Building Data Systems for Monitoring and Responding to Violence Against Women
Recommendations from a Workshop
Workshop on Building Data Systems for Monitoring and Responding to Violence Against Women (VAW)
Participants from the U.S. Department of Justice
Bernard Auchter, M.S.W.
National Institute of Justice
Washington, DC
Noel Brennan, M.A., J.D.
Office of Justice Programs
Washington, DC
Jan Chaiken, Ph.D.
Bureau of Justice Statistics
Washington, DC
Sally Hillsman, Ph.D.
National Institute of Justice
Washington, DC
Rebecca Kraus, Ph.D.
National Institute of Justice
Washington, DC
Angela Moore-Parmley, Ph.D.
National Institute of Justice
Washington, DC
Michael Rand
Bureau of Justice Statistics
Washington, DC
Leora Rosen, Ph.D.
National Institute of Justice
Washington, DC
Kathy Schwartz
Office of Justice Programs
Washington, DC
Jeremy Travis, J.D.
National Institute of Justice
Washington, DC
Christy Visher, Ph.D.
National Institute of Justice
Washington, DC
Participants from the U.S. Department of Health and Human Services
Caroline Aoyama, M.P.H.
Health Resources and Services Administration
Bethesda, MD
Marla Aron, M.A.S.
Health Care Financing Administration
Baltimore, MD
Katie Baer, M.P.H.
Centers for Disease Control and Prevention
Atlanta, GA
Kate Brett, Ph.D.
Centers for Disease Control and Prevention
Hyattsville, MD
Cathy Burt, Ed.D.
Centers for Disease Control and Prevention
Hyattsville, MD
Marsha Davenport, M.D.
Health Care Financing Administration
Baltimore, MD
Janet Fanslow, Ph.D.
Centers for Disease Control and Prevention
Atlanta, GA
Lois Fingerhut, M.A.
Centers for Disease Control and Prevention
Hyattsville, MD
Mary Goodwin, M.P.H.
Centers for Disease Control and Prevention
Atlanta, GA
Malcolm Gordon, Ph.D.
National Institute of Mental Health
Rockville, MD
Marcy Gross
Agency for Health Care Policy and Research
Rockville, MD
Rodney Hammond, Ph.D.
Centers for Disease Control and Prevention
Atlanta, GA
Martha Highsmith
Centers for Disease Control and Prevention
Atlanta, GA
John Horan, M.D., M.P.H.
Centers for Disease Control and Prevention
Atlanta, GA
Sandra Howard
Office of the Assistant Secretary for Planning and Evaluation
Washington, DC
Susan Jack, M.S.
Centers for Disease Control and Prevention
Hyattsville, MD
Lynn Jenkins, M.A.
Centers for Disease Control and Prevention
Washington, DC
Wanda Jones, Dr.P.H..
Office of Women's Health
Washington, DC
Ken Kochanek, M.A.
Centers for Disease Control and Prevention
Hyattsville, MD
Jean Kozak, Ph.D.
Centers for Disease Control and Prevention
Hyattsville, MD
Mary Ann MacKenzie
Administration for Children and Families
Washington, DC
Pamela McMahon, Ph.D., M.P.H.
Centers for Disease Control and Prevention
Atlanta, GA
James Mercy, Ph.D.
Centers for Disease Control and Prevention
Atlanta, GA
Jo Mestelle
Administration for Children and Families
Washington, DC
Francess Page, R.N., M.P.H.
Office of Women's Health
Washington, DC
Curtis Porter, M.P.A.
Administration for Children and Families
Washington, DC
Carolina Reyes, M.D.
Agency for Health Care Policy and Research
Rockville, MD
Mark Rosenberg, M.D., M.P.P.
Centers for Disease Control and Prevention
Atlanta, GA
Ann Rosewater, M.A.
Immediate Office of the Secretary
Washington, DC
Beatrice Rouse
Substance Abuse and Mental Health Services
Administration
Rockville, MD
LindaSaltzman, Ph.D.
Centers for Disease Control and Prevention
Atlanta, GA
Fred Seitz, Ph.D.
Centers for Disease Control and Prevention
Hyattsville, MD
Jerry Silverman, M.S.W.
Office of the Assistant Secretary for Planning and Evaluation
Washington, DC
Edward Sondik, Ph.D.
Centers for Disease Control and Prevention
Hyattsville, MD
Daniel Sosin, M.D., M.P.H.
Centers for Disease Control and Prevention
Atlanta, GA
Other Participants
Ronet Bachman, Ph.D.
University of Delaware
Newark, DE
Carolyn Rebecca Block, Ph.D.
Illinois Criminal Justice Information
Authority
Chicago, IL
Ruth Brandwein, Ph.D.
State University of New York
Stony
Brook, NY
Tim Bynum, Ph.D.
Michigan State University
East
Lansing, MI
Donald Camburn, B.G.S.
Research Triangle Institute
Research Triangle Park, NC
Jacquelyn Campbell, Ph.D., R.N.
Johns Hopkins University
Baltimore, MD
Linda Chamberlain, Ph.D.
Alaska Department of Health and
Social Services
Anchorage, AK
Kathleen Chard, Ph.D.
