Violence Against Women: Data on Pregnant Victims and		 
Effectiveness of Prevention Strategies Are Limited (15-MAY-02,	 
GAO-02-530).							 
                                                                 
The Violence Against Women Act funds programs that shelter	 
battered women, training for law enforcement officers and	 
prosecutors, and research on violence against women. Available	 
data on the number of pregnant women who are victims of violence 
are incomplete and lack comparability. There is no current	 
national estimate of the prevalence of violence against pregnant 
women. Available estimates cannot be generalized to all pregnant 
women, and little information is available on the number of	 
pregnant homicide victims. Health and criminal justice officials 
have designed multiple strategies to prevent violence against	 
women, but their effect is unknown. Strategies to prevent	 
violence against pregnant women are similar to those to prevent  
violence against all women and include include public health	 
efforts to prevent  violence in the first place, intervention	 
activities that identify and respond to violence after it occurs,
and criminal justice strategies that focus on incarcerating or	 
rehabilitating batterers.					 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-02-530 					        
    ACCNO:   A03329						        
  TITLE:     Violence Against Women: Data on Pregnant Victims and     
Effectiveness of Prevention Strategies Are Limited		 
     DATE:   05/15/2002 
  SUBJECT:   Data integrity					 
	     Homicide						 
	     Pregnancy						 
	     Strategic planning 				 
	     Women						 
	     Abuse						 
	     CDC National Vital Statistics System		 
	     CDC Pregancy Risk Assessment Monitoring		 
	     System						 
                                                                 
	     FBI Uniform Crime Reporting Program		 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-02-530
     
Report to the Honorable Eleanor Holmes Norton, House of Representatives

United States General Accounting Office GAO

May 2002 VIOLENCE AGAINST WOMEN

Data on Pregnant Victims and Effectiveness of Prevention Strategies Are
Limited

GAO- 02- 530

Page i GAO- 02- 530 Violence Against Women Letter 1 Results in Brief 2
Background 4 Available Data on Pregnant Victims of Violence Are Incomplete
and Lack Comparability 6 Multiple Strategies Designed to Prevent Violence,
But Effect Is

Unknown 12 Concluding Observations 17 Agency Comments 17 Appendix I Scope
and Methodology 19

Appendix II Description of the Pregnancy Risk Assessment Monitoring System
21

Appendix III Pregnancy Status Questions on States? Death Certificates 23

Appendix IV Comments from the Department of Health and Human Services 24

Appendix V GAO Contacts and Staff Acknowledgments 29

Table

Table 1: PRAMS Estimates of the Prevalence of Physical Abuse by Husband or
Partner during Pregnancy, 1998 22 Figure

Figure 1: Pregnancy Status Categories on Proposed U. S. Standard Death
Certificate Revision 11 Contents

Page ii GAO- 02- 530 Violence Against Women Abbreviations

ACOG American College of Obstetricians and Gynecologists BJS Bureau of
Justice Statistics CDC Centers for Disease Control and Prevention FBI
Federal Bureau of Investigation HHS Department of Health and Human Services
HRSA Health Resources and Services Administration

NIH National Institutes of Health NIJ National Institute of Justice NVDRS
National Violent Death Reporting System OJP Office of Justice Programs PRAMS
Pregnancy Risk Assessment Monitoring System

SAMHSA Substance Abuse and Mental Health Services Administration UCR Uniform
Crime Reporting Program VAWA Violence Against Women Act VAWO Violence
Against Women Office

Page 1 GAO- 02- 530 Violence Against Women May 15, 2002 The Honorable
Eleanor Holmes Norton

House of Representatives Dear Ms. Norton: Violence against women, including
violence that results in homicide, is a significant health and criminal
justice problem. The problem is magnified when the victim of violence is
pregnant because there are additional health risks to both the woman and her
unborn child. Objectives to decrease violence against women were included in
Healthy People 2010,

the nation?s health promotion and disease prevention strategy. 1 In response
to its concerns about violence against women, the Congress passed the
Violence Against Women Act (VAWA), 2 which funds, among other things,
programs to shelter battered women, training for law enforcement officers
and prosecutors, and research on violence against

women. Violence against women largely involves intimate partners, such as
husbands, boyfriends, and dates. A recent federal report estimated that
about 2.1 million women are raped or physically assaulted annually. 3 Of
surveyed women who reported being raped or physically assaulted since the
age of 18, about three quarters reported being victimized by a current or
former spouse, cohabiting partner, or date. Due to your concern about
pregnant women being victims of homicide and other violence, you asked us to
provide information on this problem. In response to your request, this
report will discuss (1) the availability of 1 U. S. Department of Health and
Human Services, Healthy People 2010: Understanding and Improving Health, 2nd
ed. (Washington, D. C.: U. S. Government Printing Office,

November 2000). 2 VAWA was enacted in 1994 as Title IV of the Violent Crime
Control and Law Enforcement Act of 1994, P. L. No. 103- 322, 108 Stat. 1796,
1945. In 2000, VAWA was reauthorized and amended- adding several new
programs. See Victims of Trafficking and Violence Protection Act of 2000, P.
L. No. 106- 386, 114 Stat. 1464, 1491. 3 Patricia Tjaden and Nancy Thoennes,
Prevalence, Incidence, and Consequences of Violence Against Women: Findings
From the National Violence Against Women Survey,

NCJ 172837 (Washington, D. C.: U. S. Department of Justice, November 1998).
United States General Accounting Office Washington, DC 20548

Page 2 GAO- 02- 530 Violence Against Women information on the prevalence and
risk of violence against pregnant women and on the number of pregnant women
who are victims of

homicide and (2) strategies and programs to prevent violence against
pregnant women.

