which considered both the number of types of IPV and frequency of IPV. We summarized
more types of violence or medium to high frequency physical or sexual violence) versus
‘limited to no’ experience of IPV (Psychological violence only or low frequency of physical
or sexual violence). Our summary measure yielded a point estimate for HIV risk
45
Table 5: Odds ratios for testing HIV seropositive by experience of IPV among women
attending VCT services in Addis Ababa City, 2005.
Intimate Partner Violence, life time HIV Serostatus
Types of IPV Positive Crude OR Adjusted
No % OR 95% CI
Psychological violence (no versus yes) 314 (59.1) 2.5 (1.7, 3.6) 0.9 (0.5, 1.5)
Physical Violence (no vs yes) 212 (52.2) 3.6 (2.6, 4.9) 2.2 (1.3, 3.6)
Severe physical Violence (no vs yes) 138 (46.8) 2.9 (2.1, 4.0) 2.7 (1.8, 4.3)
Sexual Violence (no vs yes) 181 (48.9) 4.2 (3.1, 5.9) 3.4 (2.2, 5.4)
46
Table 6: Crude odds ratios for testing HIV seropositive by experience of various forms of
sexual violence among women attending VCT services in Addis Ababa City, 2005.
MULTI-LEVEL COMPARISONS
Childhood sexual Abuse
Limited IPV or No GBV** 33 19.0% 141 81.0% 1.0
CSA alone 12 19.7% 49 80.3% 1.1 (0.5, 2.2)
IPV + CSA 74 46.3% 86 53.8% 3.7 (2.3, 6.0)
Forced at first intercourse
Limited IPV or No GBV 33 19.0% 141 81.0% 1.0
FFI alone 6 16.7% 30 83.3% 0.9 (0.3,2.2)
IPV + FFI 59 50.4% 58 49.6% 4.4 (2.6,7.3)
Forced First Intercourse before age 18
Limited IPV or No GBV 33 19.0% 141 81.0% 1.0
FFI before age 18 alone 7 16.3% 36 83.7% 0.8 (0.3, 2.0)
IPV+ FFI before age 18 80 48.5% 85 51.5% 4.0 (2.5, 6.5)
Unwanted first intercourse at any age
Limited IPV or No GBV 141 81.0% 33 19.0% 1.0
Unwanted FI alone 56 87.5% 8 12.5% 0.6 (0.3, 1.4)
IPV + Unwanted FI 110 51.9% 102 48.1% 4.0 (2.5, 6.3)
Any CSA, Rape, FFI , UFI
Limited IPV or No GBV
CSA/Rape /FFI/UFI at any age only 17 15.7% 91 84.3% 0.8 (0.4, 1.5)
Intimate Partner Violence only 84 45.4% 101 54.6% 3.6 (2.2, 5.7)
IPV+ CSA/Rape /FFI/UFI at any age 129 46.7% 147 53.3% 3.8 (2.4, 5.9)
* include unwanted sexual contact, forced first sex and rape before age 15 years
** women reported no CSA, FFI, Adult sexual Assault (rape) by NIP or no IPV or reported only limited
IPV
47
Women experience of forced or unwanted first intercourse seems associated with increased
odds of HIV infection, with out considering the effect of IPV (table 6). However, neither child
sexual assault, forced first intercourse, nor adult sexual assault by non partners were
associated with increased odds of being HIV positive when they occurred in absence of
partner violence. Women who had experienced both forced first intercourse and partner
Except alcohol use all other women risk behavior such as number of partners, having casual
partners, engaging in transactional sex, never used condom were not associated with women’s
risk of HIV infection. Women who knew or believed that it is very likely their partners to have
had sex with other women while they were in relationship were 4.19 (OR=4.19, 95%CI; 2.37,
7.42) times more likely to have experienced IPV in lifetime (Table 4) and such women were
almost two times (OR=1.97 95%CI: 1.06, 3.65) more likely to have a positive test result (Table
7). Women who knew or believed it very likely for their partners to have had sex with
commercial sex workers were 2 times more likely (OR=2.01, 95%CI; 1.0, 4.1) to have
experienced IPV in their life time and in the past year (OR=2.07, 95%CI; 1.05, 4.06) and they
were also 2 times more likely (OR=2.20, 95% CI: 1.09, 4.42) to have positive test results.