University of Kentucky
Lexington, KY
Mary Ellen Colten, Ph.D.
University of Massachusetts at Boston
Boston, MA
Andrea Craig, M.P.H., M.S.W.
San Francisco Injury Center for
Research and Prevention
San Francisco, CA
Walter DeKeseredy, Ph.D.
Carleton University
Ottawa, Ontario, Canada
Mary Ann Dutton, Ph.D.
George Washington University
Bethesda, MD
Patricia Edgar, Ph.D.
Carnegie Mellon University
Pittsburgh, PA
Bonnie Fisher, Ph.D.
University of Cincinnati
Cincinnati, OH
Richard Gelles, Ph.D.
University of Pennsylvania
Philadelphia, PA
Marijan Grogoza
Mansfield
Police Department
Mansfield, OH
Jeanne Hathaway, M.D.
Massachusetts Department of Public Health
Boston, MA
Nancy Isaac, Ph.D.
Northeastern University
Roxbury, MA
Susan Keilitz, J.D.
National Center for State Courts
Williamsburg, VA
Dean Kilpatrick, Ph.D.
Medical University of South Carolina
Charleston, SC
Mary Koss, Ph.D.
University of Arizona
Tucson, AZ
Colin Loftin, Ph.D.
University at Albany
State University of
New York
Albany, NY
James Lynch, Ph.D.
American University
Washington, DC
Eleanor Lyon, Ph.D.
University of Connecticut
Storrs, CT
Michael Maltz, Ph.D.
University of Illinois at Chicago
Chicago, IL
Sandra Martin, Ph.D.
University of North Carolina at Chapel Hill
Chapel Hill, NC
Wendy Max, Ph.D.
University of California
San Francisco, CA
Anne Menard
National Resource Center on Domestic
Violence
Harrisburg, PA
Susan Murty, Ph.D., M.S.W.
University of Iowa
Iowa City, IA
Stan Orchowsky, Ph.D.
Justice Research and Statistics Association
Washington, DC
Miriam Ornstein, M.P.H.
Research Triangle Institute
Research Triangle Park, NC
Carol Petrie
National Research Council
Washington, DC
Mark Prior, M.S.
Administrative Office of the Trial Court
Boston, MA
Claire Renzetti, Ph.D.
St. Joseph's University
Philadelphia, PA
Sarah Ryan
University of Nevada
Las Vegas, NV
Laura Sadowski, M.D., M.P.H.
Cook County Hospital
Chicago, IL
Joanne Schmidt, M.S.W.
City of New Orleans
New Orleans, LA
Martin Schwartz, Ph.D.
Ohio University
Athens, OH
Joslan Sepulveda, M.P.H.
University of California, Los Angeles
Los Angeles, CA
Anuradha Sharma, M.P.H.
National Resource Center on Domestic
Violence
Harrisburg, PA
Jay Silverman, Ph.D.
Massachusetts Department of Public Health
Boston, MA
Patricia Smith, M.S.
Michigan Department of Community
Health
Lansing, MI
Paula Kovanic Spiro, M.P.H.
University of Pittsburgh
Pittsburgh, PA
Murray Straus, Ph.D.
University of New Hampshire
Durham, NH
Nancy Thoennes, Ph.D.
Center for Policy Research
Denver, CO
Patricia Tjaden, Ph.D.
Center for Policy Research
Denver, CO
Wendy Verhoek-Oftedahl, Ph.D.
Brown University
Providence, RI
Anna Waller, Sc.D.
University of North Carolina
Chapel Hill, NC
Linda Williams, Ph.D.
Wellesley College
Wellesley, MA
Susan Wilt, Ph.D., M.D.
New York City Department of Health
New York, NY
The following CDC staff members prepared this report:
Linda E. Saltzman, Ph.D.
Division of Violence Prevention
National Center for Injury Prevention and Control
Lois A. Fingerhut, M.A. Office of Analysis, Epidemiology, and Health Promotion
National Center for Health Statistics
in collaboration with
Michael R. Rand
Bureau of Justice Statistics
U.S. Department of Justice
Christy Visher, Ph.D.
National Institute of Justice
U.S. Department of Justice
Summary
This report provides recommendations regarding public health
surveillance and research on violence against women developed during a
workshop, "Building Data Systems for Monitoring and Responding to Violence
Against Women." The Workshop, which was convened October 29--30, 1998, was
co-sponsored by the U.S. Department of Health and Human Services and the
U.S. Department of Justice.
BACKGROUND
Available data suggest that violence against women (VAW) (i.e., both
adolescents and adults) is a substantial public health problem in the United States.
Law enforcement data indicate that 3,419 females died in 1998 as a result of homicide
(1), and approximately one third of these women were murdered by a spouse,
ex-spouse, or boyfriend. Data regarding nonfatal cases of assault are less
accessible and are often inconsistent because of methodologic differences. However,
recent survey data collected during 1995--1996 suggest that approximately 2.1
million women are physically assaulted or raped annually; 1.5 million of these women
are physically assaulted or raped by a current or former intimate partner
(2). Based on survey data from the Bureau of Justice Statistics' National Crime
Victimization Survey, in 1998, women were victims in nearly 900,000 violent crimes committed
by an intimate partner (3). Although these and other statistics suggest the magnitude
of the problem, some experts believe that statistics on violence against
women underrepresent the problem; others believe that some studies overestimate
the extent of violence against women. Such lack of consensus and confusion about
the different findings from various data sources prompted the establishment of
the Workshop in October 1998.
