U.S. Department of Justice Office of Justice Programs Bureau of Justice Statistics Improving Access to and Integrity of Criminal History Records July 2005, NCJ 200581 -------------------------------------------------------------- This file is text only without graphics and many of the tables. A Zip archive of the tables in this report in spreadsheet format (.wk1) and the full report including tables and graphics in .pdf format are available from: http://www.ojp.usdoj.gov/bjs/abstract/iaichr.htm -------------------------------------------------------------- Peter M. Brien Bureau of Justice Statistics Lawrence A. Greenfeld Director Peter M. Brien, Statistical Policy Advisor, BJS wrote this report. Todd D. Minton, BJS Statistician, verified the findings. Tina Dorsey produced the report, and Tom Hester edited it. Contents Achievements of the National Criminal History Improvement Program Missing dispositions in criminal history records Opportunities for improving background checks Appendix I. Identifying prohibited purchasers: The mentally ill Appendix II. Identifying prohibited purchasers: Domestic violence misdemeanants and subjects of restraining orders Appendix III. BJS surveys in the aftermath of September 11, 2001 Office of Justice Programs Partnerships for Safer Communities Achievements of the National Criminal History Program NCHIP overview The National Criminal History Improvement Program (NCHIP) represents a close collaboration among the Bureau of Justice Statistics (BJS), State criminal justice agencies, and the Federal Bureau of Investigation (FBI). Since 1995, this collaboration has improved the nation's public safety by enhancing and upgrading the States' criminal history records which are used for background checks for firearms purchases, pre-employment checks for certain sensitive professions, criminal sentencing decisions, and many other purposes. The NCHIP program has facilitated the development of a national network of State criminal records which is available to State and federal law enforcement officials. The program has provided direct financial and technical assistance to States to develop and upgrade their criminal record information systems -- including records of protection orders, sex offender registries, and automated fingerprint identification. NCHIP has been instrumental in making these records accessible on an interstate basis through FBI administered systems - * the National Instant Criminal Background Check System (NICS) * the Interstate Identification Index (III) * the National Protection Order File * the National Sex Offender Registry * the Integrated Automated Fingerprint Identification System (IAFIS). Since 1995 BJS has awarded nearly $400 million to States and Territories. During the period of funding the number of criminal history records has increased 29% nationwide, and the number of automated records available for immediate use by law enforcement increased 35%. As of December 31, 2001, there were nearly 64.3 million subjects in State criminal record files, and 89% of these records were automated. This highlights section presents an overview of 11 performance measures for NCHIP: * Millions of background checks are supported by the NICS * The percentage of State criminal history records which are automated has improved * The number of records that are shareable through the FBI's III has increased * The number of States participating in III has increased * More law enforcement agencies are reporting arrest information electronically * The number of States participating in the FBI's IAFIS has increased * More courts are transmitting automated dispositions * Criminal history information processing has become more efficient * The National Protection Order File has significantly expanded * The National Sex Offender Registry has reached record levels * NCHIP has established a framework to assist in Homeland Security. These measures summarize how NCHIP has assisted States in improving criminal history record completeness and automation. The National Instant Criminal Background Check System (NICS) supports millions of background checks NCHIP funds have supported State efforts to conduct rapid and efficient checks under the NICS. In 2001, the NICS supported nearly 8 million checks at the presale stage of firearms purchases. Assisted by NCHIP funding, the State NICS infrastructure made the transition from an interim system to the current permanent system. Under the interim Brady system, the chief law enforcement officer conducted checks. Currently States operate background check systems of their own (Point-of-Contact States or POC States), rely exclusively on the NICS, or do both. From the inception of the Brady Act on March 1,1994, to December 31, 2001, nearly 38 million applications for firearm transfers were subject to background Checks. --------------------------------------------- Data sources for measuring access to and integrity of criminal history records Almost a decade before the inception of NCHIP in fiscal year 1995, BJS formed a partnership with the States and the FBI to improve the accuracy and accessibility of criminal history records. After passage of the Brady Act, expanded use of those records enlarged both BJS' role as administrator of funds to improve record systems and BJS' role as statistical monitor to meas-ure efficiency and completeness. The following provided data for this report: * Since 1989 BJS has funded the