ORIGINAL

Niger J Paed 2012; 39 (3) : 97 -104

Effect of maternal exposure to

intimate partner violence on

under-five mortality in Nigeria

Osuorah D

Antai D

Ezeudu C

Chukwujekwu E

DOI:http://dx.doi.org/10.4314/njp.v39i3.2

Accepted: 12th January 2012

violence were 0.77 times less likely

Abstract: Background: The under-

mortality rate in many developing

to lose a child under-5 compared to

women exposed to intimate partner

violence (HR=0.77 95%CI 0.64-

5

Osuorah _D (

) _A_n_t_a_i_D_,

countries has shown little or no

improvement over the years. Rav-

aged by war and poverty, violence

which is now a norm in most Afri-

can countries (including Nigeria) is

on the increase and has condensed

into most families with women and

children bearing the major brunt of

this violence.

Ezeudu C, Chukwujekwu E

Department of Paediatrics,

Nnamdi Azikiwe University

Teaching Hospital,

Nnewi,

Anambra State.

0

.81). Decision making autonomy

in family activities significantly

affected loss of a child under-5 in

the face of IPV. Similarly women

not exposed to IPV were 1.74 times

more likely to use maternal and

child health services compared to

exposed mothers (OR=1.74 95% CI

Email: chidi.osuorah@yahoo.com

Aim: Effect of maternal exposure to

intimate partner violence on under-

1

.65-1.83). Age of mother, educa-

tional status of mothers, social

class, occupation, marital status,

access to media and decision mak-

ing autonomy were retained as im-

portant maternal predictors of use

of maternal and child health ser-

vices when exposed to IPV in mul-

tivariate analysis (p<0.05).

Conclusion: Intimate Partner Vio-

lence has a significant effect on

under-5 mortality. Therefore tack-

ling this social menace will not only

reduce the effect on child mortality

but also address the ill societal ef-

fect that results from family col-

lapse following IPV.

5

Methods: Data from nationally

representative sample of mothers

mortality in Nigeria.

(

aged 15-49 years) was obtained

from the 2008 Nigeria Demoraphic

and Health Survey. Cox regression

and multiple logistic regressions

were used to identify and examine

the association between maternal

exposure to intimate partner vio-

lence and under-5 death and use of

maternal and child health services

after controlling for potential con-

founding factors.

Results: The prevalence of intimate

partner violence (IPV) in Nigeria is

3

4.9%. This may be lower than the

Key words: Intimate partner vio-

lence, under-5 mortality, use of

maternal and child health services,

Nigeria

actual prevalence due to under re-

porting of cases of IPV in most

developing countries. Women not

exposed to intimate partner

Introduction

Intimate partner violence is defined as any act of physi-

cal, sexual and emotional abuse by a current or former

partner whether cohabitating or not. The Centre for

3

Despite decline in child mortality rate globally the under

-

tries . While the major killers of children in Africa have

been identified , the effect of social determinants of

health on this increasing mortality rate remains under

studied especially in the African setting. Intimate partner

violence (IPV) is one of these social determinants which

adversely affect the capacity of the mother to care for

the child. This is particularly true in Africa where male

dominance in relationships and family is an unquestion-

able norm.

five mortality rate is still an issue in most African coun-

Disease Control prevention (CDC) also defined it as a

serious, preventable public health problem characterised

by physical, sexual or psychological harm by a current

1

2

4

or former spouse or partner . According to CDC, three

main types of intimate partner violence exists and they

include - physical abuse when a person hurts or tries to

hurt a partner by hitting, kicking, burning or other physi-

cal force. Sexual abuse forcing a partner to take part in a

sexual act when the partner does not consent and emo-

9

8

1

6

tional abuse is threatening a partner or his /her posses-

sion or loved ones or harming a partner’s sense of worth

e.g. stalking, name calling, intimidation or not allowing

a partner see friends or family.

human right issue but as a public health issue. This is

because IPV not only affects the health of the assaulted

woman but also affects the health and psychological

stability of her child leading to a vicious cycle of domes-

tic violenc₁e₇ that accounts for most of the ills of the soci-

ety today. Several studies outside Africa have docu-

mented the existence of a strong l₁in₈_,k₁₉between IPV and

childhood morbidity and mortality.