To answer these questions, we interviewed and obtained documents from
officials at the Department of Health and Human Services? (HHS) Centers for
Disease Control and Prevention (CDC), National Institutes of Health (NIH),
and Health Resources and Services Administration (HRSA) and the Department
of Justice?s Office of Justice Programs (OJP) and Federal Bureau of
Investigation (FBI). We also interviewed and collected information from
researchers and representatives of four states? departments of health and
vital statistics, medical examiners? offices, local

law enforcement, domestic violence coalitions, violence prevention programs,
health care professional organizations, and advocacy groups. We reviewed
literature on the prevalence and risk of violence toward women during
pregnancy; we identified 11 studies published since 1998 that contained
prevalence estimates. We conducted our work from July 2001 through April
2002 in accordance with generally accepted

government auditing standards. (For additional information on our
methodology, see app. I.) Available data on the number of pregnant women who
are victims of

violence, including violence that results in homicide, are incomplete and
lack comparability. Our review found that there is no current national
estimate of the prevalence of violence against pregnant women- that is,

the proportion of pregnant women who experience violence. Estimates that are
available cannot be generalized or projected to all pregnant women. For
example, CDC?s Pregnancy Risk Assessment Monitoring System (PRAMS) produces
estimates of the prevalence of violence, but

only for women whose pregnancies resulted in live births and only for
participating states. For 1998, PRAMS prevalence estimates for the 15
participating states ranged from 2.4 percent to 6.6 percent. Many studies
focus on narrowly defined populations and use varying definitions of
violence, producing prevalence estimates that are not comparable.

Research findings on whether women are at increased risk for violence during
pregnancy are inconclusive. CDC reported that, while additional research is
needed in this area, current study findings suggest that for most abused
women, the risk of physical violence does not seem to increase during
pregnancy. Moreover, some women who previously experienced violence do not
experience violence during their pregnancies. Factors that studies have
found to be associated with violence during Results in Brief

Page 3 GAO- 02- 530 Violence Against Women pregnancy include violence before
pregnancy, younger age of the victim, and unintended pregnancy.

Little information is available on the number of pregnant homicide victims.
Federal homicide data collected by CDC and the FBI do not capture the
pregnancy status of female victims. Seventeen states try to collect
pregnancy data on death certificates, but these data may understate the
number of pregnant homicide victims because autopsies, if conducted, might
not include examinations for pregnancy, and pregnancies, if identified,
might not be reported on death certificates. Officials in the four states we
contacted have attempted to improve the data by linking

multiple data sources, such as medical examiners? reports and death
certificates. However, some of these officials told us that they do not have
the resources to conduct such database links on a continuing basis. CDC has
begun two initiatives that could result in better data on homicides of
pregnant women- a revision to the U. S. standard death certificate to
include pregnancy status and a proposed national violent death reporting
system, both of which involve federal and state participation. Continued
federal- state collaboration to gather and analyze more complete and
comparable data, such as these initiatives and PRAMS, could improve
policymakers? knowledge of violence against women and guide future research
and resource allocation.

Health and criminal justice officials have designed multiple strategies to
prevent violence against women, but their effect is unknown. Strategies to
prevent violence against pregnant women are similar to those to prevent
violence against all women. These strategies include public health efforts
to keep violence from occurring in the first place and intervention
activities that identify and respond to violence after it occurs, as well as
criminal justice strategies that focus on incarcerating or rehabilitating
batterers. Screening, or asking women about their experience with violence,
is generally the initial component of interventions. However, recent studies
found that fewer than half of physicians routinely screen for violence
during prenatal visits. Reasons cited for physicians? reluctance to screen
include lack of training on how to conduct screenings and not knowing how to
respond if a woman discloses violence. Little information is available on
the effectiveness of strategies to prevent violence against women, including
batterer prevention programs and routine screening.

CDC has not recommended routine screening for intimate partner violence
because of the lack of scientific evidence about its effectiveness. HRSA is
currently funding four small prevention projects, each of which includes an
evaluation component. Evaluating the outcomes of violence prevention

Page 4 GAO- 02- 530 Violence Against Women programs and strategies could
help identify successful approaches for reducing violence against women.

We requested comments on a draft of this report from the Attorney General
and the Secretary of HHS. Justice informed us that it did not have any
comments. HHS agreed that limited information is available on

violence against pregnant women. In addition, HHS discussed several issues
and efforts that it considers important regarding violence against women.
Violence against women can include a range of behaviors such as hitting,
pushing, kicking, sexually assaulting, using a weapon, and threatening
violence. Violence sometimes includes verbal or psychological abuse,
stalking, or enforced social isolation. Victims are often subjected to
repeated physical or psychological abuse.

The federal public health agencies that address violence against women
include CDC, NIH, HRSA, and the Substance Abuse and Mental Health Services
Administration (SAMHSA). They focus on activities such as defining and
measuring the magnitude of violence, identifying causes of

violence, and evaluating and disseminating promising prevention,
intervention, and treatment strategies. CDC?s National Center for Injury
Prevention and Control and National Center for Chronic Disease Prevention
and Health Promotion have funded efforts to document the prevalence of
violence against women, improve maternal health, and prevent intimate
partner violence. CDC?s National Center for Health Statistics operates the
National Vital Statistics System, which maintains a

national database of death certificate information. The National Center for
Health Statistics has a contract with each state to support routine
production of annual vital statistics data, generally covering from
onefourth to one- third of state vital statistics operating costs. NIH has
funded research to study violence against women through several of its
institutes- the National Institute on Alcohol Abuse and Alcoholism, National
Institute of Child Health and Human Development, National Institute on Drug
Abuse, National Institute of Nursing Research, and

National Institute of Mental Health- and the National Center for Research
Resources. HRSA?s Maternal and Child Health Bureau, as part of its mission
to promote and improve the health of mothers and children, funds
demonstration grant programs that focus on violence against women during the
prenatal period. SAMHSA funds efforts focused on the mental health and
substance abuse treatment of women who have been victims of violence.
Background

Page 5 GAO- 02- 530 Violence Against Women The federal criminal justice
agencies that address violence against women are OJP?s Violence Against
Women Office (VAWO), National Institute of Justice (NIJ), and Bureau of
Justice Statistics (BJS). Using VAWA funds,

VAWO administers grants to help states, tribes, and local communities
improve the way criminal justice systems respond to intimate partner
violence, sexual assault, and stalking. VAWO also works with victims?
advocates and law enforcement agencies to develop grant programs that
support a range of services for victims, including advocacy, emergency

shelters, law enforcement protection, and legal aid. VAWO administers these
funds through both formula and discretionary grant programs. 4 NIJ conducts
and funds research on a variety of topics, including violence,

drug abuse, criminal behavior, and victimization. BJS collects, analyzes,
publishes, and disseminates information on crime, criminal offenders,
victims of crime, and the operation of justice systems at all levels of
government.