Besides, women reported current partner use of alcohol four or more days per week were 5
times more likely (OR=5.01, 95%CI; 2.33, 10.8) to report IPV and 2 times more likely
48
Table 7: Multiple logistic regression model, showing association between risk behavior, and women
HIV serostatus a) after adjustment for socio-demographic characteristics, and risk behavior of women
and partner, b) ajested for socio-demographics, and Intimate Partner Violence among women
attending VCT services in Addis Ababa City, 2005.
HIV Positive Serostatus
Crude Odds Ratio Adjusted for SDCX& Adjusted for
a b
RB SDCX, RB & IPV
OR 95% CI OR 95% CI OR 95% CI
Broad IPV 3.99 (2.79, 5.69) 3.54 (2.28, 5.50)
Women risk behavior
Number of Male partners in life time
1 Male Partner 1.0
2 Male Partners 1.3 (0.8, 2.0) 0.9 (0.5, 1.6) 1.1 (0.7, 1.9)
3 Male Partners 2.0 (1.1, 3.4) 1.1 (0.5, 2.1) 1.5 (0.8, 2.9)
4 Male Partners 2.6 (1.3, 5.3) 1.2 (0.5, 3.0) 1.7 (0.7, 3.8)
5-8 Male Partners 2.1 (1.1, 4.1) 0.9 (0.4, 2.1) 1.4 (0.6, 3.2)
≥9 Male Partners 2.7 (1.1, 6.7) 1.8 (0.6, 5.4) 2.4 (0.8, 7.0)
Casual partner, lifetime 0.6 (0.3, 0.9) 0.8 (0.4, 1.3) 0.8 (0.5, 1.4)
Ever Transactional Sex 0.7 (0.4, 1.1) 0.5 (0.3, 0.9) 0.6 (0.4, 1.1)
Never used condom 1.4 (1.0, 2.0) 0.7 (0.4, 1.1) 0.9 (0.5, 1.3)
Women Alcohol Use
1-3days/week 2.0 (1.2, 3.3) 1.4 (0.8, 2.6) 1.9 (1.1, 3.4)
≥ 4days/week 4.7 (1.2, 18.9) 7.9 (1.5, 41.) 10.9 (2.3, 51)
Women use of Khat
≤ 3days/week 1.4 (0.9, 2.2) 1.6 (0.9, 2.78) 1.6 (1.0, 2.7)
≥4 days/week 2.0 (0.8, 5.3) 2.0 (0.6, 6.4) 2.6 (0.9, 7.6)
c
Partner Risk Behavior
Partner has sex with other women 1.0
Yes, I know /I think/may be 3.9 (2.6, 5.9) 2.0 (1.1, 3.7) 1.9 (1.0, 3.8)
Don’t Know 3.0 (1.8, 5.1) 2.0 (1.0, 3.9) 1.5 (0.7, 3.2)
Partner has sex with CSWs 5.1 (3.2, 8.0) 2.2 (1.1, 4.4) 2.1 (1.0, 4.6)
49
6. DISCUSSION
We assessed the prevalence and associations between newly diagnosed HIV infection and
experience of intimate partner violence, childhood sexual assault, forced first intercourse and
adult sexual assault by non partner among 743 women seeking VCT services in six health
institutions in Addis Ababa. After adjustment for the women’s social and demographic risk
factors and women’s risk behavior, intimate partner violence was associated with HIV
seropositivity. Childhood sexual assault, forced first intercourse and adult sexual assault by
From 743 women who participated in the study 263 (35.4%) tested positive for HIV. The
prevalence of HIV observed among study was higher than the prevalence observed in 2003
among VCT clients in Ethiopia (2). The possible reasons for higher prevalence may be due to
our eligibility criteria, which exclude women who have no steady sexual relationship seem
contributed for the elevated prevalence. To address ethical and safety issues, we excluded
those women who come to these VCT centers with their partners (couples), the majority for
premarital purpose, which probably represent lower prevalence groups (2). Even if we tried to
exclude those women who come to confirm positive result and women who previously worked
as commercial sex worker, it is difficult to assume genuine response from all participants.
