Violence in Families

Questions? Call 888-624-8373

Rosemary Chalk and Patricia A. King, Editors

Committee on the Assessment of Family Violence Interventions

Board on Children, Youth, and Families

Commission on Behavioral and Social Sciences and Education

National Research Council

and

Institute of Medicine

NATIONAL ACADEMY PRESS
Washington, D.C. 1998

NATIONAL ACADEMY PRESS • 2101 Constitution Avenue, NW • Washington, DC 20418

NOTICE: The project that is the subject of this report was approved by the Governing Board of the National Research Council, whose members are drawn from the councils of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine. The members of the committee responsible for the report were chosen for their special competences and with regard for appropriate balance.

This report has been reviewed by a group other than the authors according to procedures approved by a Report Review Committee consisting of members of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine.

This study was supported by the Carnegie Corporation of New York under contract number B5936, the U.S. Department of Health and Human Services under contract number HPU 940003, and the U.S. Department of Justice under contract 95-1J-CX-0001. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the organizations or agencies that provided support for this project.

Library of Congress Cataloging-in-Publication Data

Violence in families : assessing prevention and treatment programs / Rosemary Chalk and Patricia A. King, editors ; Committee on the Assessment of Family Violence Interventions, Board on Children, Youth, and Families, National Research Council and Institute of Medicine.
p. cm.
Includes bibliographical references (p. ) and index.
ISBN 0-309-05496-6 (cloth)
1. Family violence—United States—Prevention—Evaluation. 2. Crisis intervention (Psychiatry)United States—Evaluation. 3. Evaluation research (Social action programs)—United States. I. Chalk, Rosemary A. II. King, Patricia A., 1942- . III. Board on Children, Youth, and Families (U.S.). Committee on the Assessment of Family Violence Interventions.
HV6626.2.V56 1998
362.82'927'0973—dc21 97-45375

This book is available for sale from the National Academy Press, 2101 Constitution Avenue NW, Washington, DC 20418. Call 800-624-6242 or 202-334-3313 (in the Washington Metropolitan Area). This report is also available on line at http://www.nap.edu

Printed in the United States of America

COMMITTEE ON THE ASSESSMENT OF
FAMILY VIOLENCE INTERVENTIONS

PATRICIA A. KING (Chair), Georgetown University Law Center

JACQUELYN C. CAMPBELL, Johns Hopkins University School of Nursing

DAVID S. CORDRAY, Vanderbilt Institute for Public Policy Studies

DIANA J. ENGLISH, Office of Children's Administration Research, Department of Social and Health Services, State of Washington

JEFFREY A. FAGAN, School of Public Health, Columbia University

RICHARD J. GELLES, Family Violence Research Program, University of Rhode Island

JOEL B. GREENHOUSE, Department of Statistics, Carnegie Mellon University

THE HONORABLE SCOTT HARSHBARGER, Office of the Attorney General, Commonwealth of Massachusetts

DARNELL F. HAWKINS, Departments of African-American Studies and Sociology, University of Illinois, Chicago

THE HONORABLE CINDY LEDERMAN, Eleventh Judicial Circuit of Florida, Miami

ELIZABETH McLOUGHLIN, San Francisco Injury Center, San Francisco General Hospital

ELI NEWBERGER, Family Development Program, Children's Hospital, Boston

JOY D. OSOFSKY, Department of Pediatrics and Psychiatry, Louisiana State University Medical Center

HELEN RODRIGUEZ-TRIAS, Pediatrician/Consultant in Health Programming, Brookdale, California

SUSAN SCHECHTER, School of Social Work, University of Iowa

MICHAEL E. SMITH, School of Law, University of Wisconsin, Madison

BILL WALSH, Investigations Unit, Youth and Family Crimes Bureau, Dallas Police Department

CAROLE L. WARSHAW, Cook County Hospital, Chicago

ROSALIE WOLF, Institute on Aging, The Medical Center of Central Massachusetts, Worcester

JACK P. SHONKOFF (Liaison), Board on Children, Youth, and Families

CATHY SPATZ WIDOM (Liaison), Committee on Law and Justice, Commission on Behavioral and Social Sciences and Education

ROSEMARY CHALK, Study Director
NANCY CROWELL, Staff Officer
KATHERINE DARKE, Research Assistant
SEBLE MENKIR, Research Assistant (through August 1995)
CINDY PRINCE, Project Assistant
NIANI SUTARDJO, Project Assistant (through August 1996)

BOARD ON CHILDREN, YOUTH, AND FAMILIES

JACK P. SHONKOFF (Chair), Heller Graduate School, Brandeis University

DAVID V.B. BRITT, Children's Television Workshop, New York City

LARRY BUMPASS, Center for Demography and Ecology, University of Wisconsin

FERNANDO A. GUERRA, San Antonio Metropolitan Health District, Texas

BERNARD GUYER, Department of Maternal and Child Health, The Johns Hopkins University

ALETHA C. HUSTON, Department of Human Ecology, University of Texas at Austin

RENEE JENKINS, Department of Pediatrics and Child Health, Howard University Hospital

SARA McLANAHAN, Office of Population Research, Princeton University

ROBERT MICHAEL, Harris Graduate School of Public Policy Studies, University of Chicago

PAUL NEWACHECK, Institute of Health Policy Studies and Department of Pediatrics, University of California, San Francisco

MARTHA PHILLIPS, The Concord Coalition, Washington, D.C.