biennial Survey of Criminal History Information Systems, conducted by SEARCH, The National Consortium for Justice Information and Statistics. * In 2001 and 2002 BJS initiated research with the Regional Justice Information Service (REJIS) and SEARCH. These collaborations addressed issues related to the completeness and accuracy of criminal history records, mental health records, restraining orders, and misdemeanor domestic violence records. * Upon passage of the Brady Act and creation of NCHIP, BJS initiated the Firearm Inquiry Statistics program as part of NCHIP. The program operates through a cooperative agreement with REJIS. --------------------------------------------- Number of firearm purchase applicants: 7,806,000 A. Checks conducted by NICS: 4, 248,900 1. Number denied for purchase by NICS: 60,700 (1.4%) (a) Number of denials appealed: 10,400 (17%) (Number of denials reversed: 3,200) (b) Number of denials not appealed: 50,300 (83%) 2. Number approved for purchase by NICS: 4,188,200 (98.6%) (a) Retrievals due to erroneous sale: 3,400 B. Checks conducted by POC's: 3,556,900 1. Number denied for purchase by POC's: 75,200 (2.1%) (a) Number of denials appealed: 13,500 (18%) (Number of denials reversed: 6,300) (b) Number of denials not appealed: 61,800 (82%) 2. Number approved for purchase by POC's: 3,481,700 (97.9%) (a) Retrievals due to erroneous sale: 3,400 Sources: BJS; Bureau of Alcohol, Tobacco, Firearms and Explosives; and the NICS Program of the FBI ------------------------------------------------ Of the nearly 8 million checks conducted in 2001, 98% of applicants were approved whether the check was conducted by the State or the FBI.***Footnote 1: This 2% rejection rate is consistent with the rejection rates from 2001, 2000, 1999, and 1998 and during the interim Brady period from 1994 to 1998.*** There are two principal ways to evaluate the performance of the background check system: * Examine the percentage of applicants for a firearm purchase who were incorrectly prohibited (Error I) * Examine the percentage of persons who were allowed to purchase a firearm who should have been prohibited from doing so (Error II). Error I can be measured by calculating the ratio of successful appellants of a rejection to the total number of applicants to purchase a firearm. In 2002 there were 9,500 successful appeals of rejections from among the 7,806,000 applicants that year. This translates into an error rate of 0.1%. Error II can be measured by calculating the ratio of firearm retrievals by law enforcement to the total number of applicants to purchase a firearm. In 2002 there were 7,400 retrievals due to an incorrect sale to a prohibited purchaser against a total of 7,806,000 applicants. This also translates into an error rate of approximately 0.1%. Among the 136,000 applicants rejected as prohibited purchasers during 2002, 8 in 10 did not appeal the rejection. Among those who filed an appeal, almost 6 in 10 rejection decisions were sustained. Levels of State criminal record automation have increased There have been notable improvements in the automation of State criminal history records since 1995. As the number of criminal records has increased in recent years, the NCHIP program has helped the States to ensure that levels of automation remain high. In 2001, 30 States had at least 90% of their criminal history records automated, compared to 22 States in 1995. The number of States with relatively low levels of automation (defined as having less than 70% of records automated) declined from 13 States in 1995 to 6 States in 2001. 2001 represents the first year in which more than half of the States (27) indicated that 100% of their criminal history records were automated. In 1999, 21 States reported having 100% automation, as did 18 States in 1995. Number of records shareable through the FBI's Interstate Identification Index (III) has increased The number of Interstate Identification Index (III) records maintained by the States as compared to the records maintained by the FBI is an important measure of the decentralized nature of the background check system. To ensure that criminal history records are compatible across all States, record improvements funded by BJS under NCHIP are required to conform to FBI standards for III participation. BJS has designated State participation in the III as one of the highest priorities of the NCHIP program. From yearend 1993 to April 2003, the number of State records available for sharing under the FBI's III nearly doubled from 25.5 million to 48.5 million. Since 1993 the number of records available for III has been increasing at a much faster rate than the overall number of criminal records. In 1993 half of all criminal history records were III accessible, and 70% were in 2001. It is projected that in 2005, more than three-fourths of all criminal history records will be accessible through the III. At yearend 2001, 22 States had at least 75% of their records accessible through III, and 15 States had 50% to 74% of their records III accessible. Number of States participating in III has increased As part of NCHIP's commitment to achieving full State participation in the FBI's III system, BJS has used the NCHIP program to encourage all States to make their records available to the FBI and other States. In 1989, 20 States were members of the III system. The number of participating States grew to 26 in 1993. As of May 2003, 45 States were