This study also documented high prevalence of intimate partner violence. The life time and
recent intimate partner physical violence reported in this study was higher than rates reported
by women in rural district, Butajira- in south central Ethiopia (lifetime 54.6% versus 49% and in
the past year 41% versus 29%) (5). However the prevalence of life time and 12 months sexual
50
violence was lower than that reported from women in Butajira (lifetime 49.8% vs 59% and past
In our study, 127 (38.6%) of the women were suffered from all three types of abuse
(psychological, physical and sexual). This high prevalence and overlap of different types,
severity and frequency of violence is of particular concern, given previous studies linked IPV to
different health outcomes, both in short term and long term. Many studies documented that
those women who experienced violence faced ill health more frequently than women with out
such histories with regard to physical functioning, psychological well being like depression,
suicidal attempts, and gynecological disorders, infertility, pelvic inflammatory disease, and
unwanted pregnancy, unsafe abortion, miscarriage, still birth, premature labor and birth, fetal
injury and low birth weight (4). We also observed that most of the women who experienced
violence were still living with the abusive partners in an atmosphere of terror, treats and
achieve openness from respondents (4), we identified high prevalence of childhood abuse
(29.7%) and 10.9% of women were either raped or forced at their first sex before age of 15.
This finding is higher than the prevalence rates reported from a community survey in Ethiopia
(7% prevalence) (5), and comparable to the prevalence rates reported from a study conducted
in Debark, North West Ethiopia (6), where the prevalence of performed and attempted rape
were 8.8% and 11.5%, respectively. But, our finding was lower than the prevalence observed
51
among street adolescents in Addis Ababa city (15% prevalence within three months), (7) and
About 153(21.8%) study participants reported as they were physically forced or raped at their
first intercourse. Overall, 266(39.3%) women reported that their first sex was forced, coerced,
unwanted, or unexpected. This result is comparable with studies done in Cameron (37.3%)
Peru (40%), and with nine Caribbean countries (47%) (4). Although, no significant association
found between forced first intercourse and women seropositive status, these results can not be
considered definitive since reported frequency of such types of violence was low. However, the
reported high prevalence of forced first sex is of particular concern, since it is known to affect
the biological transmission of HIV and to lead to early pregnancy, affect sexual life course and
longer the risk and exposure to HIV(9). This finding has important implication for HIV
prevention, as it indicated the possibility that violence or coercion limit a young girls ability to
chose when to be sexually active and implement ‘abstinence’ as a means of HIV prevention
and condom use (since it is less likely to use condom at such incidents) (9).
While the evidence is not conclusive, some studies suggested that women in violent
relationship do not usually discus or request condom use. Our finding contradicts this
hypothesis, but support findings reported less condom use by women in abusive relationship
(9, 13, 19, and 21). In our study, women reported intimate partner violence were more likely to
request partners to use condom, though, the request refused by partners. Those women who
reported IPV also perceived their partner risk behavior or reported that their partners have had
extra marital or extra-relational sex with other women while they were in relationship. From
52
these women who reported IPV, 255 (44.3%) women mentioned refusal of sex as one of
reasons for intimate partner violence. A plausible explanation is that women who perceive their
male partner to be at significant risk of HIV infection may be reluctant to engage in sexual
relations with this partner or request condom use; this resistance may be met, in turn, with
physical violence or coercion into sex by the male partner (15). Overall, this study provide
evidence that, condom use, women’s refusal to sex perceived risky was affected by violence,
and they were not in a position to demand and implement mutual monogamous relationship as
During the pre test counseling from women currently married or living with a partner or women
in a regular partnership, 145 (26.7%) were either refused or were not sure to disclose their
serostatus. Moreover these women were significantly more likely to report intimate partner
violence and to have positive test result. This is a source of concern since disclosure of HIV
status is considered to be important for protecting sero-negative partner and for ensuring that
HIV positive individuals are able to access a range of services including prevention of mother
to child transmission, anti-retroviral treatment, and psychosocial support. We also know that
disclosure by HIV positive women to their sexual partners could enable couples to make
informed reproductive health choices such as seeking family planning services to reduce
unintended pregnancies or it could lead to changes in HIV risk behavior (9, 13, and 22).