JULIUS B. RICHMOND, Department of Social Medicine, Harvard University Medical School

TIMOTHY M. SANDOS, National Digital Television Center, Littleton, Colorado

DEBORAH STIPEK, Graduate School of Education, University of California, Los Angeles

DIANA TAYLOR, Women's Health Program, Department of Family Health Care Nursing, University of California, San Francisco

GAIL WILENSKY, Project Hope, Bethesda, Maryland

ELEANOR MACCOBY (Liaison), Commission on Behavioral and Social Sciences and Education

RUTH T. GROSS (Liaison), Board on Health Promotion and Disease Prevention, Institute of Medicine

DEBORAH A. PHILLIPS, Director
ROSEMARY CHALK, Deputy Director
ANNE BRIDGMAN, Program Officer for Communications
DRUSILLA BARNES, Administrative Associate

Contents

TABLES AND FIGURES

PREFACE

EXECUTIVE SUMMARY

1 INTRODUCTION

2 FAMILY VIOLENCE AND FAMILY VIOLENCE INTERVENTIONS

3 IMPROVING EVALUATION

4 SOCIAL SERVICE INTERVENTIONS

5 LEGAL INTERVENTIONS

6 HEALTH CARE INTERVENTIONS

7 COMPREHENSIVE AND COLLABORATIVE INTERVENTIONS

8 CROSS-CUTTING ISSUES

9 CONCLUSIONS AND RECOMMENDATIONS

REFERENCES

APPENDICES

INDEX

Tables and Figures

TABLES

S-1 Total Number of Quasi-Experimental Evaluations of Family Violence Intervention by Service Sector, 1980-1996

1-1 Array of Interventions by Type of Family Violence and Institutional Setting

1-2 Total Number of Quasi-Experimental Evaluations of Family Violence Interventions by Service Sector, 1980-1996

2-1 Past Year Rates of Family Violence (per 1,000 persons)

2-2 Array of Services for Family Violence by Service Sector and Purpose

2-3 Federal Programs That Provide Services or Sponsor Research Relevant to Family Violence

2-4 Estimated Annual Costs of Family Violence

3-1 Interventions by Type of Strategy and Relevant Quasi-Experimental Evaluations, 1980-1996

3-2 Reviews of Multiple Studies and Evaluations

3-3 Outcome Measures Used in Evaluations of Family Violence Interventions

4-1 Expected Outcomes of Social Service Interventions for Child Maltreatment

4-2 Range of Family Support Interventions

4-3 Responses to Reports of Child Maltreatment by Child Protective Services

4-4 Expected Outcomes of Social Service Interventions for Domestic Violence

4A-1 Quasi-Experimental Evaluations of Parenting Practices and Family Support Services

4A-2 Quasi-Experimental Evaluations of School-Based Sexual Abuse Prevention

4A-4 Quasi-Experimental Evaluations of Intensive Family Preservation Services

4A-5 Quasi-Experimental Evaluations of Child Placement Services

4A-6 Quasi-Experimental Evaluations of Individualized Service Programs

4B-1 Quasi-Experimental Evaluation of Shelters for Battered Women

4B-3 Quasi-Experimental Evaluations of Advocacy Services for Battered Women

4B-4 Quasi-Experimental Evaluations of Domestic Violence Prevention Programs

4C-2 Quasi-Experimental Evaluation of Training for Caregivers

4C-3 Quasi-Experimental Evaluation of Advocacy Services to Prevent Elder Abuse

5A-3 Quasi-Experimental Evaluations of Court-Mandated Treatment for Child Abuse Offenders

5A-4 Quasi-Experimental Evaluations of Treatment for Sexual Abuse Offenders

5B-3 Quasi-Experimental Evaluations of Arrest Procedures

5B-4 Quasi-Experimental Evaluations of Court-Mandated Treatment for Domestic Violence Offenders

5B-5 Quasi-Experimental Evaluations of Criminal Prosecution

5B-7 Quasi-Experimental Evaluations of Systemic Approaches

6-1 Public Health Strategies for Preventing Violence and Its Consequences

6A-1 Quasi-Experimental Evaluations of Identification and Screening of Child Maltreatment