participating in III. The application for NCHIP funds requires nonparticipating States to include a discussion of their III status and a plan for future III participation. More law enforcement agencies are reporting arrest information automatically State law enforcement agencies have frequently relied on the NCHIP program to help them transition from ink-based fingerprinting systems to electronic systems. From 1999 to 2003 NCHIP has provided States with approximately $31 million for Livescan systems and their participation in the FBI's automated fingerprint system. These systems allow fingerprints to be scanned and instantly transmitted electronically to criminal justice agencies throughout the State or Nation. Among the 37 States reporting complete data, the number of arresting agencies that electronically transmit arrest information and fingerprints increased by more than 2,000 from 1997 to 2001. Nationwide, 2,594 agencies electronically submitted arrest information in 2001, a substantial increase from 493 agencies in 1997. An increasing number of States participate in the FBI's Integrated Automated Fingerprint Identification System (IAFIS) NCHIP funds are available to all States for the conversion to IAFIS. As of March 2003, 43 States, the District of Columbia, and 3 Territories participated in IAFIS. In fiscal year 2002 NCHIP distributed over $3.8 million to 19 States and the District of Columbia to purchase equipment for the automatic capture and transfer of fingerprint images. More courts are transmitting automated dispositions Whether the States can maintain complete and accurate criminal history records depends heavily on the ability of the courts to report final dispositions automatically to the criminal history repository or court administrator's office. For several years NCHIP has been committed to supporting the record and case management systems of State and local courts. From 1997 to 2001 NCHIP funds assisted nearly 725 courts to develop the ability to report final dispositions automatically. As of yearend 2001, over 2,000 courts were reporting disposition information electronically, based on 25 States reporting complete data. As part of NCHIP's support for capture of judicial decisions for criminal history records, participating courts across the United States have developed the ability to -- (1) electronically link fingerprints and other biometric information to dispositions and share this information with corrections and law enforcement agencies (2) provide electronic protection order/restraining order files that can be accessed by law enforcement officers in the field (3) establish and maintain sex offender registries (4) convert juvenile records to the adult case management system (5) establish databases of offender-based information. Processing criminal history information becoming more efficient Since 1995 NCHIP has assisted States in developing and implementing integrated criminal history record systems. Through NCHIP support, new information technologies have enabled State criminal justice agencies to share critical information more efficiently. The average time required for agencies to transmit arrest information, final court dispositions, and prison admissions to criminal history repositories has declined significantly since 1995. In addition, these repositories are processing and posting this information more quickly than in 1995. From 1995 to 2001 the average time between an arrest and the addition of fingerprints and information about that arrest to a criminal history record was cut nearly in half from 45 days to 24 days. The average time required to receive, process, and add final court disposition information to a criminal history record declined more than 30%, from 68 days to 46 days. In 1995 information about an individual's admission to prison took an average of 94 days to be received and then posted to a record. By 2001 this information was processed 3 times faster, in 31 days. Despite progress from 1995 to 2001, not all States have adopted integrated or automated systems for transferring criminal history information. In 2001 States were still more likely to transmit information through the U.S. Postal Service than by electronic or online transmissions. The National Protection Order File has significantly expanded A 1994 amendment of the Federal Gun Control Act made illegal the sale of firearms to persons subject to a qualifying protection order. A qualifying protection order meets conditions that refer to domestic violence. NCHIP has placed special emphasis on ensuring that domestic violence-related offenses and protection orders are included in criminal records. Funds have been specifically awarded for development of State protection order files that are compatible with the FBI's national file, NCIC National Protection Order File, to permit interstate enforcement of protection orders. The National Protection Order File, the fastest growing national set of criminal history records maintained by the FBI, has tripled in size since 2000. On February 3, 2003, the National Protection Order File held over 750,000 protection orders from 43 States. The National Sex Offender Registry has reached record levels Since 1998, the NCHIP program has assisted States in establishing sex offender registries which interface with the FBI's National Sex Offender Registry. This assistance has enabled State sex offender registry information to be obtained