forced first intercourse was associated with having multiple partner, casual partners,
transactional sex, and life time use of alcohol or Khat (13,17). Even though, the cross sectional
design limit us to establish temporality between early age experience of abuse and later
53
development of risk behavior, we expect that childhood abuse would have preceded alcohol or
Khat use or sexual interactions with many partners. Therefore our data support the hypothesis
of continuum of risk – that early abuse leads to increased risk behavior that may leads to HIV
infection (17).
Our study did not show an association between childhood sexual abuse and HIV sero-positive
status. However, childhood sexual assault or forced first intercourse, and adult sexual assault
by non-partners were associated with intimate partner violence and intimate partner violence
was associated increased number of male partners and women use of Khat. However, these
risk behaviors (having multiple partners and use of Khat) were not associated with increased
risk of HIV infection in our study participants. Taken together, the result suggests that
increased risk behavior subsequent to experience of violence is not the key mechanism by
Perhaps, the only risk factor for our study participants to have HIV seropositive test result was
alcohol use, but those women who reported use of alcohol were not at increased risk of
partner violence. Even if alcohol use associated with increased risk of HIV infection in our
study participants, this does not account for or explain the association between intimate
partner violence and HIV infection. Previous studies suggested that although experience of
violence may lead to risk behavior (or vice versa), and risk behavior to HIV infection, there is
also the possibility that women in abusive relationship to be directly infected by their abusive
partners, as some studies on men linked men’s use of violence to their own sexual risk taking
and to their own as well as their female partners risk of HIV (14). In our study participants,
54
accounting for women’s risk behavior and other potential confounders, partner violence,
partner’s sexual risk behavior and alcohol use were associated with both women’s risk of
Hence, the observed association between partner violent behavior, partner HIV risk behavior,
and HIV infection, suggest that abusive partners were more likely to take sexual risk or
practice risk behavior, and to have HIV. It seems that perpetration of intimate partner violence,
as captured in this study, serves at least partially as a proxy indicator of HIV risk in men, since
unprotected sex with a partner cannot result in HIV infection unless that partner is HIV positive
Therefore, our study supported the hypothesis that women in relationship with violent male
partners might be at increased risk to be directly infected with HIV by abusive partner or in
other words abusive partners seems to have HIV and impose risky sexual practice on partners
(13).The evidence with regard refusal to condom use also supports the hypothesis that
abusive partners impose risky sexual practice on partners and constrain women’s ability to
implement safe sexual behavior. Further support comes from the finding that refusal of sex by
the women was one of the most commonly cited reasons for IPV (cited by 280 (44.3%) women
(table 5).).