6A-2 Quasi-Experimental Evaluations of Mental Health Services for Child Victims of Physical Abuse and Neglect

6A-3 Quasi-Experimental Evaluations of Mental Health Services for Child Victims of Sexual Abuse

6A-4 Quasi-Experimental Evaluations of Mental Health Services for Children Who Witness Domestic Violence

6A-5 Quasi-Experimental Evaluation of Interventions of Mental Health Services for Adult Survivors of Child Abuse

6A-6 Quasi-Experimental Evaluations of Home Visitation and Family Support Programs

6B-1 Quasi-Experimental Evaluations of Domestic Violence Screening, Identification, and Medical Care Responses

6B-2 Quasi-Experimental Evaluations of Mental Health Services for Domestic Violence Victims

FIGURES

1-1 Family Violence Interventions by Type of Strategy

3-1 Hierarchy of Strength of Evidence in Research Evaluations

6-1 Public Health Scientific Method and Its Role in Family Violence Research

7-1 Examples of Service Integration Initiatives

8-1 Systems That Influence Family Violence and Interventions to Address Them

Preface

In May 1993, a group of 35 research scholars, state and federal officials, and representatives from law enforcement, social services, and health care systems met at the Wingspread Conference Center in Racine, Wisconsin. The purpose of this meeting was to examine whether it would be feasible to synthesize the body of research knowledge that had emerged in the past few decades regarding the development, implementation, and effectiveness of interventions designed to treat and prevent family violence. The participants agreed that efforts are needed to bridge the gap that now exists between research resources and policy needs in addressing the problem of family violence, and that one way to address this gap is to synthesize the rigorous evaluations of public-sector programs designed to treat or reduce incidents of child and spousal abuse and abuse of the elderly. They emphasized that, although no single strategy for prevention or treatment has yet proven to be effective in the research literature, the existing evaluations of relevant program interventions should be identified and analyzed to disseminate important lessons learned from past efforts to reduce family violence.

In response to the guidance of the Wingspread meeting participants, the Board on Children, Youth, and Families of the Commission on Behavioral and Social Sciences and Education (CBASSE) of the National Research Council (NRC) and the Institute of Medicine (IOM) established a Committee on the Assessment of Family Violence Interventions in August 1994. Funding was provided by several agencies within the U.S. Department of Health and Human Services (DHHS) and the U.S. Department of Justice (DOJ). The sponsoring agencies within DHHS include the Centers for Disease Control and Prevention, the Administration for Children and Families, the Office of Maternal and Child Health, the National Institute of Mental Health, and the Substance Abuse and Mental Health Services Administration. The National Institute of Justice was the DOJ sponsor. Funding was also provided by the Carnegie Corporation of New York. The Office of Health Promotion and Disease Prevention within DHHS served a valuable administrative role in coordinating the DHHS agency contributions for this project.

This study is the latest in a series of reports by the NRC that examine the emerging social science research base on violence and families. It builds on five earlier NRC publications related to this topic.

Understanding and Preventing Violence—Volume 1 (National Academy Press, 1993). This report is a comprehensive review of America's experience of violence, taking an interdisciplinary approach to examining the causes and consequences of interpersonal violence. The report includes a chapter on violence in families that describes the array of family violence interventions, research findings about police interventions and battered women's shelters, and the difficulties of evaluating and comparing interventions in this area.

Understanding and Preventing Violence—Volume 3: Social Influences (National Academy Press, 1994). This volume includes four background papers that review research on violent victimization; violence between spouses and intimates; gender and interpersonal violence; and the role of alcohol and psychoactive drugs in violent events. The paper on spousal and intimate violence by Jeffrey Fagan and Angela Browne examines the state of empirical and theoretical knowledge on violence between adult partners and presents a social epidemiology of intimate violence, characteristics of victims and assailants, and an assessment of risk markers for marital violence.

Understanding Child Abuse and Neglect (National Academy Press, 1993). This report presents a research agenda for studies of child maltreatment. It reviews the state of research on different forms of child maltreatment, including physical and sexual abuse, emotional maltreatment, and neglect. The research agenda emphasizes the importance of studies that address the nature and scope of child maltreatment, its causes and consequences, the assessment of prevention and treatment interventions, and the need for a science policy to guide the development of research in this field.

Violence and the American Family (National Academy Press, 1994). This workshop report presents a summary of the Wingspread meeting that called for the development of an in-depth analysis of the state of knowledge regarding family violence interventions.

Understanding Violence Against Women (National Academy Press, 1996). The result of a study requested by Congress in the 1994 Omnibus Crime Prevention Act, this report presents an agenda for research on intimate partner violence and sexual assault. The study identifies gaps in the knowledge base in this area and recommends a strategy for building comprehensive and interdisciplinary studies that can examine the causes and consequences, nature and scope, and prevention and intervention for violence against women.