and tracked from one jurisdiction to another. The National Sex Offender Registry has seen dramatic growth in recent years, from nearly 50,000 records in 2000 to over 280,000 records in 2003. On February 3, 2003, the National Sex Offender Registry contained records from every State, the District of Columbia, and 3 Territories. NCHIP helped to create tools for accomplishing the work of Homeland Security A signal contribution of NCHIP has been to assist the States and the FBI in construction of a nationwide network of criminal history and fingerprint archives. The contents can be examined for both criminal justice purposes (for example, presale firearms checks) and noncriminal justice purposes, such as employment security clearances. In 1992 State repositories held 36.4 million automated criminal history records, and 4 States relied completely on paper records. Since 1995 NCHIP has assisted the States' transition toward automation, so that in 2001 State repositories held 57.4 million automated criminal history records. As of March 1, 2003, there were 48.5 million criminal records that were shareable through III. Future challenges for NCHIP With a history of success, the role of NCHIP has evolved to support the States in pursuing emerging technologies and new criminal justice priorities as identified by Federal and Sate policymakers. According to the administrators of NCHIP, there exist areas of opportunity where Federal resources will continue to be needed. Discussed in greater detail in the following pages, these future challenges for NCHIP include: Updating relevant mental health records -- BJS has conducted surveys to examine the status of mental health records among the States. Analysis of responses can help to determine how NCHIP can further enable States to make the information in mental health records available for background checks. Improving access to records for domestic violence misdemeanor convictions -- Since 1996, when Congress added conviction for a domestic violence misdemeanor to the list of Federal firearm prohibitors, NCHIP has worked with the States to improve access to these records; however, significant challenges remain. Recent BJS surveys examined the persistent impediments to accessing these criminal records. Ensuring that court disposition reporting systems are automated -- All court disposition information should be transmitted electronically using established technologies. The automation of court dispositions will provide complete and accurate information electronically to Federal and State criminal justice officials. Encouraging prosecutors to complete criminal history records by reporting their declinations to prosecute -- When prosecutors decline to prosecute arrestees, frequently the declinations are not added to the criminal records. Checks of these records require significant effort to discover whether the arrested individual was prosecuted. Converting older paper records into an electronic format -- Many State criminal history record holdings contain older records on paper. Full automation awaits the entering of these records into the system. Linking criminal history transactions with NIBRS incident reports -- This linkage allows criminal history records to reflect a wealth of detailed information concerning the nature and characteristics of criminal incidents. Incident-based information allows for better informed decisions regarding criminal punishment, crime control strategies, and policy analysis. Developing a uniform national criminal history format for RAP sheets -- Because of variation in criminal history record formats among the States, the interstate exchange and interpretation of criminal history information are unnecessarily complicated. The development and adoption of a uniform format for RAP sheets will make criminal history information more easily understood across jurisdictions. Continuing to address privacy and confidentiality issues as they relate to non-criminal justice background checks -- The Prosecutorial Remedies and Other Tools to end the Exploitation of Children Today Act of 2003 or PROTECT Act contains provisions that will make it easier for private organizations that work with children, the elderly, or the disabled to conduct background checks on volunteers. The PROTECT Act established a pilot program that permits volunteer organizations to submit 100,000 requests for background checks to Federal authorities. States will want to ensure that privacy and confidentiality protections are respected as non-criminal justice background checks become more prevalent. ------------------------------------ NCHIP goals NCHIP remains committed to its responsibility to provide Federal assistance to State criminal justice record systems. That bedrock commitment remains as the program addresses these challenges: * Identify how NCHIP can be more effective in promoting the use of State mental health records during background checks * Create initiatives to remove impediments to the availability of conviction records for domestic violence misdemeanors * Continue NCHIP efforts to introduce technology into all courtrooms so that disposition records will be transmitted electronically to criminal history record repositories * Develop procedures to encourage prosecutors to promptly report declinations to the repositories * Continue NCHIP's encouragement of States to remove backlogs of paper records * Support the development and