Intimate partner violence is the third highest cause of

5

death among people 15-44 years of age, and the most

common form of violence against women. Its negative

effects on women’s health are serious enough to be rec-

ognized as a_-₈public health crisis with extensive effects

Children who witness such violence are at greater risk of

being affected in various aspects of their life including

physical, behavioural, cognitive development and social

adjustment which ar₀e all necessary determinants for the

6

on children. In most cases of IPV these several types

occur together usually starting from emotional abuse

4

then progressing to physical or sexual assault . Although

2

women can be violent in a relationship, bulk of the

global burden of IPV are borne by women, this is par-

ticularly true in Africa where women are emotionally

survival of a child. In a study done in the North of In-

dia it was found that the risk of child death during peri-

natal and neonatal period is twice higher in women who

experienced domestic violen₁ce during pregnancy com-

2 22

pared to those who did not. Another study in Uganda

associated IPV with childhood diseases like fever, cough

and fast breathing.

9

and economically dependent on men . Owing to cross-

cultural differences in expected gender roles, IPV varies

between geographical locations.

A global lifetime prevalence of 10-70% has be₀en stated

1

for women in marriage or current relationship while in

Several studies have proved that a major link between

maternal exposure to IPV and child mortality₂₃i_,s₂₄poor

The

pathway between IPV and child mortality is multi-

faceted and complex. Some linkages between IPV and

child mortality have been proposed, some of which in-

cluded maternal depression and direct physical injury

assault on the child. This study probes further other pos-

sible relationship between IPV and child mortality using

2008 Nigeria DHS data.

sub-Saharan Africa a life time prevalence of IPV is be-

1

tween 20-71% . This has been argued to be non repre-

sentative due to the low reporting trends of IPV in many

care from mothers who are mentally stressed.

1

2

sub-Saharan African countries. According to W.H.O

almost 50% of women who die of homicide are killed by

either their current or former partners, while in some

countries it can be as high as 70%. Also 25% of women

will ex₂ perience violence by intimate partner in their life

1

time. While domestic violence is a universal problem;

it is a problem of extreme magnitude in less developed

countries. A recent survey in Uganda and Bangladesh

reported that more than 80% and 94% of women sur-

veyed respectively had experienced physical, sexual or

psychological violence a₃t some point in their marriage

Conceptual framework

2

5

A conceptual framework adapted from Emily Rico et

al was modified ( figure 1) and used to hypothesize the

assumed association between IPV and under-five mor-

tality. It outlines how IPV could lead to child mortality

and how other confounding variables could possibly

play a role in enhancing or reducing these associations.

Pathway1 shows how IPV through maternal factors lead

to under- five mortality. These health problems include

mainly the phy₃sical health effect and the psychosocial/

1

or intimate relationship. In many parts of Africa, IPV

1

4

is a justifiable punishment accepted as a societal norm.

This acceptance is enhanced by the increasingly armed

conflicts in most African countries, lawlessness and al-

most₁₄absent interference of police in intra marital af-

fairs. A study in Nigeria revealed that 47% and 42% of

women in Niger-delta (South-south) and the rest of the

country respectively justified IPV for one reason or the

other. It also showed that 81% and 83% of these women

had no a₁u₅tonomy over their own health or that of their

children.

2

mental effect. The poor mental state leads to poor

breast feeding, substances abuse, poor use of

maternal and child health services, etc. All these prob-

lems tend to reduce the IPV-exposed mothers ability to

carter for herself and for the child leading to poor child

attention, child malnutrition and increased morbidity.