These NRC reports provide important insights into what is known about interventions in the field of family violence. But their assessment of rigorous evaluation studies of treatment and prevention programs is limited. In this study, our committee sought to extract knowledge from research concerning the evaluations of family violence interventions as well as insights reported in other assessments of selected interventions. The committee met six times over a 24-month period to identify major conceptual themes and to review the relevant knowledge base in formulating its conclusions and recommendations. This synthesis of research and program evaluation knowledge was augmented by expert opinion through two workshops, commissioned papers, consultant reports, and five site visits designed to draw on the experiences and insights of service providers in the health, social service, and legal communities. The study also included a review of the methodological issues associated with research in areas characterized by weak conceptual clarity and immature measurement (Institute of Medicine, 1994).

The committee benefited from an inter-agency working group organized to help guide the early stages of development and the dissemination of this study and to share agency research resources. Program officers from the sponsoring agencies also participated in the committee workshops. We are grateful to each of these officials for their thoughtful contributions over the course of the study: Bernard Auchter and Christy Visher from the National Institute of Justice; Ashley Files, Matthew Guidry, and James Harrell from the Office of Health Promotion and Disease Prevention; Lynn Short from the Centers for Disease Control and Prevention; Frank Sullivan from the Substance Abuse and Mental Health Services Administration; Audrey Yowell from the Office of Maternal and Child Health; Malcolm Gordon from the National Institute of Mental Health; William Riley from the Administration for Children and Families; and Michael Levine from the Carnegie Corporation of New York.

The committee's study identification and data collection effort required an extensive staff effort; these studies appeared in dozens of journals and had not been previously assembled into a research database. Study director Rosemary Chalk and research assistants Katherine Darke and Seble Menkir, in consultation with committee member David Cordray, identified search strategies and citation indexes to gather the appropriate studies. The results of their effort are presented in Tables 3-1 and 3-2. Katherine Darke provided an important contribution in the preparation of the individual research review tables that are included in Chapters 4 through 6.

The committee held two workshops in Washington, D.C. to inform its deliberations. The first workshop was designed to elicit expertise and perspective from service providers associated with treatment and prevention interventions in child maltreatment, domestic violence, and elder abuse. Background papers prepared by the workshop participants were published in an interim report by the committee (Service Provider Perspectives on Family Violence Interventions, 1995). The participants observed that much of the information regarding family violence programs does not appear in the research literature, and that reforms in community-based interventions have not been studied in a systematic manner.

The second workshop focused on evaluation methods and research designs associated with the assessment of family violence interventions. The participants included researchers who had studied selected interventions in health care, social services, and law enforcement settings. They reviewed specific methodological challenges and creative strategies that have been used in the selection and retention of research subjects, the ethical and legal concerns associated with research in this field, and the quality of data that is associated with administrative records in public agencies.

The site visits were coordinated by Katherine Darke, who contacted local organizations and developed comprehensive itineraries for committee members and staff in each of the five cities that served as the subjects of these meetings (Boston, Dallas, New York City, Miami, and Seattle). A detailed listing of these organizations is included in Appendix A.

Several consultants provided background information that was very helpful to the committee's work. Jodi Short and Joseph Youngblood contributed materials on the nature and scope of family violence and federal intervention programs (Chapter 2). Anne Flitcraft, Patti Culross, Patricia Mrazek, and Michelle Forcier prepared background materials on health care interventions for domestic violence (Chapter 6). Terry Fulmer and Georgia Anetzberger prepared a research review on elder abuse interventions that informed several chapters. The material in Chapter 3 that pertains to client referrals, screening, and baseline assessment benefited from a publication prepared by Georgine Pion and David Cordray (Cordray and Pion, 1993). Chapter 5 benefited from contributions by Diane Juliar and Juliana Blome and a research paper on legal interventions for family violence developed by Alissa Pollitz Worden. The committee is grateful to all these contributors.

The committee also benefited from the tremendous support of the staff of the Board on Children, Youth, and Families and the Institute of Medicine: Cynthia Abel, Nancy Crowell, Katherine Darke, Seble Menkir, Faith Mitchell, Deborah Phillips, and Michael Stoto contributed careful readings, draft chapters, and literature searches that identified relevant materials throughout the development of the project. Lauren R. Meader, Julie Walko, and Susan M. Fourt of the National Research Council Library provided invaluable assistance in identifying and collecting research materials. Special thanks are due to senior project assistants Niani Sutardjo and Cindy Prince who provided administrative support during the study, including the organization of meetings, workshops, and the preparation of several drafts of the report. Project assistants Karen Autrey and Roger Butts helped to prepare the final draft for publication. Communications director Anne Bridgman was particularly helpful in the final stages of preparing the report and planning its dissemination. We also thank our editors Rona Briere and Christine McShane, whose efforts contributed significantly to the organization and presentation of the panel's views. Most of all, thanks and acknowledgment of extraordinary effort are due to the members of the committee and our study director Rosemary Chalk.