adoption of a uniform RAP sheet * Work with States to ensure that privacy and confidentiality protections are respected as background checks for noncriminal justice purposes become more prevalent. ---------------------------------------------- Missing dispositions in criminal history records In a 2001 survey, half the States reported that from 10% to over 50% of the arrests recorded in State databases had no final disposition indicating how the arrest was resolved. These arrests that lack dispositions, known as "open" or "naked" arrests, create substantial problems for time-sensitive background checks because conducting the necessary research to complete the record is often time consuming, labor intensive, and costly. BJS examined the scope of missing dispositions as reported by the States in a 2001 survey. Factors that contribute to missing dispositions include the following: * poor communication among the courts, law enforcement, and corrections agencies in providing final disposition information to the State repository * reliance on the mail and manual research rather than on electronic communication * repository delays in posting disposition information to a criminal record * persistent backlogs of criminal justice records in criminal history repositories * failure of States to audit their data-bases to ensure record completeness * varying definitions of "missing dispositions" among States * inconsistent procedures among States for handling missing dispositions. Missing dispositions or "open arrests" an extensive problem In a 2001 survey, State repository directors were asked what percentage of arrests in their databases had final dispositions recorded. Six States reported that 90% or more of their arrests had corresponding final dispositions. Twenty-three States reported that between 50% and 89% of arrests had dispositions. In nine States less than half of the arrests had final dispositions recorded in the databases. Twelve States and the District of Columbia could not estimate the percentage of arrests with dispositions. The repository directors also reported the percentage of arrests occurring within the last 5 years ("recent arrests") that contained final dispositions. Eight States reported that 90% or more of their recent arrests had final dispositions attached. Twenty States reported that between 50% and 89% of recent arrests had dispositions. In 12 States less than half of the recent arrests in their database had final dispositions recorded. Ten States and the District of Columbia could not estimate the percentage of recent arrests with dispositions. Another measure of the missing disposition problem is the number of final court dispositions in State files that can not be linked to arrest or charging information. In 2001, 14 States estimated the number of unlinked court dispositions. These States reported a total of 704,300 final court dispositions on file that could not be linked to arrest or charging information, an average of over 50,000 unlinked dispositions per State. While 14 States estimated the number of unlinked final court dispositions, 16 States estimated the percentage of final court dispositions in their files that could not be linked to an arrest. Minnesota and Indiana reported that half of their final court dispositions could not be linked to arrest or charging information. Wisconsin, Kansas, and Pennsylvania reported that between 30% and 40% of their final court dispositions were unlinked. In Michigan and Nebraska a quarter of their final court dispositions had no link to arrest or charging information. The remaining nine States (Georgia, Maryland, Montana, Nevada, New York, South Carolina, South Dakota, Utah, and Virginia) reported that 10% or fewer of their final court dispositions contained no arrest or charging information. Poor interagency communication contributes to missing dispositions Interagency communication can be evaluated in three ways: (1) by the number of days required for the court, law enforcement agency, or corrections agency to report disposition information to the repository (2) by the extent to which criminal justice agencies rely on manual systems rather than electronic systems to transmit disposition information and research missing dispositions (3) by whether all criminal justice agencies are fully aware of their reporting requirements. Disposition data took over a week on average to reach repositories A leading reason why final dispositions are not adequately linked to their associated arrests is delay by the courts, law enforcement, and corrections agencies in reporting disposition information to repositories. Another important reason, processing delays by the repository, is described more fully in a following section. Court disposition information required 3 weeks on average On average in 2001, it took 21 days for final court dispositions to reach the State criminal history repository.***Footnote 2: The average number of days for final court disposition information to reach the repository was based on 1995-2001 data from a cohort of 28 States.