Due to the damaging effect of IPV on the society it is

now viewed by international organisations not only as a

Fig 1: Conceptual framework between association of IPV and under-5 death

Exposure variable

Intimate partner violence

Association

Maternal health

Antenatal care

Use of maternal and

child services

Path 1

Path 3

Cofounders

Maternal age

Occupation

Marital status

Wealth index

Women empowerment

Path 2

Direct Effect on Child

Breast feeding

Child immunization

Outcome variable

Under5 mortality

9

Pathway 2 shows IPV can also be related to under -five

mortality via direct effect on the child. These could oc-

cur prenatally leading to poor pregnancy outcome like

ante-partum haemorrhage, spontaneous abortion, prema-

turity, and still births due to combined effect of trauma

on the abdomen or activity of stress hormone on the

pregnancy. Postnatally, children are advertently victim

of domestic violence either accidentally by being used

as a shield or when they try to intervene due to fear, or

non-accidentally due to battering by one of the spouse

out of retaliation or reaction for the domestic violence. It

was noted in the USA that 40-60% of men who abuse

women also abuse their childr₂e₆n and 27% of violent

homicide victims were children.

In Pathway 3 social and demographic factors which in

themselves do not cause under-five mortality directly

but mediates the association between IPV and child

mortality. Thus the presence or absence of these factors

enhance or reduce the effect of IPV on under-five mor-

tality. In this study these factors will be treated as co-

founders.

interview between participant in the classroom, and

practice interviews w₉ith real respondents in area outside

2

NDHS sample point .

Exposure variable

The logistic regression analysis included the two broad

predictor variables namely-

•

Exposure to IPV was assessed using the DHS inti-

mate partner module which is based on a modified

p₂₉revious version of the conflict tactics scale (CTS).

Based on this scale respondent will be classified

as “exposed” and “non-exposed”. IPV referred to

any exposure to one or several of the types of IPV

i.e. physical, sexual or emotional acts of violence

against wome₂n₉ by a current, former husband or inti-

mate partner.

•

Social and Demographic Variables- These charac-

teristics were chosen as they were found to be im-

portant ₁p₅r_,₂e₂d_,₂i₅ctors of child immunization in previous

studies.

This variable include age of mothers,

marital status, educational status, occupation,

wealth class, decision autonomy, access to media,

literacy level and number of living children. These

variables wer₂e₉re-coded into categories using DHS

primary data.

Methods

Study area

Nigeria is in the West African sub-region, lying between

Niger in the North, Chad to the North-East, Cameroun

in the East and Benin to the West. The 2006 population

and housing census puts Nigeria’s population at

Outcome variable

• The outcome variable was under-5 mortality de-

fined as a child dying between birth and the fifth

birthday. Under-5 mortality was estimated for the 5

years preceding the survey. All deaths among chil-

dren ≤ 60 months were regarded as cases while

those >60 month were regarded as non cases. The

under-5 mortality was chosen because it reflects the

impact of social, economic and environmental cir-

cumstances as well as other causes of death on in-

fants, toddler and children and more so 90% of

death among c₃h₀ildren < 18 years occurs in the first

1

40,431,790_t_hmaking it the₂m₇ ost populous nation in Af-

rica and 14 in the world. Nigeria is made up of 36

states and a federal capital territory, grouped into six

geopolitical regions. There are 774 constitutionally rec-

ognised local g₈ overnments and about 374 identifiable

2

ethnic groups. This survey was conducted in all states

(

including the Federal Capital Territory) within the six

regions of Nigeria.

Study design

5

years of life.

This is a cross-sectional descriptive- analytical study.

The study was conducted between July 2011 and De-

cember 2011 using data from the 2008 edition of DHS

survey, which is a nationally-representative probability

sample, using a stratified two-stage cluster sampling

design consisting of 888 clusters, 286 urban and 602

rural areas. In all, a nationally representative sample of

•

The use of maternal and child health services

MCHS) such as antenatal care, place of delivery

(

and postnatal care like immunisation and child fol-

low-up clinics were outcomes that were measured

in women whether or not they were exposed to IPV.

The variable was created from DHS data on ante-

natal care visit, place of delivery and child having

ever been immunized. The MCHS variable was then

computed by summing up of the variable. Two

groups were formed, 0 for those who had never

used MCHS and 1 for people that have used MCHS

at least once or more.