INTRODUCTION
The U.S. Department of Health and Human Services (DHHS) and the
U.S. Department of Justice (DOJ) co-sponsored the workshop "Building Data Systems
for Monitoring and Responding to Violence Against Women" in October 1998. The
2-day invitational workshop, funded by CDC's National Center for Injury Prevention
and Control (NCIPC) and National Center for Health Statistics (NCHS) along with
the Bureau of Justice Statistics (BJS) and the National Institute of Justice (NIJ),
brought together researchers and practitioners from the public health and criminal
justice fields.
Earlier in 1998, the U.S. Secretary of Health and Human Services and
Attorney General held a joint briefing that focused on the nature and extent of VAW.
During the briefing, concerns were raised over differences among published estimates
of rape, sexual assault, and intimate-partner violence and the resulting difficulties
for developing and implementing effective programs and policies. The briefing
also highlighted current knowledge about the magnitude of violence against women
and identified areas in which more information is needed. The Workshop was
an outcome of this briefing and was conceived as a first step in a long-term effort
to more accurately measure VAW and to conduct sound research.
In planning the Workshop, the Steering Committee* conceptualized VAW
as encompassing many types of behaviors and relationships between victims
and perpetrators. The Committee decided to focus on that subset of VAW categorized
as intimate-partner violence and sexual violence by any perpetrator
(Figure 1). In addition, several issues were identified as needing to be addressed, including
a) collection of national, state, and local VAW data from both public health and
criminal justice sources to represent different perspectives; b) definitions and
methodologies; and c) concerns about the availability of social services for VAW victims. The
Steering Committee commissioned six background papers that targeted these issues.
All Workshop participants were provided copies of these papers before the
workshop. Each paper was presented at the Workshop, followed by comments from one
or more respondents.**
This Workshop addressed the opportunities and challenges associated with
public health surveillance (i.e., the ongoing and systematic collection, analysis,
and interpretation of information) and research relating to VAW. The goals of
the workshop were to
develop information and make recommendations enabling researchers
to better describe and track VAW;
share information about data collection for VAW, with emphasis on
intimate-partner violence and sexual violence; and
identify gaps and limitations of existing systems for ongoing data
collection regarding VAW.
THE WORK GROUPS
Workshop attendees were divided into four work groups that met twice during
the 2-day meeting. The groups were asked to develop recommendations on
the following four topics related to the background papers and presentations:
defining and measuring VAW;
state and local data for studying and monitoring VAW;
national data for studying and monitoring VAW; and
new research strategies for studying VAW.
Work Group on Defining and Measuring VAW
The purpose of this work group was to identify and make recommendations
about resolving problems resulting from the absence of uniform definitions associated
with VAW. VAW is a broad term, encompassing a wide range of behaviors. Definitions
of VAW should be established that are comprehensive enough to encompass
women's physical and psychological experiences of violence, yet that are not so broad
that they encompass behaviors that cannot be validly defined as VAW. It is
unknown which data elements are most critical, or even possible, to collect. In addition
to identifying components that are critical to defining and measuring VAW, this
work group was asked to address questions about how to develop new
measurement instruments or enhance existing ones to improve the quality of VAW data
collected. The work group was directed to address which aspects of VAW should be
measured (e.g., the occurrence of acts and the number of victims).
Work Group on State and Local Data for Studying
and Monitoring VAW
This work group was charged with developing recommendations regarding
how state and local data systems could be improved for monitoring and
characterizing VAW. They were asked to identify the key opportunities and methodologic
challenges in using state and local data sources and to offer potential solutions for
overcoming the identified challenges. The work group considered what types of data
items should be collected; which data systems have the greatest utility for monitoring
and characterizing VAW at the state and local levels; how greater uniformity in
definitions and types of data collected on VAW can be fostered; and the challenges of
data linkage.
Work Group on National Data for Studying and
Monitoring VAW
This work group was charged with developing recommendations regarding
how to improve and optimize national data for monitoring and characterizing VAW and
its key dimensions (e.g., intimate-partner violence and sexual assault). The work
group recognized that national data are collected from various data sources designed
for different purposes. The group considered 18 surveys and surveillance systems
that either contribute data or have the potential to contribute data toward
measuring some aspect of VAW (Table 1). Although this list is not comprehensive, it served as
a reference for a discussion about what makes a survey or a data system useful
for monitoring VAW.
In addition, the group considered some of the factors that determine the
utility and reliability of VAW estimates (Table 2). None of the 18 surveys or
surveillance systems considered by the work group are ideal for measuring VAW; however,
four surveys (i.e., the National Crime Victimization Survey, the National Violence
Against Women Survey, the National Youth Survey, and the National Women's Study)
are likely the most useful and reliable. Data from each of these surveys can be used
to produce estimates of prevalence, incidence, and chronicity.