Patricia A. King, Chair
Committee on the Assessment of
Family Violence Interventions

Executive Summary

In the past three decades, family violence—which includes child maltreatment, domestic violence, and elder abuse—has emerged as a major social, health, and law enforcement issue. In addition to child and adult homicides, family violence contributes to a broad array of fatal and nonfatal injuries and medical and psychiatric disorders each year. In addition, family violence has been associated with numerous social problems, including teenage pregnancy, runaway and homeless youth, alcoholism and substance abuse, and crime and delinquency. The association of family violence victimization with such an extensive range of health, mental health, and behavioral dysfunctions suggests that interventions that can lead to the reduction or prevention of family violence would contribute to the resolution of these other problems as well.

The rate of family violence victimization in the U.S. population is widely regarded as a serious problem that affects large numbers of children and adults, but there are currently no generally agreed-on national statistics that measure the experience with family violence across the life span. Varying estimates of the scope of the different forms of family violence exist that draw on self-report surveys, crime reports, and child protective services records. Recent government surveys indicate that almost 3 million children in the United States are annually reported to child protective services agencies as alleged victims of maltreatment (including neglect, physical abuse, sexual abuse, and emotional maltreatment), and at least one-third of these cases are confirmed. National crime victimization surveys indicate that the rate of reported violent attacks by family members was 9.3 per 1,000 in 1992-1993. Other self-report surveys have indicated annual rates of domestic violence that range from 12 per 1,000 (for acts of marital rape) to 116 per 1,000 (for any act of violence). There are no self-report surveys of elder abuse, and the surveys of elder abuse reporting and recognition are incomplete.

The U.S. General Accounting Office has estimated that treatment services for child maltreatment alone (such as child protective services, child mental health services, and court expenses) cost in the range of $500 million annually. Preliminary estimates of the costs of direct services and enforcement efforts aimed at family violence in Canada range between $1.5 and $4.2 billion. These estimates are generally regarded as conservative, since family violence may be a hidden but contributing factor to injuries, disorders, and crimes for which services or enforcement efforts are provided.

The urgency and magnitude of the problem of family violence have caused policy makers, service providers, and advocates to take action in the absence of scientific knowledge that could inform policy and practice. The rush to do something has resulted in a broad array of community agency services, law enforcement approaches, and health care practices—a wealth of program experimentation that is extensive but uncoordinated. Most interventions have their origins in local and national advocacy efforts, and as such they have remained largely undocumented and unanalyzed in the research literature. Programs are often put into place without collecting baseline information about existing services or client characteristics, testing preliminary designs, or specifying—let alone measuring—the outcomes that the interventions are expected to achieve.

Existing interventions include child and adult protective services, battered women's shelters, special police and prosecution units focused on child maltreatment and domestic violence, victim advocates in health and law enforcement agencies, fatality review teams, guidelines and treatment protocols for health care providers, family support services (including home visitation and intensive family preservation services), and child advocacy centers. They can be arrayed conceptually along a continuum of strategies that include prevention, identification, protection, treatment, enforcement, punishment, and deterrence. Many services have been put into place with limited resources. The extent to which interventions have been implemented as designed is seldom documented, and often they are not fully implemented because of budget shortages in local agencies.

For the research community, service providers, program sponsors, and policymakers, the challenge is to determine if and where the research evidence is sufficient to guide a critical examination of selected components of family violence interventions. This examination is complicated by the fragmentation and uneven distribution of the research and program literature. Service providers and researchers who focus on one type of family violence or one institutional setting (social services, law enforcement, or health care) are often unfamiliar with key interventions and research evaluations in other areas. Yet there are multiple interventions focused directly or indirectly on family violence alongside each other in most communities, raising important but unexamined questions about interactive and synergistic effects.

Services are separated by the institutional settings and community jurisdictions in which they are located. Family violence interventions may focus on victims, offenders, service providers, witnesses, families, or communities. Interventions are generally specific to a certain type of maltreatment. Although many different agencies play important roles in seeking to prevent, identify, treat, or punish family violence, their efforts are largely uncoordinated. As a result, they lack shared strategies and common frameworks that could guide efforts to identify common goals, create common measures of service performance, pool resources when appropriate, and guide the implementation and development of selected interventions.