The FBI administers the Uniform Crime Reporting Program (UCR). Under this
program, city, county, and state law enforcement agencies voluntarily
provide information on eight crimes occurring in their jurisdictions:

criminal homicide, forcible rape, robbery, aggravated assault, burglary,
larceny- theft, motor vehicle theft, and arson. The FBI assembles and
publishes the data and distributes them to contributing local agencies,
state UCR programs, and others interested in the nation?s crime problems.
CDC homicide data indicate that from 1995 through 1999, homicide was

the second leading cause of death for women aged 15 to 24, after accidents.
CDC data also show that almost 2,600 women of childbearing age (15 through
44) were homicide victims in 1999. BJS reported that intimate partner
homicides accounted for about 11 percent of all murders nationwide in that
year. 5 Seventy- four percent of these murders (1,218 of 1,642) were of
women. About 32 percent of all female homicide victims were murdered by an
intimate partner, in comparison to about 4 percent of all male homicide
victims.

4 VAWA funds programs in both Justice and HHS. Justice?s fiscal year 2002
appropriation for VAWA programs was $390. 6 million. HHS?s fiscal year 2002
appropriation for VAWA programs was $176. 7 million for battered women?s
shelters, a domestic violence hotline, rape prevention and education, and
community programs on intimate partner violence. 5 Callie Marie Rennison,
Intimate Partner Violence and Age of Victim, 1993- 99, Bureau of Justice
Statistics Special Report, NCJ 187635 (Washington, D. C., U. S. Department
of Justice, October 2001).

Page 6 GAO- 02- 530 Violence Against Women There is no current national
estimate of the prevalence of violence against pregnant women. Estimates
that are currently available cannot be generalized or projected to all
pregnant women. CDC?s PRAMS develops

statewide estimates of the prevalence of violence for women whose
pregnancies resulted in live births; 1998 estimates for 15 participating
states ranged from 2.4 percent to 6.6 percent. Research on whether women are
at increased risk for violence during pregnancy is inconclusive.

However, CDC reported that study findings suggest that, for most abused
women, physical violence does not seem to be initiated or to increase during
pregnancy. National data are also not available on the number of pregnant
homicide victims, and such data at the state level are limited. The

two federal agencies collecting homicide data, the FBI and CDC, do not
identify the pregnancy status of homicide victims. CDC is exploring
initiatives that could result in better data on homicides of pregnant women.
There is no current national estimate measuring the prevalence of violence
during pregnancy- that is, the proportion of pregnant women who

experience violence. Some state- and community- specific estimates are
available, but they cannot be generalized or projected to all pregnant
women. CDC developed PRAMS, an ongoing population- based surveillance system
that generates state- specific data on a number of maternal behaviors, such
as use of alcohol and tobacco, and experiences- including physical abuse-
before, during, and immediately following a woman?s pregnancy.

CDC awards grants to states to help them collect these data. The number of
states that participate in PRAMS has increased since its inception. Five
states and the District of Columbia participated in fiscal year 1987 and 32
states and New York City participated in fiscal year 2001. CDC officials

reported that lack of funds has prevented additional states from being
added; six states were approved for participation in PRAMS but were not
funded in 2002. CDC?s goal is to fund all states that want the surveillance

system. The estimated 1998 PRAMS prevalence rates of physical abuse by
husband or partner during pregnancy, which CDC reported for 15 states,
ranged Available Data on Pregnant Victims of

Violence Are Incomplete and Lack Comparability

Knowledge of Prevalence of Violence during Pregnancy Is Limited, Although
Several Risk Factors Have Been Identified

Page 7 GAO- 02- 530 Violence Against Women from 2.4 percent to 6.6 percent.
6 (See app. II for PRAMS prevalence estimates for the 15 participating
states and a description of PRAMS?s methodology.) States participating in
PRAMS use a consistent data collection methodology that allows for
comparisons among states, but it

does not allow for development of national estimates because states
participating in PRAMS were not selected to be representative of the nation.
In addition, PRAMS data cannot be generalized to all pregnant

women because they represent only those women whose pregnancies resulted in
live births; the data do not include women whose pregnancies ended with
fetal deaths or abortions or women who were victims of homicide. 7 PRAMS is
based on self- reported data and, because some women are unwilling to
disclose violence, the findings may underestimate abuse. Studies have also
estimated the prevalence of violence within certain states and communities
and among narrowly defined study populations. These estimates lack
comparability and cannot be generalized or projected to all pregnant women.
Many of the studies do not employ random samples and are disproportionately
weighted toward specific demographic or socioeconomic populations. Most of
the 11 such studies we reviewed,

which were published from 1998 through 2001, found prevalence rates of
violence during pregnancy ranging from 5. 2 percent to 14. 0 percent. In a
CDC- sponsored 1996 review of the literature, the majority of studies
reported prevalence levels of 3.9 percent to 8.3 percent. 8 The variability
in estimates could reflect differences in study populations and

methodologies, such as differences in how violence is defined, the time
period used to measure violence, and the method used to collect the data.
Research on whether being pregnant places women at increased risk for
violence is inconclusive. CDC reported that additional research is needed in
this area, but that current study findings suggest that for most abused
women, physical violence does not seem to be initiated or to increase

6 The most recent year for which CDC has reported comprehensive data for
PRAMS is 1998. CDC reported data for those 15 participating states that had
fully implemented PRAMS data collection procedures and achieved CDC?s
required response rate of at least 70 percent.

One additional state participated in PRAMS but did not meet these criteria.
7 CDC reported that, in 1997, 63 percent of pregnancies resulted in live
births.