Some surveys (e.g., the National Family Violence Survey) can be used to
derive prevalence estimates but are not conducted on an ongoing basis. One
reporting system, the National Incident-Based Reporting System, is ongoing but is being
used by only a few states and thus does not provide nationally representative data.
In addition, none of the ongoing surveys collect detailed VAW data. Some of
the surveys and surveillance systems could potentially be modified to include
additional questions related to VAW (e.g., the National Health Interview Survey and the
National Electronic Injury Surveillance System). Although several factors (e.g.,
comorbidity and etiology) are addressed by a few surveys, these surveys do not
provide incidence or prevalence estimates.
Work Group on New Research Strategies for Studying VAW
The purpose of this work group was to make recommendations for new
methods of data collection and data analysis to better understand and characterize VAW.
The group considered new data sources, ways to improve identification of VAW
in existing databases, and data linkages. In addition, they discussed new methods
of assessing a) exposure to violence and b) intervention outcomes, with emphasis
on service delivery settings that can become sources of data regarding the
prevalence and experiences of battered women.
RECOMMENDATIONS
The following recommendations, which were developed by the four work
groups, are categorized by several broad topics. Because the workshop was organized
into four work groups, similar recommendations were conceived for several topics.
Some of the recommendations could have been categorized under more than one
topic; however, to avoid repetition, these recommendations are listed only in the
most appropriate category.
Although some recommendations may seem similar, they are not identical
and were developed by different work groups and from different perspectives.
The recommendations do not reflect consensus from the entire workshop. Thus, for
each bulleted recommendation, the work group responsible for its conception is
identified in parentheses following the statement.
Defining the Scope of the Problem
CDC has initiated a process to develop and pilot test uniform
definitions associated with intimate-partner violence
(12). These uniform definitions should be used as the basis for defining and measuring VAW, with
the following modification. The term "violence and abuse against women"
(VAAW) should become standard. The "VAAW" term can provide a middle
ground between the desire not to muddle the generally understood meaning of
the term "violence" (i.e., actions that cause or threaten actual physical harm)
and the desire not to overlook psychological/emotional forms of abuse and
the trauma and social costs they cause to victims. Continuing to use only the
term "VAW" supports the misconception that a woman is only abused if she
has broken bones or other physical injuries. Both practice guidelines and
published research document the psychological and psychiatric sequelae of
violence against women (13) and the substantial use of mental health services
by victims of intimate-partner violence (14).***
(Work Group on Defining and Measuring VAW)
"Violence" is a term that encompasses a broad range of maltreatment
against women. The phrase "violence and abuse against women" should be used
to refer to the combination of all five of the following major components of
such maltreatment:
physical violence;
sexual violence;
threats of physical and/or sexual violence;
stalking; and
psychological/emotional abuse.
The first three components --- physical violence, sexual violence,
and threats of physical and/or sexual violence --- should comprise a
narrower category of VAW. Accusations have been made that VAW statistics are
falsely inflated with subjective measures of psychological abuse
(5). With the recommended terminology and classification scheme, the first
three categories can be combined and reported as VAW. All five components
of maltreatment against women can still be used to represent a larger
spectrum of behaviors harmful to women.
Consensus was reached that stalking should be included as a component
of VAAW; however, no consensus was reached regarding whether stalking
should be included in the narrower category of VAW, considered
psychological/emotional abuse, or treated as a discrete category. Whether stalking
requires the presence of a clear threat to do physical harm is an unresolved
issue. Future research on stalking may help clarify the category in which
stalking should be included.*** (Work Group on Defining and Measuring VAW)
Data should be collected on as many of the five major components of VAAW
as possible, and data collection should allow for examination of the
co-occurrence of the components.*** (Work Group on Defining and Measuring VAW)
Research, program, and public health surveillance data should
report disaggregated statistics for each of the five forms of VAAW. Presentations
of VAAW data should show cross-tabulations or Venn diagrams for all of
the forms of maltreatment.*** (Work Group on Defining and Measuring VAW)
The use of common definitions and data elements should be
encouraged. Uniformity of definitions and data elements will increase the reliability of
VAW estimates across locale and time. A CDC-sponsored panel of invited
experts developed uniform definitions and a recommended set of data elements
for intimate-partner violence surveillance that are being tested by three states
(12). In addition, guidelines for public health surveillance of
intimate-partner violence are needed on local levels, potentially serving as a model
for surveillance of other forms of VAW. Federal agencies (e.g., those
responsible for addressing the legal or public health consequences of VAW) should
jointly fund local surveillance efforts. (Work Group on State and Local Data
for Studying and Monitoring VAW)
Need for Multiple Measures/Collaboration Across
Disciplines and Agencies
Personal interview surveys (national, state, and local) are a better tool
for measuring the extent of VAW than record reviews (e.g., medical, crime,
and other service delivery); however, no single or existing tool is sufficient to
gauge and track all dimensions of VAW. Multiple data collection efforts and funding
of health, criminal justice, and social services are needed.
(Work Group on National Data for Studying and Monitoring VAW)
Because no single measurement tool can capture all of the elements of
VAAW, researchers and programs must continue drawing from existing tools
and developing new measures.*** (Work Group on Defining and Measuring VAW)
Multi-disciplinary research should be strongly encouraged.