Results from a previous studies which showed a strong association between perceptions of the
partner's perceived HIV risks and sexual coercion, provide indirect support for this hypothesis
(16). For example, in India, a study showed that men who had extramarital sex were six times
more likely to report sexual abuse of their wives than men who remained faithful. Moreover,
men who reported an STI were 2.5 times more likely to report abuse of wives than men who
did not report an STI (14). The researchers concluded that abusive men were more likely to
55
engage in extramarital sex, acquire STI, and place their wives at higher risk for STI possibly
through sexual abuse. In Cape Town, South Africa, men who reported use of sexual violence
against intimate partners were nearly twice as likely to have multiple partners compared to
Finally, the reported high prevalence of forced sex or unwanted first sex indicated that violence
limited women’s ability to chose when to be sexually active and implement ‘abstinence’ as a
means of HIV prevention. More over women in abusive relationship are not in a position to
refuse unsafe sex, to negotiate condom use, demand and implement monogamous
relationship as a means of HIV prevention and to get benefits from using VCT services and by
disclosing test results. Overall, our research suggests that intimate partner violence play a very
important role in women’s risk of HIV infection and this study clearly shows the extent violence
These results point towards the extent to which intimate partner violence and early childhood
abuse, forced first intercourse are a realities in the lives of women, but several limitations must
be noted. The direction of causal associations between childhood sexual abuse, forced first
intercourse, adult sexual abuse by non intimate partner, risk behavior and acquisition of HIV
The prevalence intimate partner violence may have been underestimated as a result of under–
reporting of violence by respondents, given the culturally sensitive nature of this behavior and
56
the possible reluctance of many respondents to acknowledge its occurrence. However, several
features of the study and the setting are likely to have increased reporting, including the close
interaction and rapport between interviewers and respondents, the demonstrated expertise of
interviewers in eliciting sensitive information and the known privacy and confidentiality of
responses in institutions providing VCT services. We also relied on self reported data on
participation in HIV risk behavior. However, since women were unaware of their serostatus at
the time of the interview, we expect any under reporting to be non deferential, thus biasing our
results towards rejecting our hypothesis. More importantly we assessed the risk behavior of
current partner’s from women and it is difficult to relay on such information, but most likely it
We didn’t collect data about attendants who were not sampled for interview and we can not
determine how our study sample represents women attending VCT services in the city. Our
primary tool for assessing the experience of intimate partner violence (IPV) was the WHO
violence against women instrument, developed for a multi country study on women’s health
and domestic violence, which helps us to conduct standard assessment of violence and
based on women’s’ perception whether it was few times or many times, as well as relying on
women memory. Despite these limitations, our results provided comprehensive information to
date on the prevalence and nature of intimate partner violence among women attending VCT
services in Addis Ababa City. Our results indicated the magnitude of the problem of intimate
partner violence in Ethiopian urban setting and its overlap and interaction with HIV infection
57
8. CONCLUSIONS AND RECOMMENDATIONS
This study examined gender-based violence as a possible risk factor for HIV infection among
women attending VCT services in Addis Ababa City. More than over half of the 743 women
interviewed reported physical or sexual assault from intimate partners at some point during
their lives, and more than one third of them had experienced such violence in the past 12
months.
Broad lifetime experience of intimate partner violence in life time as well as in the past 12
months associated with significantly increased odds of HIV seropositivity, even after
adjustment for risk behaviors known to increase women risk to experience violence from
intimate partners. Women who reported child sexual assault, forced first intercourse or adult
sexual assault by non-partners also reported higher levels of risk behavior than those who not,
but were not at increased risk of being HIV seropositive. Taken together, results strongly
suggest that increased risk behavior subsequent to experience of violence is not the key
mechanism by which violence increases risk of HIV. Our data support the hypothesis that
women in abusive are likely to be directly infected by male partners, and that child sexual
assault and forced first intercourse seem increase HIV primarily through increasing the risk of
partner violence.
Identifying women who experience violence, providing support and considering implications of
these experiences for their health is seem very important for the health service. This study also
showed that women currently in violent relationship were significantly have no intention to
58
disclose their test results to their partners, possibly due to fear of violence. This suggest the
need to develop culturally appropriate screening instrument and VCT services need to
introducing in to VCT services and further research need to be undertaken to monitor its
introduction and assess the benefits for women. Research also needed on women experience
of first sex, men’s violent behavior and sexual risk taking behavior, condom refusal. Most
importantly, our findings point to the need for intervention efforts in HIV prevention that target
male sexual risk taking, condom refusal, and violent behavior, as well working towards broader
community and societal level transformations to challenge cultures of violence and male