8 Julie A. Gazmararian and others, ?Prevalence of Violence Against Pregnant
Women,? JAMA 275, no. 24 (1996): 1915- 1920.

Page 8 GAO- 02- 530 Violence Against Women during pregnancy. 9 Although some
women experience violence for the first time during pregnancy, the majority
of abused pregnant women

experienced violence before pregnancy. In one study we reviewed, only 2
percent of women who reported not being abused before pregnancy reported
abuse during pregnancy. 10 The same study also found that, for

some women, the period of pregnancy may be less risky, with violence abating
during pregnancy; 41 percent of the women who reported abuse in the year
before pregnancy did not experience abuse during pregnancy. Studies have
found other factors to be associated with violence during pregnancy,
including younger age of the woman, lower socioeconomic status, abuse of
alcohol and other drugs by victims and perpetrators of violence, and
unintended pregnancy. 11 To increase the generalizability of research on the
prevalence and risk of

violence to women during pregnancy, researchers have reported the need for
more population- based studies that would allow for comparisons of pregnant
and nonpregnant women. These studies would draw their samples from all
pregnant women, not just those receiving health care or giving birth, as
well as nonpregnant women. Such research could indicate whether pregnant
women are at increased risk for violence compared to their nonpregnant
counterparts. Researchers have also suggested using methodologies that
consistently define and measure the prevalence of violence. A recent report
by the Institute of Medicine on family violence

recommended that the Secretary of HHS establish new, multidisciplinary
education and research centers to, among other things, conduct research on
the magnitude of family violence and the lack of comparability in current
research. 12 9 Melissa Moore, ?Reproductive Health and Intimate Partner
Violence,? Family Planning

Perspectives 31, no. 6 (1999): 302- 306, 312. 10 Sandra L. Martin and
others, ?Physical Abuse of Women Before, During, and After Pregnancy,? JAMA
285, no. 12 (2001): 1581- 1584. 11 For example, see Mary M. Goodwin and
others, ?Pregnancy Intendedness and Physical

Abuse Around the Time of Pregnancy: Findings from the Pregnancy Risk
Assessment Monitoring System, 1996- 1997,? Maternal and Child Health Journal
4, no. 2 (2000): 85- 92; and Vilma E. Cokkinides and others, ?Physical
Violence During Pregnancy: Maternal Complications and Birth Outcomes,?
Obstetrics & Gynecology 93, no. 5 (1999): 661- 666. 12 Family violence
includes intimate partner violence, child abuse and neglect, and elder
abuse. Institute of Medicine, Confronting Chronic Neglect: The Education and
Training of Health Professionals on Family Violence (Washington, D. C.:
2001).

Page 9 GAO- 02- 530 Violence Against Women There is also little information
available on violence against pregnant women that results in homicide. The
FBI and CDC are the two federal

agencies that collect and report information on homicides nationwide;
however, neither agency collects data on whether female homicide victims
were pregnant or recently pregnant. According to CDC, 17 states, New York
City, and Puerto Rico collect data related to pregnancy status on their
death certificates, but the data collected are not comparable. Included in
these data are victims who may not have been pregnant at the time of death
but had been ?recently? pregnant; in addition, states? criteria for recent
pregnancy ranged from 42 days to 1 year after birth. (See app. III for a
list of the questions on pregnancy status that states include on their death
certificates.)

The ability to identify pregnant homicide victims from death certificates is
limited. While there are questions on some states? death certificates
regarding pregnancy status, officials in the four states we contacted
(Illinois, Maryland, New Mexico, and New York) told us that these data are
incomplete and may understate the number of pregnant homicide victims.

For example, if the pregnancy item on the death certificate is left blank,
there is no way to easily determine whether an autopsy, if conducted,
included a test or examination for pregnancy. Moreover, researchers have

reported that physicians completing death certificates after a pregnant
woman?s death failed to report that the woman was pregnant or had a recent
pregnancy in at least 50 percent of the cases. 13 To address these
limitations, all four states we contacted are making

efforts to compare death certificate data with other datasets and records-
such as medical examiners? reports- to identify pregnant or recently
pregnant homicide victims. They told us that they are reviewing the data in
order to determine if there is something they can do to prevent violent

deaths of pregnant women or help women who are victimized. For example, the
Maryland medical examiner?s office conducted a study of the deaths of
females aged 10 to 50 to determine if these women were pregnant when they
died. Several sources of data- death certificates, medical examiners?
reports, and recent live birth and fetal death records- from a 6- year
period were linked. Of the 247 women who were identified as pregnant or
recently pregnant, 27 percent were identified through

13 Isabelle L. Horon and Diana Cheng, ?Enhanced Surveillance for Pregnancy-
Associated Mortality- Maryland, 1993- 1998,? Journal of the American Medical
Association 285, no. 11 (2001): 1455- 1459. Pregnancy Status Often Not
Reflected in Data on

Homicide Victims

Page 10 GAO- 02- 530 Violence Against Women examining cause of death
information on death certificates. The remaining 73 percent were identified
by matching the woman?s death certificate with

recent birth and fetal death records and by reviewing data from medical
examiners? records, such as autopsy reports or police records. Similarly,
New York officials determined through dataset links (death certificates,
fetal death records, recent birth certificates, and hospital discharge
records) that, in 1997, 9 of 174 female homicide victims aged 10 to 54 were
pregnant or recently pregnant at the time of death, rather than the 1 of 174
that death certificate data alone would have indicated. Officials from New
York and Maryland told us these efforts to link datasets are dependent on
records being computerized. Some state officials also told us they did not

have the resources to conduct these analyses on a continuing basis. There
are two federal initiatives under development that propose to collect data
on the number of homicides of pregnant women. CDC is proposing a revision of
the U. S. standard certificate of death used for the National Vital
Statistics System to include five categories related to

pregnancy status. (See fig. 1.) Each state has the option of adopting the U.
S. standard certificate for its death certificate or excluding or adding
data elements. If the revision is approved, CDC expects several states to
implement it in 2003, with an increasing number using it each year.