(Work Group on New Research Strategies for Studying VAW)
Experts in several different disciplines should be encouraged to
collaborate with researchers who specialize in VAW and to initiate similar research in
their own fields. Disciplines that currently or could potentially conduct research
on VAW include anthropology, business/management, criminal
justice, demography, economics, education, epidemiology,
geography,journalism/mass communication, philosophy/ethics, psychology, public health,
social work, sociology, substance abuse, suicidology, system
analysis/operations research, theology, urban/rural planning, and women's studies. In addition
to these discipline-based groups, such collaboration might also include
persons whose research areas focus on ethnicity, the behavior of boys and men,
and research methodology (e.g., survey methodologists).
(Work Group on New Research Strategies for Studying VAW)
A chartbook or annual report should be produced to present the
current available data regarding VAW. In addition to describing the current state
of VAW, such a report would help identify areas in the data systems that
need improvement or areas in which more information is needed.
(Work Group on National Data for Studying and Monitoring VAW)
DHHS and DOJ should jointly conduct methodologic research on VAW.
Such research could focus on several issues, such as the effect of context
on prevalence estimates (e.g., health versus criminal justice) and definitions
(e.g., narrow versus broad). (Work Group on National Data for Studying
and Monitoring VAW)
Collaboration between service providers and researchers in the conduct
of research activities will improve the quality of information collected about
VAW. Such collaboration requires the development of a true partnership at the
start of research activities (i.e., a partnership that includes the joint planning
and implementation of the research methodology, presentation and dissemination
of study findings, and using the research results to refine the services for
victims and perpetrators of violence). Such partnerships between researchers
and service providers should be studied to identify the types of activities
and procedures that are most useful. (Work Group on New Research Strategies
for Studying VAW)
Developing Strategies to Collect Data on VAW
Building/Enhancing Measures of VAW
The potential of existing data sets for characterizing and monitoring VAW
should be assessed. Data can be organized into four major categories:
nationally representative surveys, local health data, local criminal justice data, and
non-nationally representative data from service providers. Ongoing,
population-basedsurveys developed for other local or state purposes should be
considered as potential opportunities for studying VAW. Other ongoing surveys that
contain questions concerning VAW (although not all are currently conducted at the
local level or in all jurisdictions) include the Pregnancy Risk Assessment
Monitoring System (PRAMS) and the National Crime Victimization Survey (NCVS).
Modules or specific questions pertaining to VAW could also be added routinely to
the Behavior Risk Factor Surveillance System (BRFSS) or the Youth Risk
Behavior Surveillance System (YRBSS). Potential sources of local health data
include emergency departments, hospital discharge records, the Health Employer
Data Information System (HEDIS), sexual assault nurse examiner (SANE)
programs, mental health databases, medical examiner data, and trauma registries.
Possible sources for local criminal justice data include databases for
misdemeanors, restraining orders, court probation, and court-case tracking. Police
departments, forensic labs, and district attorney offices may also provide local
criminal-justice data. Service-provider data might be collected from battered women
programs, rape crisis centers, protective-service programs, victim-witness advocates,
teen dating violence prevention programs, child and family services, welfare
offices, and school counselors. (Work Group on State and Local Data for Studying
and Monitoring VAW)
Questions or supplements can be added to existing continuous surveys (e.g.,
the National Survey of Family Growth, the National Health Interview Survey,
and BRFSS). Although supplements to surveys can be costly, adding questions
to ongoing surveys or conducting periodic supplements can be more
cost-effective in producing detailed data sets than creating new surveys.
(Work Group on National Data for Studying and Monitoring VAW)
As a cost-effective and efficient strategy for gathering data, questions
or modules concerning VAW could be added to numerous ongoing surveys.
This activity might be particularly useful if the survey is representative of a
well-defined population (e.g., persons living within a particular geographic region
or persons with other common characteristics) and is ongoing (e.g., following
the same persons or monitoring a changing population over time).
(Work Group on New Research Strategies for Studying VAW)
Monitoring efforts should focus on counting the number of women who
are victimized by VAAW. Future consideration should also be given to
adding measures that capture more accurately the number of perpetrators in
the population for each of the components of VAAW.***
(Work Group on Defining and Measuring VAW)
Data used for monitoring should include past year prevalence, past
year frequency, and lifetime prevalence. The lifetime prevalence
calculation represents the physical health, mental health, and social consequences that
can occur years after violence or abuse has stopped.
(Work Group on Defining and Measuring VAW)
Improved estimation of lifetime prevalence of VAW is needed. Of the
ongoing surveys, none can estimate lifetime prevalence of violence.
(Work Group on National Data for Studying and Monitoring VAW)
Etiologic and co-morbidity information periodically should be collected
(e.g., approximately every 5 years) as a supplement to a more routine
monitoring system because these data are relatively stable and because including
such information on a more frequent basis is costly.
(Work Group on National Data for Studying and Monitoring VAW)
Collecting data within various settings and populations enhances
perspectives about VAW. Data from diverse settings and populations can
provide information regarding risk factors, consequences of violence, and service
needs of particular populations as well as how victims of violence fare in
different health, judicial, or social service systems. Settings and sources of
information concerning VAW include employment locations; faith communities;
health-care settings (e.g., emergency departments, migrant-health programs,
community-health programs, maternal- and child-health programs, managed
care programs, and military/veterans health services); community-based
service agencies (e.g., welfare offices, child development and child care services,
Head Start locations, and day care centers); and programs for children (e.g.,
schools, Boys and Girls Clubs, gang programs, and programs for runaway children).