59
9. REFERENCES
1. UNAIDS, WORLD REPORT ON HIV/AIDS EPIDEMIC, UNAIDS, 2003 : 9-10
2. MOH, DISEASE PREVENTION AND CONTROL DEPARTMENT, AIDS IN ETHIOPIA.
5TH ED. ADDIS ABABA, ETHIOPIA, MINISTRY OF HEALTH, 2004: 7-15
3. UNAIDS, RESOURCE CD, 2001
4. Krug EG et al. , eds, WORLD REPORT ON VIOLENCE AND HEALYH, GENEVA,
WORLD HEALTH ORGANIZATION 2002, 89-113, 149-175
5. YEGOMAWORK G, Negussie D, Yemane B, etl. WOMEN’S HEALTH AND LIFE EVENTS
STUDY IN RURAL ETHIOPIA, ETHIOPIAN JOURNAL OF HEALTH DEVELOPMENT,
SPECIAL ISSUE, ETHIOPIA, ISSN 1021-6790,
6. WORKU A, ADDISIE M. SEXUAL VIOLENCE AMONG FEMALE HIGH SCHOOL
STUDENTS IN DEBARK, NORTH WEST ETHIOPIA.DEPARTMENT OF COMMUNITY
HEALTH, FACULTY OF MEDICINE, ADDIS ABABA UNIVERSITY, ETHIOPIA. PMID:
12380887 [PUBMED - INDEXED FOR MEDLINE]
7. MOLLA M; ISMAIL S; KUMIE A; KEBEDE F, SEXUAL VIOLENCE AMONG FEMALE
STREET ADOLESCENTS IN ADDIS ABABA, ETHIOPIAN JOURNAL OF HEALTH
DEVELOPMENT. APRIL 2002; 16(2):119-128.
8. MULUGETA E, KASSAYE M, BEREHANE Y. PREVALENCE AND OUTCOMES OF
SEXUAL VIOLENCE AMONG HIGH SCHOOL STUDENTS, ETHIOPIAN MEDIACAL
JOURNAL 1998; 7-53
9. WORLD HEALTH ORGANISATION, SETTING THE, RESEARCH AGENDA, VIOLENCE
AGAINST WOMEN AND HIV/AIDS: GENDER AND WOMEN’S HEALTH MEETING
REPORT, SWITZERLAND, GENEVA, OCTOBER 2000; 23-25,
10. MOH, HIV/AIDS BEHAVIOURAL SURVEILLANCE SURVEY ETHIOPIA 2002 ROUND
ONE, MOH, Addis Ababa 2002, 85-87
11. Central Statistic Authority, 2001 ETHIOPIA DEMOGRAPHIC AND HEALTH SURVEY
JUNE 2000, CSA, ADDIS ABABA AND MARCO INTERNATIONAL INC, USA, 2002
12. MOH/HAPCO ETHIOPIAN MULTISECTORAL HIV/AIDS PREVENTION AND CONTROL
STRATEGY, MOH/HAPCO , ADDIS ABABA, 2005
13. KRISTIN L DUNKLE E.T. GENDER BASED VIOLENCE, RELATIONSHIP POWER, AND
RISK OF HIV INFECTION IN WOMEN ATTENDING ANTENATAL CLINICS IN SOUTH
AFRICA, THE LANCET VOL. 363. MAY 2004
14. WHO, VIOLENCE AGAINST WOMEN AND HIV/AIDS; CRETICAL INTERSECTIONS,
INTIMATE PARTNER VIOLENCE AND HIV/AIDS, WHO, INFOREMATION BULLETIN
SERIOUS, NUMBER 1, 2004
60
15. MICHAEL A. TOM L., FENG Z., FRED N.,etl. DOMESTIC VIOLENCE IN RURAL
UGANDA: EVIDENCE FROM A COMMUNITY–BASED STUDY UGANDA (FUENTE:
DECS, BIREME).