Page 11 GAO- 02- 530 Violence Against Women Figure 1: Pregnancy Status
Categories on Proposed U. S. Standard Death Certificate Revision

Source: CDC. CDC is also beginning to implement the National Violent Death
Reporting System (NVDRS), which, as currently envisioned, would collect data
that could determine the number of pregnant homicide victims. CDC plans to
collect data from a variety of state and local government databases on
deaths resulting from homicide and suicide. Like the Maryland and New York
efforts, NVDRS would link several databases, such as death and

medical examiners? records, to identify pregnant homicide victims. According
to CDC, implementation of NVDRS depends on future funding; full
implementation would take at least 5 years. The estimated federal cost of
this system is $10 million in start- up costs and $20 million in annual
operating costs; these estimates primarily consist of expenditures for

providing technical assistance to the states and funding for state personnel
to collect the data.

Not pregnant within past year Pregnant at time of death Not pregnant, but
pregnant within 42 days of death Not pregnant, but pregnant 43 days to 1
year before death Unknown if pregnant within the past year If female:

Page 12 GAO- 02- 530 Violence Against Women Violence prevention strategies
for both pregnant and nonpregnant women include measures to prevent initial
incidents of violence, such as educating

women about warning signs of abuse, and intervention activities that
identify and respond to violence after it has occurred. Typically, the
initial component of an intervention is screening, or asking women about
their experiences with violence. Many health care organizations and
providers recommend routine screening for intimate partner violence. Studies
have found, however, that fewer than half of physicians routinely screen for

violence during prenatal visits. Reasons for physicians? reluctance to
screen include lack of training on how to screen and how to respond if a
woman discloses violence. Violence prevention strategies also include
criminal justice measures, which focus on apprehending, sentencing,
incarcerating, and rehabilitating batterers. Little information is available
on the effectiveness of violence prevention strategies and programs.
Researchers have reported the need for evaluations of the effectiveness of
screening protocols and batterer intervention programs.

Measures to prevent violence against pregnant women are similar to those to
prevent violence against all women. Public health violence prevention
programs can include primary prevention measures to keep violence from
occurring in the first place and interventions that ask women about their

experiences with violence and respond if violence has occurred. Criminal
justice strategies to prevent violence against women focus on apprehending,
sentencing, incarcerating, and rehabilitating batterers. Efforts to prevent
initial incidents of violence concentrate on attitudes and behaviors that
result in violence against women. These efforts include educating children,
male and female, about ways to handle conflict and anger without violence
and social norms about violence, such as attitudes about the acceptability
of violence toward women. They also include training parents, police
officers, and other community officials to be resources for youth seeking
assistance about teenage dating violence. Primary prevention efforts also
have been targeted to pregnant women. For example, the Domestic Violence
During Pregnancy Prevention Program in Saginaw, Michigan, provided 15-
minute counseling sessions to pregnant women who reported that they had not
experienced violence. 14 Women were educated about intimate partner violence
and given tools

and information to help prevent abuse in their lives, including information
14 This program is a component of the Saginaw Fetal- Infant Mortality Review
Program. Multiple Strategies Designed to Prevent

Violence, But Effect Is Unknown

Violence Prevention Programs Use Health and Criminal Justice Strategies

Page 13 GAO- 02- 530 Violence Against Women on behaviors typical of abusive
men, warning signs of abuse, and community resources.

Interventions to deal with violence that has occurred are designed to
identify victims and to prevent additional violence through such actions as
providing an assessment of danger, developing a safety plan, and providing
information about and referral to community resources. For example, HRSA has
funded a demonstration program to develop or enhance systems that identify
pregnant women experiencing intimate partner violence and provide
appropriate information and links to services. The HRSA program funds four
projects; each project is funded at $150,000 a year for 3 years. 15
Screening for the presence of violence is generally the initial component of
intervention efforts to prevent additional violence against pregnant

women. Many experts view the period of pregnancy as a unique opportunity for
intervention. Pregnant women who receive prenatal care may have frequent
contact with providers, which allows for the development of relationships
that may facilitate disclosure of violence. For example, the American
College of Obstetricians and Gynecologists

(ACOG) recommends that physicians screen all patients for intimate partner
violence and that screening for pregnant women occur at several times over
the course of their pregnancies. Some women do not disclose

abuse the first time they are asked, or abuse may begin later in pregnancy.
Some of the barriers to women?s disclosure of violence are fear of
escalating violence, feelings of shame and embarrassment, concern about
confidentiality, fear of police involvement, and denial of abuse. In
addition, some health care officials told us that the period of pregnancy
may be a difficult time for a woman to leave or take action against the
abuser because of financial concerns and pressures to provide the child with
a father.

15 The projects are the Comprehensive Services for Pregnant Women
Experiencing Substance Abuse and Violence in Baltimore, Maryland; Systems
for Pregnancy Education and Awareness of Safety in New York, New York;
Improving Systems of Care for Pregnant Women Experiencing Domestic Violence
in St. Clair County, Illinois; and Perinatal Partnership Against Domestic
Violence: Improving Systems of Care for Pregnant/ Post Partum Women in the
Asian and Pacific Islander Community in Seattle, Washington. HRSA is
planning to initiate another demonstration program in June 2002 to address
family violence during or around the period of pregnancy. The primary focus
of the program is women experiencing violence, but its projects will also
link to child abuse, elder abuse, and

perpetrator rehabilitation programs.

Page 14 GAO- 02- 530 Violence Against Women Studies have found that fewer
than half of physicians routinely screen women for violence during
pregnancy. For example, a survey of ACOG

fellows reported that 39 percent of respondents routinely screened for
violence at the first prenatal visit. 16 The study found that screening was
more likely to occur when the obstetrician- gynecologist suspected a

patient was being abused. Another study that surveyed primary care
physicians who provide prenatal care found that only 17 percent of
respondents routinely screened at the first prenatal visit and 5 percent at
follow- up visits. 17 Across the 15 states with PRAMS data for 1998, from 25
percent to 40 percent of women reported that a physician or other health

care provider talked to them about intimate partner violence during any of
their prenatal care visits. CDC and providers of prevention services have
reported that reasons for

physicians? reluctance to screen women for violence include lack of time and
resources, personal discomfort about discussing the topic, concern about
offending patients, belief that asking invades family privacy, and
frustration with patients who are not ready to leave or who return to their
abusers. Lack of training and education on how to screen for intimate
partner violence and lack of knowledge about what to do if a woman reports
experiencing intimate partner violence have also been cited as barriers to
physician screening. In its report on family violence, the Institute of
Medicine stated that health professionals? training and

education about family violence are inadequate and recommended that the
Secretary of HHS establish education and research centers to develop
training programs that prepare health professionals to respond to family
violence. Criminal justice approaches to preventing violence against women
include apprehending and sanctioning the batterer, preventing further
contact between the abuser and the victim, and connecting the victim to

community services. In addition, batterer intervention programs, which have
existed for over 20 years as a criminal justice intervention, are often used
as a component of pretrial or diversion programs or as part of sentencing.
Batterer programs can include classes or treatment groups, 16 Deborah L.
Horan and others, ?Domestic Violence Screening Practices of
ObstetricianGynecologists,?