3

6,298 household was selected for the 2008 NDHS sur-

vey. From these household 34,596 women were eligible

for interview out of which 33,385 were succes₉sfully

2

interviewed yielding a response rate of 96.5%. Data

collection was by 368 trained people (37 supervisors, 37

editors, 37 quality control interviewers, 152 female in-

terviewers, and 74 male interviewer, reserve interview-

ers, etc). Training for the field staff who conducted the

state interviews was done during a three week period in

May-June 2008. The training consisted of instruction

regarding interviewing techniques and field procedures,

a detailed review of item on the questionnaire, mock

Ethical Consideration

The survey procedure and instruments for DHS for Ni-

geria was ethically approved by the Ethics Committee of

the Opinion Research Corporation (ORC) Macro Inter-

national Inc, Calverton, USA, and by the National

1

00

Ethics Committee in the Federal Ministry of Health of

Nigeria. Informed consent was obtained from all partici-

pants prior to participation in the₉ survey, and collection

regression (model 1) showed that mothers not exposed

to IPV were 0.69 times less likely to lose a child under-5

than those exposed to IPV (HR 0.69 CI 0.62-0.78). In

other words those exposed to IPV were about 1.5 times

more likely to lose a child under-5 compared to mothers

not exposed to IPV. In model 2, after correcting for co-

founders using multivariate Cox regression, the likeli-

hood of losing a child when mother is not exposed to

IPV remained significant (HR 0.77 CI 0.64-0.81) com-

pared to when she is exposed to IPV. Stratified analysis

showed that women with no decision making autonomy

were 1.36 times more likely to lose a child under-5 com-

pared to those with decision making autonomy in events

of IPV. See figure 2

2

of information was confidential.

Analysis

The Predictive Analytics Software (PASW) statistical

package version 19.0 (formerly called SPSS) was used

for data analysis. The Pearson chi-square (χ2) test was

used to study the differences in proportion between the

independent variables and immunization practice. The

largest category in each predictor variable was used as

the reference category. For all statistical tests performed,

it was ensured that the assumptions for carrying out

these specific tests were met. Statistical significance was

set at p-value < 0.05. Results are presented using per-

centages, Odds Ratios and 95% CIs where appropriate.

Fig 2: Survival plot from Kaplan-Meier survival analy-

sis and number of censored observation for women

exposed and those not exposed to IPV.

Results

The mean age of mothers surveyed is this study was

2

8.7years with an average of 2.54 children per woman.

The overall prevalence of IPV for women from age of

5 in this study was 34.9%. Cross tabulation shows that

1

women with no education had the highest rate of death

of a child under-5 though this was not statistically sig-

nificant (p=0.1025) while those 20-29 had a signifi-

cantly higher proportion of child death under-5 years

Censored

Observa-

tion

(

p=0.001) compared to women in other age category.

Expectedly women in the poor wealth category had the

most under-5 death (p=0.034) while women who are

employed and skilled had most child death in the occu-

pation category; this proportion was however not sig-

nificantly different from other occupational category

(

p=0.889). Respondents who were not married, those

with no access to media and respondent with no decision

making autonomy had higher rate of under-5 death p

=

0.3688, 0.0951 and 0.0870 respectively. These were

Respondents with no education used MCHS the least

(p˂0.001) while similarly mothers more than 40 years

old used MCHS the least (p˂0.001). Women in the poor

wealth category, women with no form of employment

and those not literate used MCHS less often compared

to others in their respective category (p˂0.001). Lastly

women with access to media and decision making

autonomy used MCHS more than those with no access

to media and no decision making autonomy (p˂0.001).

however not statistically significant. Finally under-5

death was not significantly different in women with

more than 4 children compared to those with less num-

ber of children (p=0.0977).