In addition, other places where women and men congregate may provide
venues for collecting information, including laundromats, hair salons, Internet
chat rooms, and job training programs. Data should be collected from
underserved populations, including Native American, Asian, Latino, and
African-American communities. (Work Group on New Research Strategies for Studying VAW)
Because some victims and perpetrators of violence never seek
violence-related services, monitoring systems should be implemented to estimate a)
the prevalence and incidence of VAW in the general community and
b) the number of persons in need of services who are not receiving them. Persons who
seek such services are not likely to be representative of all victims or perpetrators
of violence. (Work Group on New Research Strategies for Studying VAW)
A nationally representative system for monitoring VAW should be
developed. Although data from state and local agencies (e.g., social service and
criminal justice agencies) help document the extent of the problem, data from
these sources are likely to be skewed because few female victims of violence
ever seek help from those agencies. Therefore, core monitoring efforts should
be based on national samples of the total population (i.e., population-based).
In addition, BJS should explore the feasibility of developing local or
state estimates of VAW from representative samples in states, cities, or
defined metropolitan areas. However, measuring VAW (especially
intimate-partner violence, rape, and sexual assault) in smaller geographic areas is
problematic because of infrequent occurrence of VAW.
(Work Group on State and Local Data for Studying and Monitoring VAW)
Incident-based reporting that includes information on the
victim-perpetrator relationship should be employed within the criminal justice system. Use
of incident-based data would allow estimation not only of how many women
are affected by VAW but the frequency of its occurrence.
(Work Group on State and Local Data for Studying and Monitoring VAW)
Offender-based data systems should be considered for measuring and
tracking VAW. Offender-based data sources (e.g., arrests and court-based statistics)
can help estimate some elements of the VAW problem. However, these
data sources exclude victims and offenders who do not come to the attention of
the criminal justice system; hence, these data sources should not be used as a
sole method for estimating VAW. (Work Group on State and Local Data
for Studying and Monitoring VAW)
An improved identification system for homicides is needed. Only
three identified data systems --- the Supplementary Homicide Reporting
System (SHR) and NIBRS (both part of the Uniform Crime Reporting System) and
the National Vital Statistics System (NVSS) --- measure the incidence of
homicide. However, NIBRS has not been implemented nationally, SHR is
missing substantial amounts of data regarding victim-offender relationships, and
NVSS can not identify offenders or specifically identify victims of
intimate-partner violence. (Work Group on National Data for Studying and Monitoring VAW)
Building Partnerships
Each state should provide funds for a position to oversee data collection
and monitoring of VAW. The interests of both the criminal justice and health
fields must be represented, and technical assistance must be provided to state
and local entities collecting data for studying VAW.
(Work Group on State and Local Data for Studying and Monitoring VAW)
Stakeholders should be involved in the development of data systems. From
its inception, any data system should include input from victims and
service providers. Service providers need to be better informed about data systems
to understand the purposes of public health surveillance and the usefulness of
the information that such systems provide. (Work Group on State and Local
Data for Studying and Monitoring VAW)
Developing Strategies Related to Subpopulations
Data should be gathered for groups that have been omitted from
national surveys. No national studies focus on immigrant or homeless women,
women with disabilities, women in the military, or women in other
institutional populations. (Work Group on National Data for Studying and Monitoring VAW)
The terms "cultural sensitivity" and "competency" must be clearly
defined. Research strategies should then be designed to meet those definitions
and should be sensitive to the situations of victims of violence. Populations
at higher risk for VAW must be identified to ensure the implementation
of appropriate preventive and therapeutic services. Several
methodologic concerns may arise when researching VAW among persons in these
high-risk groups. The research conducted must be relevant to the community
being studied. In addition, to thoroughly understand the role of violence in the
lives of culturally diverse populations, researchers must examine both
protective factors and risk factors that may affect those populations. Developing
true partnerships with service providers and recipients may improve data
quality. (Work Group on New Research Strategies for Studying VAW)
Improving Measures of Service Provision
Service providers should be involved in local data-collection efforts, both
to enhance data collection and to encourage wider acceptance, use,
and dissemination of results. (Work Group on Defining and Measuring VAW)
Data concerning how VAW victims utilize health and social services should
be collected periodically. Collection of such data has been limited, often
because of ethical issues (e.g., privacy, confidentiality, and safety). Methods
of documenting the use of health, social, and legal services that will
not compromise the privacy and safety of the respondent should be
developed. (Work Group on National Data for Studying and Monitoring VAW)
Rigorous evaluations of the effectiveness of various services are
needed. Limited information is available regarding the effectiveness of services
for victims and perpetrators, and this information is needed to guide program
and policy development. Service providers and recipients may define
positive outcomes in different ways. Evaluation activities should address the
financial costs of various violence-related services, including primary
prevention activities. (Work Group on New Research Strategies for Studying VAW)
The feasibility of universal screening and documentation within local
health systems (e.g., emergency departments, health departments, mental
health centers, primary outpatient care centers, and school health centers) should
be investigated as a possible mechanism for surveillance of VAW. In addition,
the reliability and validity of screening questions should be assessed.