THE ROLE OF SCIENCE IN THE
ASSESSMENT OF INTERVENTIONS

In reviewing the research literature on evaluations of family violence interventions, the Committee on the Assessment of Family Violence Interventions identified 114 evaluation studies conducted in the period 1980-1996 that have sufficient scientific strength to provide inferences about the effects of specific interventions in the area of child maltreatment, domestic violence, and elder abuse. Because the committee was charged with focusing on questions about the effectiveness of interventions (rather than, for example, questions about the severity of the problem of family violence or the process of implementing interventions), we identified a core group of studies for review that included a control or comparison group (rather than simply conducting pre- and postintervention studies on the clients of a specific treatment or prevention program). Hundreds of other descriptive research studies or process evaluations have been published that provide other insights into the characteristics of an intervention and its clientele, but they do not have sufficient rigor to examine the impact or the relative effectiveness of the intervention compared with other treatment or prevention efforts.

In most cases, the set of 114 studies that forms the evidentiary base for this report also used reliable research instrumentation to measure the impact of the intervention itself on child or adult cognitive skills, attitudes, behavior, and physical and mental health as well as the impact on rates of family violence reports. However, it is important to note that comparison groups can be severely distorted by selection bias, differential attrition, and failure to consider critical unmeasured differences among groups—factors that have plagued the literature in the family violence area. In addition, the number of participants in these individual evaluation studies is often very small, which reduces their strength in examining the overall effectiveness of the intervention in addressing family violence. What can be said about the relative effectiveness of specific innovative interventions is also limited by the absence of reliable research information about the nature, quality, and impact of existing community services.

A critical review of this core group of studies illuminates both the quality of the family violence evaluation research and the developmental history of this field. Almost half of the studies identified (50 of 114 studies) involve evaluations of social service interventions for child maltreatment (see Table S-1). The majority of evaluations of domestic violence interventions (19 of 34 studies) focus on law enforcement strategies. Rigorous evaluations for elder abuse interventions in any institutional setting are almost nonexistent. Evaluations of any type of family violence intervention in health care settings (with the exception of home visitation programs) are comparatively rare.

In some areas, nevertheless, the body of research is sufficient to inform policy choices, program development, evaluation research, data collection, and theory building; the committee's conclusions and policy recommendations about these interventions are highlighted below. In other areas, although the research base is not yet mature enough to guide policy and program development, in the committee's judgment some interventions are ready for rigorous evaluation studies. For this second tier of interventions, the committee makes recommendations for future evaluation studies, also presented below. The committee has also identified a set of four topics for basic research that reflect current insights into the nature of family violence and trends in family violence interventions.

TABLE S-1 Total Number of Quasi-Experimental Evaluations of Family Violence Interventions by Service Sector, 1980-1996

Type of Family Violence

Service
Sector Child
Maltreatment Domestic
Violence Elder
Abuse

Social service 50 7 2

Legal 4 19 0

Health care 24 8 0

SOURCE: Committee on the Assessment of Family Violence Interventions, National Research Council and Institute of Medicine, 1998.

CONCLUSIONS

The committee's conclusions are derived from our analysis of the research literature as well as discussions with service providers:

• Findings from small-scale studies are often adopted into policy and professional practice without sufficient independent replication or reflection on their possible shortcomings;

• Identification and treatment interventions predominate over preventive strategies in all areas of family violence, reflecting a current emphasis on after-the-fact interventions rather than proactive approaches in the design of interventions;

• Interventions exist in an uncoordinated system of services whose effects interact on the problem of family violence in a way that presents a major challenge to their evaluation;

• Secondary prevention efforts have emerged in some areas (such as home visitation services and child witness to violence interventions) that show some promise of impact on the problem of family violence by concentrating services on targeted populations at risk;

• An increasing emphasis on the need for integration of services is stimulating interest in comprehensive and cross-problem approaches that can address family violence in the context of other problem behaviors; and

• The duration and intensity of the mental health and social support services needed to influence behaviors that result from or contribute to family violence may be greater than initially estimated.

RECOMMENDATIONS FOR CURRENT
POLICIES AND PRACTICES

It is premature to offer policy recommendations for most family violence interventions in the absence of a research base that consists of well-designed evaluations. However, the committee has identified two areas (home visitation and intensive family preservation services) in which a rigorous set of studies offers important guidance to policy makers and service providers. In four other areas—reporting practices, batterer treatment programs, record keeping, and collaborative law enforcement strategies—the committee has drawn on its judgment and deliberations to encourage policy makers and service providers to take actions that are consistent with the state of the current research base.

Recommendation 1:The committee recommends that states initiate evaluations of their current reporting laws addressing family violence to examine whether and how early case detection leads to improved outcomes for the victims or families and promote changes based on sound research. In particular, the committee recommends that states refrain from enacting mandatory reporting laws for domestic violence until such systems have been tested and evaluated by research.