*** The time required for repositories to receive disposition information ranged from 1 day or less in six States to 80 days in North Dakota. Repositories in most of the reporting jurisdictions (21) received final court disposition information within 15 days. The average time required for final court disposition information to reach the repository steadily decreased from 36 days in 1997 and 33 days in 1999. In 2001 final court dispositions reached the repository 14 days faster on average than they did in 1995; how- ever, as six States have demonstrated, the average time for submission could be 1 day rather than 21 days. Arrest information required 11 days on average In 2001, as in 1999, arrest data and fingerprints took 11 days on average to be received by State repositories. The time required to receive arrest information ranged from 1 day or less in five States and the District of Columbia to 169 days in Mississippi. In 22 States and the District of Columbia, repositories received arrest information within 7 days. In 1995 and 1997 transmission of arrest information to the repository took an average of 13 days and 14 days, respectively.***Footnote 3: The average number of days for arrest information to reach repositories was based on 1995-2001 data from a cohort of 40 States and the District of Columbia.*** Corrections information required 9 days on average In 2001 an average of 9 days elapsed between a prison admission and the receipt of that information by the State repository. The time required for repositories to receive prison admissions ranged from 1 day or less in nine States to 60 days in North Carolina. Repositories in just over half of the reporting jurisdictions (17) received prison admission information in 7 days or less. In 2001 the average prison admission reached the repository 5 times faster than in 1995, when 45 days were required. In 1997 and 1999 prison admissions reached the repository in 16 days on average. ***Footnote 4: The average number of days for prison admission information to reach repositories was based on 1995-2001 data from a cohort of 25 States.*** Prosecutor declinations often did not reach repositories In 2001, 12 States were able to estimate the number of prosecutor declinations -- decisions to drop or modify charges -- sent to the repository in the prior year. Thirty-two States indicated that State law or policy required that prosecutorial declinations be transmitted to criminal history repositories. Local prosecutors often fail to report declinations to State record repositories. In 2001 five States indicated that all prosecutorial declinations were transmitted to the repository. In 1999 three States, and in 1997 six States, reported that all prosecutors' declinations were transmitted to the repository. BJS, with the American Prosecutors Research Institute, is scheduling a survey to delineate more clearly the processes by which disposition information reaches or fails to reach a repository. The study will examine the following: whether statute, regulation, office practice, or some other authority governs the transmission of declination information by prosecutors; what penalties (if any) exist for noncompliance; the average time from declination to the notification of the repository; and the impediments that exist in reporting declinations promptly to the repository. Postal services used instead of electronic communication Trial courts provide most of the information about final dispositions. Criminal history repositories in the 47 States and the District of Columbia receive final disposition information from trial courts or the State court administrator's office. Fifteen of those repositories receive all the information through the Postal Service; 14 repositories receive part of that information by mail. Thirty-five States request final disposition information from prosecutors. In 16 States the repositories receive all information from prosecutors by post; in 12 States, part of that information. Twenty-six States request law enforcement agencies to report information on final dispositions to the repository. In 13 States law enforcement agencies rely entirely on the Postal Service; in 11 States, partially. Repositories receive final disposition information from appellate courts or court administrators in 28 States. More than half of these States (15) rely on the Postal Service for delivery of all this information; five rely on the Postal Service for some of the information. Manual systems used to research missing dispositions A 2001 BJS survey revealed that while researching criminal records with missing dispositions, agencies in 46 States rely on manual procedures. Most of these States (34 of the 46) use only manual systems to locate missing final dispositions. Only Colorado, Michigan, Pennsylvania, and South Dakota fully utilize automated means to obtain missing final dispositions. Agencies unaware of their reporting requirements In certain States the criminal history repository does not capture final disposition information because the agencies responsible for providing this information are not aware of all reporting requirements. Other States report that courts and law enforcement agencies are maintaining incomplete or inaccurate statistics on final case dispositions. Arrest and corrections information does not always reach repositories The transmission of arrest and corrections information to the repository is not required by law or policy in all States. In 39 States and the District of Columbia, there are policies to provide the repository with data on the admission or release of felons from State prison. In 26 States and the District of Columbia, there are policies to give the repository information on the admission or release of a felon from a local jail. Thirty-one States and the