Obstetrics & Gynecology 92, no. 5 (1998): 785- 789. 17 Linda Chamberlain and
Katherine A. Perham- Hester, ?Physicians? Screening Practices for Female
Partner Abuse During Prenatal Visits,? Maternal and Child Health Journal 4,
no. 2 (2000): 141- 148.

Page 15 GAO- 02- 530 Violence Against Women evaluation, individual
counseling, or case management; their goals are rehabilitation and
behavioral change.

To assist communities, policymakers, and individuals in combating violence
against women, the National Advisory Council on Violence Against Women and
VAWO developed a Web- based resource for instruction and guidance. 18 These
guidelines include recommendations for strengthening prevention efforts and
improving services and advocacy for victims. For example, the guidelines
recommend that communities increase the cultural and linguistic competence
of their sexual assault, intimate partner violence, and stalking programs by
recruiting and hiring staff, volunteers, and board members who reflect the
composition of the community the program serves. The guidelines also
recommend that all health and mental health care professional school and
continuing education curricula include information on the prevention,
detection, and treatment of sexual assault and intimate partner violence.

Researchers have reported that little information is available on the
effectiveness of strategies to prevent and reduce violence against women.
For example, many health care organizations and providers advocate routine
screening of pregnant women for intimate partner violence, but questions
have been raised about the effectiveness of screening, the most effective
way to conduct screening, and the optimal times for conducting screening. In
addition, limited information is available on the impact of screening on
women and their children.

A CDC official told us that CDC has not issued guidelines or recommendations
related to routine screening for violence in health care settings, primarily
due to the lack of scientific evidence about the effectiveness of screening.
CDC recently funded a cooperative agreement to measure the effectiveness of
an intimate partner violence intervention

that includes evaluation of a screening protocol and computerized screening.
19 The results of the study are expected to provide data on the array of
outcomes that need to be considered in implementing

18 National Advisory Council on Violence Against Women and the Violence
Against Women Office, Toolkit to End Violence Against Women (Washington, D.
C.: U. S. Department of Justice, November 2001). http:// toolkit. ncjrs. org
(downloaded on February 12, 2002). 19 The cooperative agreement is between
CDC, Johns Hopkins University, and the State

University of New York at Albany. Little Information Is Available on the

Effectiveness of Violence Prevention Programs

Page 16 GAO- 02- 530 Violence Against Women intervention programs to
decrease intimate partner violence. CDC officials told us that additional
studies are necessary to evaluate screening and

intervention strategies and that CDC is in the process of identifying
additional study topics and designs that could complement this effort.

CDC and other researchers on violence against women and providers of
prevention services have identified several other areas in which research
could be fruitful. For example, they have reported the need to

 develop information on the most effective ways to promote women?s safety
after screening;

 develop and evaluate the effectiveness of programs that coordinate
community resources from the medical, social services, law enforcement,
judicial, and legal systems; and

 develop and evaluate the effectiveness of prevention strategies that
incorporate cultural perspectives in serving ethnic and immigrant
populations.

An example of an effort to conduct such research is HRSA?s program to
improve interventions for pregnant women experiencing violence; however, the
projects? evaluation components are small and, according to HHS, their
results may not be generalizable to the nation. Each funded project will
evaluate whether its intervention was effective in improving rates of
screening, assessment, and referral or links to community services; the
projects may also assess the impact of the intervention on

women?s behaviors. For example, the Comprehensive Services program in
Baltimore is assessing whether the project was effective in linking families
to needed services and whether women report improvement in their physical or
psychosocial status after the intervention. The Systems for Pregnancy
Education and Awareness of Safety in New York is evaluating whether the
project increases the number of women who disclose violence and receive
services and referrals to community services, such as shelters. The
Perinatal Partnership Against Domestic Violence in Seattle is evaluating the
effectiveness of screening protocols and interventions that

are tailored to the culture and values of women who are Asian and Pacific
Islanders.

Researchers have also reported that there is little evaluative information
on the effectiveness of violence prevention programs for batterers. A VAWO-
funded study of the effectiveness of batterer programs concluded that they
have modest effects on violence prevention when compared with traditional
probationary practices and that there is little evidence to support the
effectiveness of one batterer program over another in

Page 17 GAO- 02- 530 Violence Against Women reducing recidivism. 20 The
study concluded, however, that batterer programs are a small but critical
element in an overall violence prevention

effort that includes education, arrest, prosecution, probation, and victim
services. The study authors advocated experimenting with different program
approaches and performing outcome evaluations of batterer programs.

The magnitude of the problem of violence against pregnant women is unknown.
Current collaborative efforts by federal and state governments to gather and
analyze more complete and comparable data could improve policymakers?
knowledge of the extent of this violence and guide future

research and resource allocation. These efforts can also help in setting
priorities for prevention strategies. Continuing evaluation of prevention
strategies and programs could help identify successful approaches for

reducing violence against women. We provided a draft of this report to
Justice and HHS for comment. Justice informed us that it did not have any
comments. HHS agreed with our finding that limited information is available
regarding violence against pregnant women. HHS also noted reasons why the
data are incomplete,

such as the difficulty of collecting data from a representative sample of
pregnant victims because they are such a small percentage of the U. S.
population. Other reasons HHS cited are legal and ethical issues in
conducting research on this population, such as maintaining privacy and

confidentiality. HHS commented that several states are conducting mortality
reviews to better understand pregnancy- related deaths and their underlying
causes.