In reviewing the research base associated with the relationship between reporting systems and the treatment and prevention of family violence, we observed that no existing evaluation studies can demonstrate the value of mandatory reporting systems compared with voluntary reporting procedures in addressing child maltreatment or domestic violence. For elder abuse, studies suggest that a high level of public and professional awareness and the availability of comprehensive services to identify, treat, and prevent violence are preferable to reporting requirements in improving rates of case detection.

The absence of a research base to support mandatory reporting systems raises questions as to whether they should be recommended for all areas of family violence. The committee therefore suggests that it is important for the states to proceed cautiously at this time and to delay adopting a mandatory reporting system in the area of domestic violence until the positive and negative impacts of such a system have been rigorously examined in states in which domestic violence reports are now required by law. In the committee's view, mandatory reporting systems have some disadvantages in cases involving domestic violence, especially if the victim objects to such reports, if comprehensive community protections and services are not available, and if the victim is able to gain access to therapeutic treatment or support services in the absence of a reporting system.

Recommendation 2:In the absence of research that demonstrates that a specific model of treatment can reduce violent behavior for many domestic violence offenders, courts need to put in place early warning systems to detect failure to comply with or complete treatment and signs of new abuse or retaliation against victims, as well as to address unintended or inadvertent results that may arise from the referral to or experience with treatment.

Court mandates for treatment are becoming increasingly widespread in the area of domestic violence, but the effectiveness of batterer treatment has not been examined in rigorous scientific studies. The research base does not yet suggest a specific treatment model that is appropriate for most batterers. Batterer treatment programs may be helpful for some offenders but require stronger mechanisms to enforce referrals, to establish penalties for failure to comply with program requirements, to identify and develop program components that can address the needs of different types of batterers, and to consider the unintended or inadvertent results that may be a consequence of the treatment program, such as the possible desensitizing effects of an offender's recognition that other individuals are batterers or exposure to experiences with diverse forms of violent or coercive behavior.

Recommendation 3:The committee recommends that health care and social service providers develop safeguards to strengthen their documentation of abuse and histories of family violence in both individual and group records, regardless of whether the abuse is reported to authorities.

Such documentation should be designed to record voluntary disclosures by both victims and offenders and to enhance early and coordinated interventions that can provide a therapeutic response to experiences with abuse or neglect. Safeguards are required, however, to ensure that such documentation does not lead to victim stigmatization, encourage discriminatory practices, or violate assurances of privacy and confidentiality.

Recommendation 4:Collaborative strategies among caseworkers, police, prosecutors, and judges are recommended as law enforcement interventions that have the potential to improve the batterer's compliance with treatment as well as the certainty of the use of sanctions in addressing domestic violence.

In the committee's view, collaborative law enforcement strategies that create a web of social control for offenders are worth testing to determine if such efforts can achieve a significant deterrent effect in addressing domestic violence. Collaborative strategies include such efforts as victim support and offender tracking systems that are designed to increase the likelihood that domestic violence cases will be prosecuted when an arrest has been made, that sanctions and treatment services will be imposed when evidence exists to confirm the charges brought against the offender, and that penalties will be invoked for failure to comply with treatment conditions. Creating the deterrent effect, however, requires extensive coordination and reciprocity among diverse sectors of the law enforcement and social services community that may be difficult to implement and evaluate.

Early efforts to control domestic violence through deterrence relied on single strategies (such as the use of arrest) that have now been studied. Although the arrest studies provide empirical support for the use of deterrence in dealing with specific groups of batterers, the differing effects of arrest for employed/unemployed and married/unmarried individuals call into question the reliance of law enforcement officers on arrest as the sole or central component of their response to domestic violence incidents in communities where domestic violence cases are not routinely prosecuted, where sanctions are not imposed by the courts, or where victim support programs are not readily available. What remains to be determined is whether collaborative approaches have the ability to establish deterrence for larger numbers or different types of batterers and how the costs and benefits of increased agency coordination compare with those that could be achieved by a single law enforcement strategy (such as arrest) in dealing with different populations of offenders and victims.

Recommendation 5:As part of a comprehensive prevention strategy for child maltreatment, the committee recommends that home visitation programs should be particularly encouraged for first-time parents living in social settings with high rates of child maltreatment reports.

Evaluation studies are needed to determine the factors that may influence the effectiveness of home visitation programs, including (1) the conditions under which home visitation services are provided as part of a continuum of family support programs, (2) the types of parenting behaviors that are most and least amenable to change as a result of home visitation, (3) the duration and intensity of services (including amounts and types of training for home visitors) that are necessary to achieve positive outcomes for high-risk families, (4) the experience of fathers in general and of families in diverse ethnic communities in particular with home visitation interventions, and (5) the need for follow-up services once the period of home visitation has ended.

