The National Child Abuse and Neglect Data System

The National Child Abuse and Neglect Data System (NCANDS): development and utility of an existing system for collecting data and monitoring child abuse in the US John D. Fluke Vice President Research Walter R. Mc. Donald & Associates, Inc. EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE Child abuse: which kind of data for monitoring? Florence, Istituto degli Innocenti, 18 January 2007

Overview » Brief History of National US Child Maltreatment Data Collection » NCANDS Scope and Methods » NCANDS Data » Policy and Trends (examples) » Example: Recurrence » General Discussion of NCANDS and » The Value of NCANDS Like Systems January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 2

National Child Maltreatment Data: Brief History and Context » Early Studies in the US » Incidents Extracted from Media Reports (1950 s – 1970 s) » National Reporting Study on Child Abuse and Neglect » 1976 – 1988 » Other Studies of Incidence » Strauss and Gelles » Finkelhor » Prevent Child Abuse America » Major US National Data Collection Activities » National Child Abuse and Neglect Data System (NCANDS) » 1988 - » Other National Data Collection Programs » National Incidence Study (NIS) » Adoption and Foster Care Analysis and Reporting System (AFCARS) » National Study of Child and Adolescent Well Being (NSCAW) January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 3

Scope, Goals and Purposes of the NCANDS Data Collection Program » Meets US legal requirements to collect child maltreatment data » Provides data that are critical to policymakers and administrators of CPS programs » Data are used to support a range of US initiatives, notably » ASFA Outcome Measures » Child and Family Services Review Process » Office of Management and Budget Program Assessment Rating Tool (PART) » Contribute to the understanding of child maltreatment and the impact of intervention policy January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 4

NCANDS Data Collection and Analysis Methods » Sample » Universe uses no systematic “sampling” approach. » Data are administrative and derived from state Child Protective Service Agencies. » Representation » Represents only cases known to one particular service sector (child welfare). » Includes data from all US states (case level data from 46 of 52 States in 2005). » Data Collection » Aggregate Data are survey. » Case level from State information systems. January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 5

NCANDS Data Collection and Analysis Methods » Mapping and Validation of Data » Data are mapped to national categories » Data are validated for coding, internal consistency, and analytical accuracy and comprehensiveness » Measures of Maltreatment » Six major categories (Physical, neglect, medical neglect, sexual, emotional, other). » Longitudinal » Aggregate data since 1990; case level data since 1993. » Encrypted unique child and perpetrator identifiers. » Time Period » Data are collected annually. January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 6

Investigation and Victimization Rates, 1990 -2004 January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 7

January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 8

Figure 3 -3 Victimization Rates by Maltreatment Type, 2000 -2004 January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 9

Policy and Trends (Example) : Child and Family Service Review (CFSR) Maltreatment Recurrence » In August 2001, the US Children’s Bureau set the first national standard for recurrence : » A State meets the national standard if of all children who were victims of substantiated or indicated child abuse and/or neglect during the first 6 months of the period under review, 6. 1 percent or fewer children had another substantiated or indicated report within 6 months. January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 10

Child and Family Service Review (CFSR) Maltreatment Recurrence (continued) » As of June 2006 the Standard has been revised » Of all children who were victims of a substantiated or indicated maltreatment allegation during the first 6 months of FY 2004, what percent were not victims of another substantiated or indicated maltreatment allegation within the 6 -months following that maltreatment incident? » National Standard Based on 45 States: 94. 6% or higher January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 11

Survival Analysis Plot for 2002 CM Data: Recurrence During a 12 Month Period, Over 75% Recurred Within 6 Months (n = 210, 641 – 26 States) January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 12

Policy and Trends (Example) : How is Data Translated Into Policy and Programs? » Specific States not meeting standard must implement a 3 year Program Improvement Plan (PIP) approved by the US government » Plans address a range of programmatic and intervention strategies, for example: » » Multiple maltreatment and chronic neglect Alternative response systems Safety and risk assessment systems Additional research and evaluation » States not in compliance may be subject to penalties » After each three year cycle the US standards are reviewed and revised. January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 13

Policy and Trends (Example 1) : Policy Related Research Questions 1. What are the trends in recurrence rates over time? 2. In the presence of other factors that are associated with recurrence, what is the basis for a possible trend in recurrence rates over time? January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 14