HHS raised several issues that it considers important regarding violence
against women, such as the need to evaluate factors correlated with violence
against women, and identified additional efforts within the department that
focus on intimate partner violence. We recognize that there are many issues
and efforts related to violence against women; however, our focus was on
violence against pregnant women, and therefore much of our discussion
relates to this population. HHS noted 20 Larry Bennett and Oliver Williams,
Controversies and Recent Studies of Batterer

Intervention Program Effectiveness, Grant number 98- WT- VX- K001
(Washington, D. C.: U. S. Department of Justice, 2001). Concluding
Observations

Agency Comments

Page 18 GAO- 02- 530 Violence Against Women that although HRSA?s
demonstration program to improve interventions for pregnant women
experiencing violence will result in new qualitative

information, the evaluation component is small and the findings would likely
be limited. We modified our discussion of this program to indicate that it
is a small demonstration program and its results may not be generalizable to
the nation. In response to HHS?s comments, we added a

description of another demonstration program focused on violence against
pregnant women that HRSA plans to initiate in June 2002. HHS also provided
technical comments, which we incorporated where appropriate. (HHS?s comments
are reprinted in app. IV.) As arranged with your office, unless you publicly
announce its contents earlier, we plan no further distribution of this
report until 30 days after its

issue date. We will then send copies to the Secretary of Health and Human
Services; the Attorney General; the Administrator of the Health Resources
and Services Administration; the Directors of the Centers for Disease
Control and Prevention, National Institutes of Health, Office of Justice
Programs, and Federal Bureau of Investigation; appropriate congressional
committees; and others who are interested. We will also make copies
available to others on request.

If you or your staff have any questions, please contact me at (202) 512-
8777 or Janet Heinrich, Director, Health Care- Public Health Issues, at
(202) 512- 7119. Additional GAO contacts and the names of other staff
members who made contributions to this report are listed in appendix V.

Sincerely yours, Paul L. Jones Director, Tax Administration and Justice

Appendix I: Scope and Methodology Page 19 GAO- 02- 530 Violence Against
Women To do our work, we interviewed and obtained information from officials
at the Department of Health and Human Services? Centers for Disease Control
and Prevention (CDC), Health Resources and Services

Administration (HRSA), and National Institutes of Health, and the Department
of Justice?s Office of Justice Programs (OJP) and Federal Bureau of
Investigation (FBI). We also interviewed representatives of and obtained
information from the American College of Obstetricians and Gynecologists,
Institute of Medicine, Family Violence Prevention Fund, National Coalition
Against Domestic Violence, and National Association of Medical Examiners;
several state domestic violence coalitions; and

researchers. To determine the availability of information on the prevalence
and risk of violence against pregnant women, we reviewed literature on the
prevalence and risk of violence to women during pregnancy. We identified 11
studies published since 1998 that contained prevalence estimates and
assessed their methodologies to ensure the appropriateness of the data
collection and analysis methods and the conclusions. We also interviewed

CDC officials and reviewed data collected through CDC?s Pregnancy Risk
Assessment Monitoring System (PRAMS).

To determine the availability of data on the number of pregnant women who
are victims of homicide in the United States, we interviewed officials and
collected and analyzed homicide statistics and reports from CDC, the FBI,
and OJP?s Bureau of Justice Statistics. We also interviewed officials from
state departments of health and vital statistics in Illinois, Maryland, New
Mexico, and New York to determine how they collect and use data on pregnant
homicide victims. We selected these states because, in addition to
collecting pregnancy data on their state death certificates, they are active
in collecting and analyzing information from various sources to

study maternal health issues. The states were not intended to be
representative of all states. We also interviewed and obtained information
from CDC and Justice officials to identify federal initiatives that are
under way to improve the availability of information on homicides of
pregnant women. To identify strategies and programs to prevent violence
against pregnant

women, we gathered information through a literature review and interviews
with and information collected from researchers and officials from federal
agencies, health care associations, and advocacy groups. We

reviewed a HRSA- funded program (with projects located in Illinois,
Maryland, New York, and Washington) and two other programs (located in
Michigan and Pennsylvania) because they focused specifically on violence
Appendix I: Scope and Methodology

Appendix I: Scope and Methodology Page 20 GAO- 02- 530 Violence Against
Women against pregnant women and served varied populations, including
adolescents, diverse ethnic groups, and women with substance abuse problems.

We conducted our work from July 2001 through April 2002 in accordance with
generally accepted government auditing standards.

Appendix II: Description of the Pregnancy Risk Assessment Monitoring System

Page 21 GAO- 02- 530 Violence Against Women CDC developed PRAMS, a
population- based survey of women whose pregnancies resulted in live births.
CDC awards grants to states to help

them collect information on women?s experiences and behaviors before,
during, and immediately following pregnancy. CDC funded about $6.2 million
for PRAMS in fiscal year 2001; grant awards to states ranged from $100,000
to $150,000. CDC?s funding for PRAMS also includes costs for CDC staff and
contractors to provide technical support to the states.

States participating in PRAMS use a consistent methodology to collect data.
Each state selects a stratified sample of new mothers every month from
eligible birth certificates and then collects data through mailings and
follow- up telephone calls to nonrespondents. A birth certificate is
eligible for the PRAMS sample if the mother was a resident of the state. For
1998,

the most recent year for which CDC has reported comprehensive data for
PRAMS, states used a standardized questionnaire that asked women if their
husbands or partners physically abused them during their most recent
pregnancy. PRAMS defined physical abuse as pushing, hitting,

slapping, kicking, or any other way of physically hurting someone. 1 Table 1
lists 1998 PRAMS estimates of the prevalence of intimate partner violence
during pregnancy.