Respondent not exposed to IPV had a total of 867 under-

5

those exposed to IPV, with an average age at death of

death compared to 499 under-5 deaths recorded by

2

3

period for those not exposed to IPV compared to 55.9%

in those exposed to IPV. Similarly 26.2% and 42.1% of

under-5 death occurred in the infant and early childhood

period in respondents not exposed to IPV compared to

.01±0.85 years and 1.67±0.87 years respectively.

1.7% of the under-5 death occurred in the neonatal

The use of MCHS was significantly better in women not

exposed than those exposed to IPV (p˂0.001). Logistic

regression in model 1 showed that women not exposed

to IPV had 1.65 times more likelihood than those ex-

posed to use MCHS (OR 1.65, CI 1.58-1.72). Correcting

for confounders in model 2 showed that the association

still remained significant with likelihood increasing to

1.74 (OR 1.74, CI 1.65-1.83). Women who were 40

years and older were 1.71 more likely to use MCHS

when exposed to IPV than those 15-19 years (OR 1.71

CI 1.28-2.28) while those between 20-29 and 30-39

2

1.2% and 22.8% respectively in those exposed to IPV.

The proportion of deaths within this categories were

statistically significant (p<0.001). Even though respon-

dent not exposed to IPV proportionately had higher un-

der-5 death compared to those exposed, this proportion

was not statistically significant (p=0.112). Bivariate Cox

1

01

were 1.37 times more probable to use these services than

the reference category (OR 1.37, CI 1.06-1.63). Respon-

dents in the rich and middle wealth category were 4 and

respectively more likely to use MCHS compared to the

unemployed women [(OR 1.57, CI 1.34- 1.84) and OR

1.37, CI 1.21- 1.55)]. Similarly respondent who had no

access to media and those without decision making auton-

omy were 0.8 and 0.65 times less likely to use MCHS

than those with access to media (OR 0.81, CI 0.72- 0.92)

and decision making autonomy (OR 0.65, CI 0.58- 0.73)

respectively. Finally women who were not married were

0.66 times less likely to use MCHS compared to those

who are married (OR 0.66, CI 0.41- 0.93). See table 1.

2

times respectively more likely to use MCHS than re-

spondents in the poor wealth class [(OR 4.78, CI 3.38-

.47) and (OR 2.66, CI 2.24- 3.15)]. Employed women

5

had more likelihood to use MCHS compared to those

unemployed. Women within the unskilled and skilled

employed category being 1.57 and 1.37 times

Table 1: Crude and adjusted hazard/odd ratio for under-5 death/use of MCHS and IPV variables

Variable

Death of_*under-5 child

Model 1

Use of M_*CHS

Model 1

**

Model 2

HR (95% CI)

OR (95% CI)

Exposure IPV

No

+

0.69(0.62, 0.78)

0.77(0.64, 0.81)

1

1.65(1.58, 1.72)

1

1.74(1.65, 1.83)

1

Yes

1

Age of mother

<

19yr

1.13(0.87, 1.46)

1.08(0.92, 1.28)

1.12(0.95, 1.33)

1

+

2

3

0-29

0-39

1.37(1.06, 1.63)

1.37(1.07, 1.63)

1.71(1.28, 2.28)

>

40

Education

None

Primary

Higher

1

1.05(0.80, 1.38)

0.66(0.36, 1.22)

1

2.66(2.24, 3.15)

4.78(2.24, 8.90)

+

Wealth index

Poor

Middle

1

0.76(0.63, 1.37)

0.87(070, 1.09)

1

2.36(2.01, 2.78)

4.30(3.38, 5.47)

+

Rich

Occupation

None

Unskilled

Skilled

1

0.73(0.43, 1.25)

1.06(0.94, 1.19)

1

1.57(1.34, 1.84)

1.37(1.21, 1.55)

+

Literacy

No

1

Yes

1.56(0.89, 2.74)

1.25(0.71, 2.19)

Marital status

No

Yes

+

0.67(0.40,1.18)

1

0.66(0.41,0.93)

1

Media access

No

Yes

+

0.90(0.80, 1.01)

1

0.81(0.72, 0.92)