Consensus has not been reached regarding whether universal documentation of
intimate-partner violence should be used within health-care settings, because
such documentation could have negative effects for victims of VAW. For
example, documentation of repetitive injuries resulting from intimate-partner
violence could result in denial of health insurance claims or future denial of
health insurance benefits. (Work Group on State and Local Data for Studying
and Monitoring VAW)
Methodologic Concerns
When feasible, measurements should include open-ended questions
or variables. Data from such questions can be re-coded into existing categories
or may serve to clarify the need for additional categories. In situations where
data are gathered using survey methodology, these open-ended questions can
serve to humanize the data-collection process and add rapport with the
respondents. (Work Group on Defining and Measuring VAW)
Questions and data elements should be pretested (e.g., through focus
groups and in-depth interviews) to explore how respondents interpret
questions. (Work Group on Defining and Measuring VAW)
Information is needed regarding which data elements are common
across surveys and whether data can be linked. Data rarely are coordinated
between existing data sources, despite the need for comparability of estimates
across data systems. With new data sources, using variables and questions similar
to those used in existing surveys should be explored.
(Work Group on National Data for Studying and Monitoring VAW)
Several scientific methods should be used to study VAW. No "gold
standard" scientific methodology exists. The study methodology should fit the
study question being posed, and some study questions may be best addressed
by using multiple types of study designs and assessment measures.
(Work Group on New Research Strategies for Studying VAW)
Both quantitative and qualitative methods may be useful in the study of
VAW, particularly when used in combination. To better understand the complexity
of VAW, study methodologies should account for contextual issues
surrounding the violence (e.g., whether a violent episode represented a discrete event
or was part of ongoing violence in the relationship or whether violence
was defensive in nature). (Work Group on New Research Strategies for
Studying VAW)
The development and use of psychometrically sound assessment
techniques should be encouraged within all areas of VAW research, including
assessments based in service settings. Research on the reliability and validity of
various assessment techniques for measuring VAW is limited.
(Work Group on New Research Strategies for Studying VAW)
Whenever data about VAAW are reported, the actual data elements
or questions used to gather the information (i.e., the operational definitions
of VAAW) and a description of the human subjects methods used to protect
the
confidentiality and safety of those from whom data are gathered should
also be reported. Because data on VAAW can be affected by the wording of
a survey question or the method of data collection used, making this
information available allows users of the data to more accurately interpret the
numbers presented.*** (Work Group on Defining and Measuring VAW)
Establishing a unique identifier for victims of VAW is essential
for recordkeeping and protecting confidentiality. However, each system may
have its own method of coding: one victim may be assigned a unique identifier
by the local police department and another by a rape crisis center. The
feasibility of using common unique identifiers to enhance linkage across data
systems and to ensure that victim safety is not compromised should be
explored. Linking criminal-justice, health, and service-provider data for
monitoring purposes could minimize the probability of duplicating counts and allow for
the analysis of repeat victimization. Common unique identifiers would make such
a linkage feasible. (Work Group on State and Local Data for Studying
and Monitoring VAW)
The context of a survey (e.g., whether it addresses health, crime, or
personal safety issues) should be explicit to allow appropriate interpretation of
findings. (Work Group on National Data for Studying and Monitoring VAW)
Confidentiality and Safety
The safety of victims and the confidentiality of data collected must be given
a high priority. Data collected regarding VAW must be designed to
ensure confidentiality and to avoid potentially dangerous situations that
could compromise the safety of victims. (Work Group on State and Local Data
for Studying and Monitoring VAW)
The confidentiality and safety of VAW study participants must be
protected. Although standard procedures used in conducting research with
human populations should be followed, sometimes procedures must be modified
to ensure the safety of VAW victims. Although several specific actions have
been developed to increase safety for victims, no guidelines are available
for researchers concerning the safety and confidentiality issues that can arise
in VAW studies and the practices that have been used to address these
issues. Therefore, guidelines concerning confidentiality should be developed
and disseminated. For example, federal agencies could solicit papers on
these issues and then use them to prepare a handbook to guide future
research. (Work Group on New Research Strategies for Studying VAW)
The safety of staff members who conduct research (e.g., interviewers)
should also be considered. Study staff may suffer psychological distress
after interviewing multiple violence victims or may fear attack from
violent perpetrators. (Work Group on New Research Strategies for Studying VAW)
Research should be conducted on the potential effects of participating in
VAW studies. Limited empirical evidence exists concerning how participating in
such research affects study participants. (Work Group on New Research
Strategies for Studying VAW)
CONCLUSIONS
Summary remarks presented by representatives from all four work
groups emphasized that the work group deliberations represented only a beginning to
the process of developing uniformity across the numerous sectors and
disciplines concerned with VAW. Further input from researchers and practitioners concerning
the feasibility of these recommendations is needed. In addition, the
specific recommendations that are most essential to the process of building VAW
data systems must be identified. Agency leaders from BJS, NIJ, and two centers
within CDC (NCHS and NCIPC) affirmed that the Workshop itself was an initial
cross-departmental step in a long-term, coordinated
effort to improve the monitoring of VAW and to develop programs to respond to such
violence.