1 Some states have also added questions on verbal and emotional abuse.
Appendix II: Description of the Pregnancy Risk Assessment Monitoring System

Appendix II: Description of the Pregnancy Risk Assessment Monitoring System

Page 22 GAO- 02- 530 Violence Against Women Table 1: PRAMS Estimates of the
Prevalence of Physical Abuse by Husband or Partner during Pregnancy, 1998
State Percentage a Alabama 3.8

Alaska 3.8 Arkansas 5.5 Colorado 2.8 Florida 4.1 Illinois 4.1 Louisiana 5.2
Maine 2.5 New Mexico b 6.6 New York c 2.4 North Carolina 4.2 Oklahoma 5.1
South Carolina 3.9 Washington 3.5 West Virginia 4.7

Note: PRAMS includes data only for women whose pregnancies resulted in live
births. a This column represents the proportion of pregnant women who
reported physical abuse (i. e., pushing, hitting, slapping, kicking, or any
other way of physically hurting someone). b Data represent births from July
1997 through December 1998. c Data do not include New York City. Source: L.
E. Lipscomb and others, PRAMS 1998 Surveillance Report (Atlanta, Ga.:
Division of Reproductive Health, National Center for Chronic Disease
Prevention and Health Promotion, Centers for Disease Control and Prevention,
2000).

Appendix III: Pregnancy Status Questions on States? Death Certificates

Page 23 GAO- 02- 530 Violence Against Women State Question

Alabama Was there a pregnancy in last 90 days or 42 days? Florida If female,
was there a pregnancy in the past 3 months? Georgia If female, indicate if
pregnant or birth occurred within 90 days of

death. Illinois If female, was there a pregnancy in the past 3 months?
Indiana Was decedent pregnant or 90 days postpartum? Iowa If female, was
there a pregnancy in the past 12 months? Louisiana If deceased was female
10- 49, was she pregnant in the last 90 days?

Maine Indicate if the decedent was pregnant or less than 90 days postpartum
at time of death. Maryland If female, was decedent pregnant in the past 12
months? Missouri If deceased was female 10- 49, was she pregnant in the last
90 days?

Nebraska If female, was there a pregnancy in the past 3 months? New Jersey
If female, was she pregnant at death or any time 90 days prior to

death? New Mexico Was decedent pregnant within last 6 weeks? New York If
female, was decedent pregnant in last 6 months? New York City a If female
under 54, pregnancy in last 12 months? North Dakota Was deceased pregnant
within 18 months of death? Puerto Rico If female, was deceased pregnant?

Texas Was decedent pregnant at time of death; within last 12 months?
Virginia If female, was there a pregnancy in last 3 months?

Note: According to CDC, these are the only states that include questions on
pregnancy status on their death certificates. The term ?states? includes New
York City and Puerto Rico. a According to New York state officials, New York
City uses a different death certificate from the rest of the state. The New
York City death certificate is used for the five boroughs of the city:
Manhattan, Brooklyn, Queens, the Bronx, and Staten Island. Source: CDC.
Appendix III: Pregnancy Status Questions on States? Death Certificates

Appendix IV: Comments from the Department of Health and Human Services Page
24 GAO- 02- 530 Violence Against Women Appendix IV: Comments from the
Department of Health and Human Services

Appendix IV: Comments from the Department of Health and Human Services Page
25 GAO- 02- 530 Violence Against Women

Appendix IV: Comments from the Department of Health and Human Services Page
26 GAO- 02- 530 Violence Against Women

Appendix IV: Comments from the Department of Health and Human Services Page
27 GAO- 02- 530 Violence Against Women

Appendix IV: Comments from the Department of Health and Human Services Page
28 GAO- 02- 530 Violence Against Women

Appendix V: GAO Contacts and Staff Acknowledgments Page 29 GAO- 02- 530
Violence Against Women Weldon McPhail, (202) 512- 8644

Helene F. Toiv, (202) 512- 7162 In addition to those named above,
contributors to this report were Janina Austin, Nancy Kawahara, Emily Gamble
Gardiner, Geoffrey Hamilton, Anthony Hill, Hiroshi Ishikawa, Alice London,
Behn Miller, and Sara- Ann Moessbauer. Appendix V: GAO Contacts and Staff

Acknowledgments GAO Contacts Staff Acknowledgments (440056)

The General Accounting Office, the investigative arm of Congress, exists to
support Congress in meeting its constitutional responsibilities and to help
improve the performance and accountability of the federal government for the
American people. GAO examines the use of public funds; evaluates federal
programs and policies; and provides analyses, recommendations, and other

assistance to help Congress make informed oversight, policy, and funding
decisions. GAO?s commitment to good government is reflected in its core
values of accountability, integrity, and reliability. The fastest and
easiest way to obtain copies of GAO documents at no cost is through the
Internet. GAO?s Web site (www. gao. gov) contains abstracts and fulltext

files of current reports and testimony and an expanding archive of older
products. The Web site features a search engine to help you locate documents
using key words and phrases. You can print these documents in their
entirety, including charts and other graphics.

Each day, GAO issues a list of newly released reports, testimony, and
correspondence. GAO posts this list, known as ?Today?s Reports,? on its Web
site daily. The list contains links to the full- text document files. To
have GAO e- mail this list to you every afternoon, go to www. gao. gov and
select ?Subscribe to daily E- mail alert for newly released products? under
the GAO Reports heading. The first copy of each printed report is free.
Additional copies are $2 each. A check or money order should be made out to
the Superintendent of Documents.

GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed
to a single address are discounted 25 percent. Orders should be sent to:

U. S. General Accounting Office 441 G Street NW, Room LM Washington, D. C.
20548 To order by Phone: Voice: (202) 512- 6000 TDD: (202) 512- 2537

Fax: (202) 512- 6061 Contact: Web site: www. gao. gov/ fraudnet/ fraudnet.
htm E- mail: fraudnet@ gao. gov Automated answering system: (800) 424- 5454
or (202) 512- 7470

Jeff Nelligan, managing director, NelliganJ@ gao. gov (202) 512- 4800 U. S.
General Accounting Office, 441 G Street NW, Room 7149 Washington, D. C.
20548 GAO?s Mission

Obtaining Copies of GAO Reports and Testimony

Order by Mail or Phone To Report Fraud, Waste, and Abuse in Federal Programs
Public Affairs
*** End of document. ***