Children Who Experience Recurrence Within 6 Months (12 States) January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 15

Research Questions …and Answers 1. What are the trends in recurrence rates over time? We see a decline in recurrence rates over time. January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 16

Research Questions …and Answers What factors predict the likelihood of recurrence? Child Age, Disability, Maltreatment Type, Perpetrator Relationship, Report Source, Services, Prior Victimization Does the trend in reduced recurrence operate on its own? No, the trend doesn’t exist on its own and operates best when it interacts with other factors such as maltreatment, report source, and the provision of foster care placement. Conclusion: The decline in recurrence among the States is not just a function of time but is associated with reductions in risk factors over time. The possibility that State Program Improvement Plans may have been successful in mitigating some risk factors thereby reducing recurrence cannot be dismissed. January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 17

How do the NCANDS Data Support National and State Level Policy and Intervention? » Some of examples of data supporting policy: » Problem Recognition » Re-entry (rereporting, recurrence, recidivism, etc. ) » Racial Disproportionality and Disparity » Neglect » Standard Setting » Child and Family Services Review Standard Setting » Information System Guidelines (SACWIS) » Monitoring » Child and Family Services Review » Alternative Response Implementation January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 18

General Discussion of NCANDS » Facilitators of Case Level Data Collection in the US » US Legislation » Related Programs State Automated Child Welfare » Information Systems (SACWIS) » Child and Family Service Review Process (CFSR) » System has evolved » Almost all States are Now Providing Case Level Data (46 in 2005) » Data are timely and available annually » Range of applications has expanded dramatically January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 19

NCANDS Characteristics/Features » Action (regulations, policy, programs) » Data collection program is authorized by US Office of Management and Budget (OMB) » Supports US HHS Children’s Bureau Child and Family Services Review Process (CSFR) » Support Children’s Bureau OMB Program Assessment Rating Tool (PART) measures » Evaluation » Contract is competitively bid by the US HHS Children’s Bureau » Feedback loop » Project convenes a State Advisory Group each year » Data program is reviewed for revision every three year January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 20

NCANDS Characteristics/Features » Training » Ongoing training and technical assistance for States » Quality Assessment » Evolving set of procedures for data mapping (definitions), data consistency analysis, data review. » Increased use of information system platforms to insure data integrity » Continuous Quality Improvement (CQI) process with States » Analysis » » » Annual report Descriptive analysis Compliance reports Policy studies Longitudinal studies » Dissemination » Annual reports (most recent Child Maltreatment 2004) » http: //www. acf. hhs. gov/programs/cb/stats_research/ » The National Data Archive on Child Abuse and Neglect (NDACAN) at Cornell University » http: //www. ndacan. cornell. edu/ » Special Reports, Presentations, Publications January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 21

Discussion continued » How is the existence of multiple definitions handled within the US? » Categories are general, capturing a universal view of the measurement objective » Specific definitions at the State level are mapped to NCANDS Categories » Differences are acknowledged in describing the data » States are encouraged to improve their data and address issues of consistency over time » How are different needs for various audiences such as policy, finance, performance outcomes, and research dealt with? » A range of analytic products are developed for various audiences ranging from the general public to researchers » Data are disseminated for secondary analysis » Specific analysis and analytic infrastructure is developed to meet various US reporting requirements January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 22

The Value of NCANDS Like Systems » Current and timely » Who is served » How many » Characteristics » Continuity » Ability to monitor trends » Outcomes and performance January 18, 2007 » Research » » » Longitudinal data Special populations Infrequent events Decision making Sampling frames » Assess Resource » Staffing » Cost EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 23

Disclaimer The views expressed in written conference materials or publications and by speakers, do not necessarily reflect the official policies of the Department of Health and Human Services, nor does mention of trade names, commercial practices, or organizations imply endorsement by the U. S. Government. Contact John D. Fluke Ph. D. Vice President of Research Walter R. Mc. Donald & Associates, Inc. 5445 DTC Parkway Penthouse Four Greenwood Village, CO 80111 USA (303) 488 -3437 jfluke@wrma. com January 18, 2007 EUROPEAN SEMINAR ON MONITORING SYSTEMS OF CHILD ABUSE 24