Acknowledgment
The following persons are acknowledged for their efforts in initiating the Workshop:
Jan Chaiken, Ph.D., Director, Bureau of Justice Statistics, Department of Justice; Mark
Rosenberg, M.D., M.P.P., Director, National Center for Injury Prevention and Control, CDC; Edward
Sondik, Ph.D., Director, National Center for Health Statistics, CDC; and Jeremy Travis, J.D.,
Director, National Institute of Justice, Department of Justice. The following persons are
also acknowledged for their leadership within the four work groups: Tim Bynum, Ph.D. (Work
Group on State and Local Data for Studying and Monitoring VAW); Nancy Isaac, Ph.D. (Work Group
on Defining and Measuring VAW); Sandra Martin, Ph.D. (Work Group on New Research
Strategies for Studying VAW); and Carol Petrie (Work Group on National Data for Studying and
Monitoring VAW). Additionally, Nancy Isaac, Ph.D., Sandra Martin, Ph.D., and Pamela McMahon,
Ph.D., M.P.H. are acknowledged for their contributions to the writing of this report.
References
Federal Bureau of Investigation. Uniform crime reports: crime in the United States
1998. Washington, DC: US Department of Justice, 1999;14.
Tjaden P, Thoennes N. Prevalence, incidence, and consequences of violence
against women: findings from the National Violence Against Women Survey --- research in
brief. Washington, DC: National Institute of Justice and Centers for Disease Control, 1998.
NCJ 172837.
Rennison CM, Welchans S. Intimate partner violence. Washington, DC: Bureau of
Statistics special report, May 2000. NCJ 178247.
Campbell JC. Promise and perils of surveillance in addressing violence against
women. Violence Against Women 2000;6(7):705--27.
DeKeseredy WS. Current controversies on defining nonlethal violence against women
in intimate heterosexual relationships: empirical implications. Violence Against
Women 2000;6(7):728--46.
Gordon M. Definitional issues in violence against women: sureillance and research from
a violence research perspective. Violence Against Women 2000;6(7):747--83.
Gelles RJ. Estimating the incidence and prevalence of violence against women:
national data systems and sources. Violence Against Women 2000;6(7):784--804.
Schwartz MD. Methodological issues in the use of survey data for measuring
and characterizing violence against women. Violence Against Women 2000;6(8):815--38.
Bachman R. A comparison of annual incidence rates and contextual characteristics
of intimate-partner violence against women from the National Crime Victimization
Survey (NCVS) and the National Violence Against Women Survey (NVAWS). Violence
Against Women 2000;6(8):839--67.
Waller AE, Martin SL, Ornstein ML. Health related surveillance data on violence
against women: state and local sources. Violence Against Women 2000;6(8):868--903.
Orchowsky S, Weiss J. Domestic violence and sexual assault data collection systems
in the United States. Violence Against Women 2000;6(8):904--11.
Saltzman LE, Fanslow JL, McMahon PM, Shelley GA. Intimate partner
violence surveillance: uniform definitions and recommended data elements. Version 1.0.
Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention
and Control, 1999.
American Medical Association. Diagnostic and treatment guidelines on domestic
violence. Chicago, IL: American Medical Association, 1992.
Wisner CL, Gilmer TP, Saltzman LE, Zink TM. Intimate partner violence against women:
do victims cost health plans more? J Fam Pract 1999;48:439--43.
*Steering Committee members from the U.S. Department of Health and Human
Services (DHHS) included Linda E. Saltzman (National Center for Injury Prevention and
Control [NCIPC], CDC), Lois A. Fingerhut (National Center for Health Statistics, CDC), James A.
Mercy (NCIPC, CDC), Jerry Silverman (DHHS), and Malcolm Gordon (National Institute of
Mental Health, National Institutes of Health). Members from the U.S. Department of Justice
included Christy Visher (National Institute of Justice [NIJ], Office of Justice Programs [OJP]),
Michael R. Rand (Bureau of Justice Statistics, OJP), and Bernard Auchter (NIJ, OJP).
**Revisions of the background papers have been peer-reviewed and published
(4--11).
***In this report, the terms "VAW" and "VAAW" are used by the Work Group on Defining
and Measuring VAW to represent different components of violence against women. This
work group suggested the use of specific terminology to differentiate the term "violence"
from "abuse." Because each work group's recommendations were not presented to the
other groups until the conclusion of the workshop, whether consensus might have been
reached by the entire workshop is unknown. In this report, the term "VAAW" was not
incorporated into recommendations from other work groups.
Disclaimer
All MMWR HTML versions of articles are electronic conversions from ASCII text
into HTML. This conversion may have resulted in character translation or format errors in the HTML version.
Users should not rely on this HTML document, but are referred to the electronic PDF version and/or
the original MMWR paper copy for the official text, figures, and tables.
An original paper copy of this issue can be obtained from the Superintendent of Documents,
U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800.
Contact GPO for current prices.
**Questions or messages regarding errors in formatting should be addressed to
mmwrq@cdc.gov.