1

Decision autonomy

No

Yes

+

1.36(01.21, 1.54)

1

0.65(0.58, 0.73)

1

No of children

0

1

2

4

1.21(0.90, 1.63)

1.03(0.84, 1.26)

1.08 (0.925, 1.26)

0(0.00, 0.00)β

0.91(0.80, 1.04)

1.03(0.93, 1.15)

-3

1

(

* ) Crude Hazard ratio/odd ratio (**) Adjusted Hazard ratio/odd ratio (+) statistically significant (β) Not computable

1

02

Discussion

had better usage of MCHS than those without decision

making autonomy and without media access. The reason

for this is not far-fetched. A woman who has say in the

decision of her family and has access to media is more

likely to be better empowered than one without such

privileges. This power puts her in charge of decisions

regarding health issue for herself and the child which in

turns translates to less mortality for the child.

The finding of the survey puts the prevalence rate of IPV

in Nigeria at 34.9%. Th₁is falls within the life time preva-

1

lence of IPV in Africa put at 20-71%. A study done in

2

5

Kenya by Rico et al found a prevalence rate of 42.6%.

The prevalence rate found by this study is believed to be

lower than the actual prevalence in Nigeria. This is be-

cause in Nigeria like in many developing coun₄tries, IPV

1

which is regarded as a justifiable punishment is not or

The fin₁d₈_,i₁n₉g_,₂s₁_,o₂f₃ this study though collaborated by other

should be interpreted in the light of

under reported in most cases and regarded as a family

issue.

studies

some limitations. Firstly, being a cross-sectional study,

the temporality in causality could not be ascertained

since both intimate partner violence and under-5 mortal-

ity could precede one another. Since some of the vari-

ables used in this study were re-categorised using DHS

primary data, errors in classification might have oc-

curred giving rise to misclassification and interpretation

errors. Lastly most confounding variables had skewed

distribution in the number of respondents in various

categories, which made regression analysis impossible,

inconsequential or resulting in incorrect associations.

Finding from this study showed that women exposed to

violence lost a child at earlier ages of the child life when

compared to those not exposed. This is particularly im-

portant in Nigeria and other developing countries where

majority of the under-5 death occurs in the first year of

3

1

life with the highest risk within the first 28 days . Dur-

ing this period care of the mother is critical to survival of

the child therefore experience of IPV by mother means

higher chances of neonatal and infant death. Analysis

also indicates that women who experienced IPV were 1.5

times more likely to lose a child under-5 years than those

who were not exposed. This is expected as these women

not only suffer physical injury but also social, emotional

and psychological trauma which affects their capacity as

mothers to cater for their children. Furthermore women

exposed to IPV were more likely to separate temporary

from their husbands after spousal abuse leaving the care

of the child to the man who is usually less skilled in the

care of children. For those who stay on after IPV, re-

duced financial support for them from their partners

Conclusion and Recommendation

Despite all the short comings, this analysis supports

evidence that IPV has an influential role in under-5

mortality and use of maternal and child health services

in Nigeria and by extension to developing countries

where child mortality is still high. This indicates that

violence against women represent an important public

health concern not only for women but also for child’s

survival. Therefore tackling the increasing incidence of

intimate partner violence in families of developing coun-

tries will go a long way to impact on the ever increasing

childhood mortality in many c₄ountries in Africa, if the

millennium development goal is to be achieved in the

continent by 2015. It is recommended that women

should be empowered by ways of education and

employment in Nigeria as the factors were seen not only

to reduce under-5 death in face of IPV and but also

encouraged the use of MCHS. Also early marriage

should be discouraged by effective legislation as lower

maternal age was seen to negatively affect child care and

use of MCHS. Further studies to further collaborate

these findings, to evaluate other possible pathways not

captured in this study and exploration of causality and

temporality is also recommended.

(

especially those who are unemployed and completely

dependent on their husbands for finances) will definitely

translate to poor use of MCHS and invariably higher

chances of death of their under-5 children. The findings

2

1,

o₂₃f this study are in line with the results of other studies

which showed a significant association between IPV

and child mortality.