District of Columbia have policies to provide probation information to the repository, and 30 States and the District of Columbia have policies to provide parole information. If an arrestee is not charged with a crime after the fingerprints are submitted to the repository, 19 States indicated that State law does not require that the repository be notified of the failure to charge the arrestee. Criminal history repositories face challenges in processing records Recent BJS surveys have suggested that criminal history repositories are encountering several problems including significant backlogs, older records that have no dispositions, and infrequent audits to ensure accuracy of records. The backlogs faced by criminal history repositories can be measured in two ways: (1) by the length of time between the repository's receipt of information and when that information is entered into a record (2)by the number of forms that await entry into the record. By both measures, State repositories face significant challenges that must be met to improve the NICS background check process. Overall average processing times by repositories decreased, 1995-2001 Separate from the issues whether the repositories receive final court dispositions, arrest information, or correctional information, and whether the received information is recent, is how well a repository processes information it does receive. Many State repositories are unable to quickly process critical information and include it in the record, despite progress in recent years). Final court dispositions Repositories took an average of 25 days to process final court disposition information and post it to a criminal record in 2001. The time required to post court disposition information ranged from 1 day or less in 12 States to 330 days in Washington. The majority of responding jurisdictions (23) entered these data in 10 days or less. Average processing time has decreased since 1995 and 1997 when repositories took 33 and 34 days, respectively, to post final court dispositions. In 1999 an average of 24 days was required.***Footnote 5: The average number of days required for repositories to process and post final court dispositions to a record was based on 1995- 2001 data from a cohort of 32 States and the District of Columbia.*** Prison admission information -- In 2001 an average of 22 days elapsed between the repository's receipt of prison admission information and the posting of that information to the record. The time required to post prison admission information ranged from 1 day or less in nine States to 90 days in Illinois. The majority of responding jurisdictions (21) enter these data in 10 days or less. The average number of days required to post prison admission information has declined significantly in recent years. In 2001 repositories processed and posted prison admission information in nearly half the time required in 1999, when the process took an average of 43 days. The average processing time for prison admissions in 2001 was considerably shorter than in 1995 and 1997 when processing took 49 days and 29 days, respectively.***Footnote 6: The average number of days required for repositories to process and post prison admission information to a record was based on 1995-2001 data from a cohort of 28 States.*** Arrest information -- Repositories took an average of 13 days to post arrest information to a criminal record in 2001. Eleven State repositories posted arrest information in 1 day or less. Oklahoma took 180 days -- the longest time period of any State. Most States (27) posted arrest data in 7 days or less. In 2001 repositories posted arrest information to a criminal record in less than half the time required in 1995, when the process took an average of 32 days.***Footnote 7: The average number of days required for repositories to process and post arrest information to a record was based on 1995-2001 data from a cohort of 41 States and the District of Columbia.*** In 1997 and 1999 State repositories took an average of 31 days and 28 days, respectively, to process arrest information and post it to a record. Repository backlogs exceed 2.5 million records In 2001 most States reported that they were encountering backlogs in entering certain criminal history information into State criminal databases. Twenty-six States reported a backlog in entering arrest and fingerprint information; 27 States had a backlog of court disposition forms; and 12 States had a backlog of custody supervision reports. Repositories in States that could estimate the size of their backlogs in 2001 reported that 2.5 million records of arrest, disposition, and custody information were unprocessed or only partially processed. The backlog comprised over 2 million court disposition forms, 350,000 fingerprint cards, and 130,000 custody supervision reports. In 2001 State repository backlogs were smaller in all three categories of information than they were in 1995. Compared to 1997, however, the total backlog increased 50% in 2001. Most of the increase after 1997 was attributable to unprocessed court disposition forms, increasing from 940,000 in 1997 to approximately 2 million in 2001. The number of unprocessed custody supervision reports nearly tripled from 1997 to 2001 from 46,500 to 134,000. From 1997 to 2001, the backlog of unprocessed fingerprint cards was cut in half from 710,000 to 354,000. The composition of State criminal record backlogs remained