16. VAN DER STRATEN A, KING R, GRINSTEAO, etl. SEXUAL COERCION, PHYSICAL
VIOLENCE, AND HIV INFECTION AMONG WOMEN IN A STEADY RELATIONSHIPS IN
KIGALI, REWANDA JOURNAL OF AIDS BEHAV 1998 2: 61-73
17. MANDGE C. DOMESTIC VIOLENCE AND CHILDHOOD SEXUAL ABUSE IN HIV
INFECTED WOMEN AND WOMEN AT RISK FOR HIV, AMERICAN JOURNAL OF
PUBLIC HEALTH APPRIL 200, VOL 90 NO 4
18. WHO. WHO MULTI-COUNTRY STUDY ON WOMENS HEALTH AND DOMESTIC
VIOLENC: CORE QUESTIONNAIRE AND WHO INSTRUMENT-VERSION 9, GENEVA,
WHO 2000
19. UNAIDS, CD ROAM ON SEXUAL RISK NETWORKING, 2002
20. JESSIE MBWAMBW M, MICHEAL D, GAD P. etl. Women’s Barriers to HIV testing and
Disclosure, challenges for VCT, FORTCOMING AIDS CARE, Dare salaam Tanzania, 2002
21. WORLD HEALTH ORGANIZATION. WHO/WHD VIOLENCE AGAINST WOMEN: A
PRIORITY HEALTH ISSUE. GENEVA: WORLD HEALTH ORGANIZATION, 1997. WHO
DOCUMENT WHO/FRH/WHD/97.8#.
22. BLAIR BEADNELL, S.A. BAKER. D.M. MORRRISON, K.KNOR: HIV/STD RISK FACTORS
FOR WOMEN WITH VIOLENT MALE PARTNER, SEX ROLES:, APRIL 2000; 1445-1452
23. JEWKES RK; LEVIN JB; PENN-KEKANA LA GENDER INEQUALITIES, INTIMATE
PARTNER VIOLENCE AND HIV PREVENTIVE PRACTICES: FINDINGS OF A SOUTH
AFRICAN CROSS SECTIONAL STUDY.SOCIAL SCIENCE AND MEDICINE. 2003 JAN;
56(1):125-134.
24. WATTS C, KEOUGH E, NDLOVU M, KWARAMBA R. WITHHOLDING OF SEX AND
FORCED SEX: DIMENSIONS OF VIOLENCE AGAINST ZIMBABWEAN WOMEN.
REPRODUCTIVE HEALTH MATTERS 1998; 6:57–65.
25. COKER AL, RICHTER DL.: FREQUENCY AND CORRELATES OF INTIMATE PARTNER
VIOLENCE AND FORCED SEXUAL INTERCOURSE IN SIERRA LEONE. AFRICAN
JOURNAL OF REPRODUCTIVE HEALTH 1998; 2:61–72.
26. MAMAN S, CAMPBELL J, SWEAT MD, GIELEN AC. THE INTERSECTIONS OF HIV AND
VIOLENCE: DIRECTIONS FOR FUTURE RESEARCH AND INTERVENTIONS. SOCIAL
SCIENCE AND MEDICINE 2000; 50:459–78.
27. WHO, PUTTING WOMEN FIRST: ETHICAL AND SAFTEY RECOMMENDATIONS FOR
RESEARCH ON DOMESTIC VIOLENCE AGAINST WOMEN. GENEVA WORLD HEALTH
ORGANIZATION HTTP.WHO.INT/GENDER/DOCUMENTS/VAWETHICS/EN
61
HIGH RISK SEXUAL BEHAVIOR GENDER BASED VIOLENCE
VCT
• Reason
for
testing
• Outcome
WOMEN CHARACTERISTICS
• Demographics (age, religion, marital
status, primary residence, education)
• Economic (occupation, income)
• Social (women status – perceived gender
inequality (as measured by sexual
relationship power) reproductive history,
economic status relative to partner)
• Behavioral risk factors
o Chat use, problematic alcohol use
o Commercial/transactional sex
• Biological - History of STDs
PARTNER CHARACTERISTICS
o Demographics (age, religion, education,
occupation, income,
o Alcohol62and Chat use
63