Higher educational level, higher wealth index and better

employment status were also seen to significantly affect

the use of MCHS. These variables which reflect the so-

cioeconomic class of a woman empowers the woman

better in seeking healthcare for herself and her child.

Because they are less dependent on the man financially

and psychosocially they tend to act more independently

from their husbands during spousal crises. This also pos-

sibly explains the lower rate of under-5 death experi-

enced by women of higher education level, rich class and

employed status found in this study.

Authors contributions

The findings that married women uses MCHS more than

unmarried women is worthy of note. Apart from the fi-

nancial and emotional support enjoyed by most married

women in seeking health services for themselves and

their children, the belief in most African countries that

having a child outside marriage is an abomination usu-

ally discourages mo₂st single mothers from visiting

Osuorah D : Conceptualised the frame work, developed

work and data analysis.

Antai D:

Ezeudu C :

Supervised the work,

Literature review

Chukwujekwu E: Methodology, discussion and

Recommendations

3

healthcare facilities in Nigeria. Finally the study

Conflict of interest: None

Funding : None

showed that women who have a say in the running of the

family activities and have access to media

1

03

Acknowledgements

used in this study and the Nigeria Federal Ministry of

Health and National Population Commission for their ethical

clearance. The contents of this study are solely responsibility

of the authors and are in no way representative of the official

views of the ORC.

We thank the officials of the Opinion research company

(ORC) owners of the Demographic and health survey for their

kind assistance in permitting us to use the data

References

1

0. Ellsberg MC, Peña R, Herrera A,

21. Saifuddin A, Koenig M, Stephen-

son R. Effect of domestic violence

on perinatal and early childhood

mortality; evidence from North

India. Am J Pub Hth 2006; 96 (8)

22. Karamagi CA, Tumwine JK, Tylle-

skar T, Heggenhougen K. Intimate

partner violence and infant mor-

bidity: evidence of an association

from a population-based study in

eastern Uganda in 2003. BMC

Pediatr 2007; 7:7-34.

23. Campbell, JC, Lewandowski, LA.

Mental and physical health effects

of intimate partner violence on

women and children. Psych Clinics

of N Am, 1997. 20: 353–74.

24. Cascardi M, O’Leary K, & Schlee,

K. Co-occurrence and correlates of

posttraumatic stress disorder and

major depression in physically

abused women. J Fam Viol, 1999.

14:227–50.

25. Rico E, Fenn B, Abramsky T,

Watts C. Associations between

maternal experiences of intimate

partner violence and child nutrition

and mortality: findings from

1

.

Black RE, Saul S Morris, Jennifer

Bryce. Why and where are 10

million children dying every year?

Department of International

Health, John Hopkins Bloomberg

School of Public Health, Baltimore

USA 2003; 361(9378): 2226-34.

UNICEF article on Under-five

mortality by cause: Child Health

Epidemiology Reference Group

Liljestrand J, Winkvist A. Wife

abuse among women of childbear-

ing age in Nicaragua. Am J Public

Health. 1999; 89(2):241-4.

1. Jewkes R, Levin J, Penn-Kekana

L. Risk factors for domestic vio-

lence: findings from a south Afri-

can cross-sectional study. Soc Sci

Med. 2002; 55(9):1603-17.

2

3

.

2. World Health Organisation report

on violence against women, 2010.

Available at www.who.int/

(CHERG). http://www.unicef.org/

health/index_problem.html.

Krug EG, Mercy JA, Dahlberg LL,

Zwi AB. The world report on vio-

lence and health. Injuries and Vio-

lence Prevention Department, Non

Communicable Disease and Men-

tal health, WHO, Geneva, Switzer-

land 2002; 360: 1083-8.

Saltzman LE, Fanslow JL, McMa-

hon PM, Shelley GA. Intimate

partner violence surveillance: uni-

form definitions and recommended

data elements, version 1.0. Atlanta

violence-injury-prevention

(Assessed 25-07-2011)

3. ICRW. Intimate partner violence:

High cost to household and com-

munities. International centre for

research on women, Washington

D.C, 2009.