Recommendation 6:Intensive family preservation services represent an important part of the continuum of family support services, but they should not be required in every situation in which a child is recommended for out-of-home placement.

Research findings suggest that intensive family preservation services do not show an ability to resolve underlying family dysfunction or to improve child well-being or family functioning in most families. However, methodological shortcomings in these studies suggest that measures of child health, safety, and well-being often are not included in evaluations of intensive family preservation services, so it is difficult to determine their impact on children's outcomes as well as placement rates and levels of family functioning. It is particularly important to include evidence of recurrence of abuse of the child or other family members. Intensive family preservation services may provide important benefits to the child, family, and community in the form of emergency assistance, improved family functioning, better housing and environmental conditions, and increased collaboration among discrete service systems. These services may also result in child endangerment, however, when a child remains in family environments that threaten the health or physical safety of the child or other family members.

RESEARCH RECOMMENDATIONS

Determining which interventions should be selected for rigorous and in-depth evaluations in the future will acquire increased importance as the array of family violence interventions expands in social services, law, and health care settings. For this reason, clear criteria and guiding principles are necessary to guide sponsoring agencies in their efforts to determine which types of interventions are suitable for evaluation research. Recognizing that all promising interventions cannot be evaluated, public and private agencies need to consider how to invest research resources in areas that show programmatic potential as well as an adequate research foundation. To assist in this evaluation selection process, the committee developed the following guiding principles:

1. An intervention should be mature enough to warrant evaluation.

2. An intervention should be different enough from existing services that its critical components can be evaluated.

3. Service providers should be willing to collaborate with the researchers and appropriate data should be accessible in the service records.

4. Satisfactory measures should exist to assess service processes and client outcomes.

5. Adequate time and resources should be available to conduct a quality assessment.

With these principles in mind, the committee has identified a set of interventions that are the focus of current policy attention and service innovation efforts but have not received significant attention from research. In the committee's judgment, each of these nine interventions has reached a level of maturation and preliminary description in the research literature to justify their selection as strong candidates for future evaluation studies: (1) family violence training for health and social service providers and law enforcement officials; (2) universal screening for family violence victims in health care and child welfare settings; (3) comprehensive community initiatives; (4) shelter programs and other domestic violence services; (5) protective orders; (6) child fatality review panels; (7) mental health and counselling services for child maltreatment and domestic violence; (8) child witness to violence prevention and treatment programs; and (9) elder abuse services.

The committee identified four research topics that require further development to inform policy and practice. These topics raise fundamental questions about the approaches that should be used in designing treatment, prevention, and deterrence strategies:

• Cross-problem research (such as the relationship between substance abuse and family violence);

• Studies of family dynamics and processes that interact with family violence;

• Cost analysis and service system studies that describe the existing set and distribution of services focused on different forms of family violence;

• Social setting issues that warrant study because of their implications for the design of treatment, support, prevention, and deterrence strategies.

CHALLENGES TO EFFECTIVE EVALUATIONS

In recommending this research agenda, the committee recognizes that such research is complicated by a number of factors. Service providers and researchers are realizing that family violence is an interactive, dynamic, and complex problem that requires approaches across multiple levels of analysis (individual, family, community) and multiple service systems. The presence or absence of policies and programs in one domain may directly affect the implementation and outcomes of service strategies in another. Services also interact with the characteristics of the client: some people need more support or stronger sanctions, depending on their histories and life circumstances; others need only a limited amount of assistance, treatment, or sanction to improve.

The lack of collaboration between researchers and service providers has impeded the development of appropriate measures and study designs in assessing the effectiveness of programs. It has also discouraged research on the design and implementation of service interventions and the multiple pathways to services that address the causes and consequences of family violence.

To improve evaluations of family violence interventions and to provide a research base that can inform policy and practice, major challenges must be addressed. These challenges include issues of study design and methodology, as well as logistical concerns, that must be resolved in order to conduct rigorous research in open service systems in which many factors are not under the control of the research investigator. Meeting these challenges will require collaborative partnerships among researchers, service providers, and policy makers to generate approaches and data sources that are useful to all. The establishment and documentation of a series of consensus conferences on relevant outcomes, and appropriate measurement tools, will strengthen and enhance evaluations of family violence interventions and lead to improvements in the design of programs, interventions, and strategies.

The development of the next generation of evaluation studies will benefit from the building blocks of knowledge that have been put into place over the past 15 years. This research base, and the convergence of the field around such issues as the recognition of the interactive nature of the service system in different institutional settings, the existence of multiple subgroups of offenders, the need for research experimentation to guide treatment and prevention efforts, and the use of multiple measures of program outcomes suggest that a richer and deeper understanding of family violence interventions lies within reach in the decade ahead.

Top of Document | NAP Home Page | Document Home Page