4. Kristen A. Guns and roses: Gender

and armed violence in Africa.

UNDP 2007. http://

4

.

genevadeclaration.org/pdfs/

gender.pdf

(GA): Centre for Disease Control

5. Diddy antai, Antai J. Collective

violence and attitudes of women

toward intimate partner violence:

Evidence from the Niger Delta.

BMC International Health and

Human Rights 2009, 9:12

6. Krantz Gunilla, Farcia-Moreno

Claudia: Violence against women.

J Epidemiol Community

Health 2005; 59: 818-821.

7. Levinson D: Family violence in

cross-cultural perspec-

tive. Thousand Oaks, CA: Sage

Publications; 1989.

8. Silverman JG, Gupta J, Decker

MR, Kapur N, Raj A. Intimate

partner violence and unwanted

pregnancy, miscarriage, induced

abortion, and stillbirth among a

national sample of Bangladeshi

women. Brit J Obstet Gyne 2007;

and Prevention, National Centre

for Injury Prevention and Control;

2

002. Available at: http://

Demographic and Health Surveys

in Egypt, Honduras, Kenya, Ma-

lawi and Rwanda. J Epidemiol

Comm Hlth. 2010.

26. American Psychological Associa-

tion. Violence and the Family:

Report of the APA Presidential

Task Force on Violence and the

Family, 1996.

www.cdc.gov/ncipc/pubres/

ipvsurveillance/intimate.htm

Krug EG, Dahlberg LL, Mercy JA,

Zwi AB, Lozano R. World report

on violence and health. Geneva:

W.HO 2007

5

6

.

1

Grisso JA, Schwarz DF,

Hirschinger N, Sammal M,

Brensinger C, Santanna J et al.

Violent injuries among women in

an urban area. New Eng J Med

27. United Nations World Population

Prospects: 2011 revision.

28. CIA world fact book. People and

ethnic group: Available at http://

www.cia.gov/library/publications/

the-world-factbook/geos/ni.html.

(Accessed -24-10-2011)

29. Macro International. Demographic

and Health Surveys (DHS), 2008.

Available at http://www. meas-

uredhs.com (Accessed 27-07-

2011).

1

999; 341: 1899-1905.

7

8

9

.

Kyriacou DN, Anglin D, Taliaf-

erro E, Stone S, Tubb T, Linden

JA et al. Risk factors for injury to

women from domestic violence

against women. New Eng J Med

1

999; 341: 1842-98.

1

14(10):1246-52.

Bacchus L, Mezey G, Bewley S.

Domestic violence: prevalence in

pregnant women and association

with physical and psychological

health. Eur J Obstet Gyne Rep Biol

1

9. Bogat GA, DeJonghe E, Leven-

dosky AA, Davidson WS, von Eye

A. Trauma symptoms among in-

fants exposed to intimate partner

violence. Child Abuse Negl. 2006;

2

004; 113: 6-11.

3

0(2):109-25.

Heise L, Ellsberg M, Gottemoeller

M. A global overview of gender-

based violence. Int J Gynaecol

Obstet. 2002; 78 Suppl 1:S5-14.

2

0. Adam CM. The consequence of

witnessing violence on children

and implication for the family

counsellor. Fam J 2006; 14: 334-

4

1

04

3

0. The United Nations Children’s

Fund. UNICEF fast-fact on chil-

dren, 2008. Available at http://

www.unicef.org/esaro/

31. UNICEF Eastern and Southern

Africa. Maternal and child health.

Young child survival and develop-

ment, 2009. Available at

www.unicef.org/

esaro/5497_maternal_newborn_he

alth.html

32. Treffer P et al. Care for adolescent

pregnancy and child birth, 2001

cited in Heidi W, Emelita L, Heidi

T. Adolescent use of maternal and

child health service in developing

countries. Inter fam plan persp,

2006, 32(1):3

factsonchildren_5775.html.