unchanged in recent years. Since 1995 court disposition forms have comprised the majority of State criminal record backlogs, followed by arrest fingerprint cards and custody supervision reports. Because not all States can estimate the size of their repository backlogs, the true number of unprocessed or partially processed criminal justice records is likely to be higher than the estimates reported here. Older records on paper or microfilm account for large portion of "open arrests" Many State criminal history records contain arrests from several years ago for which no final disposition was reported. At the request of BJS, the FBI drew a sample of NICS checks that could not be completed instantly due to the presence of "open arrests." More than 75% of the arrests from this sample were from 1984 or earlier. Audits of criminal history records remain infrequent In 2001, 23 State criminal history repository directors reported that their databases had not been audited for completeness in the prior 5 years. An audit would have told them how accurate and complete were the criminal histories in their holdings. Over half of those States (13) reported that they had not planned or scheduled a data quality audit to occur within the next 3 years. Overall, 24 States did not plan to perform a data quality audit within 3 years of the survey. Of the 27 States with completed audits after 1995, changes were made to improve data quality as a result of 22 of these audits. National standards can help assure complete criminal history records The problem of missing dispositions in State criminal history records is perpetuated in part because there is no national consensus on how a missing disposition is defined, measured, or researched by State agencies. Improvements in criminal history records have fallen short of earlier BJS efforts to establish voluntary reporting standards.***Footnote 8: Recommended Voluntary Standards for Improving the Quality of Criminal History Record Information (56 FR 5849).*** The presence of a national standard on the time required for final court disposition information to reach the repository would encourage better performance by States that currently process this information more slowly. National standards are lacking on the following five issues concerning criminal history record completeness. States define "missing dispositions" differently States' definitions of an "open" arrest or "missing disposition" vary. It is difficult to accurately identify missing dispositions and obtain a national estimate of the extent of the problem because the term has different meanings in different States. Most States (34) and the District of Columbia consider any arrest event or charge without a disposition to be a "missing disposition," while 16 States require a certain amount of time to pass before an arrest event or charge is officially considered a missing disposition. Ten States reported that they define a missing disposition as an arrest that is not being actively prosecuted. These 10 States rely on a variety of methods to determine that an arrest is not being actively prosecuted, including notification by the prosecutor or court, searching court records, and regular audits of records. States are divided as to whether they measure a missing final disposition by using an arrest event or an arrest charge. An event has one or more charges. Twenty-one States measure a final disposition by an arrest event; 17 States identify it by an arrest charge; and 10 States and the District of Columbia measure a missing final disposition by both an arrest event and an arrest charge. There are also two exceptions to the basic information used to determine whether a disposition is missing. Vermont uses arraignments rather than arrests, and Pennsylvania does not specify any particular event to measure a missing disposition. States require long delays before designating a disposition as "missing" Sixteen State repositories require that time elapse before a disposition is officially considered to be missing. If the subject of a background check has been convicted of a disqualifying offense and the conviction is not included in the record, in certain States months will pass before the record is designated as having a missing disposition. During those months the subject will not be prohibited from purchasing a firearm because the NICS check will not reveal the disqualifying conviction. Of the 16 States that require a certain amount of time to pass before an arrest event or charge will be officially considered to have a missing disposition. Two States require that more than 1 year must pass before a record without a disposition is officially declared to have a missing disposition (Connecticut and Washington). In seven States an entire year must elapse before a disposition is considered missing (Alaska, Indiana, Kansas, Michigan, Missouri, South Carolina, and Virginia). In three States a disposition is not counted as missing until more than 6 months have passed (Arizona, Maine, and New York). The remaining four States use another time interval. State procedures for handling missing dispositions differ When State repositories receive new disposition information that cannot be linked to an arrest charge or event, States follow varying policies. From July 1, 2000, to July 1, 2001, 45 States and the District of Columbia reported receiving final court dispositions that could not be linked to arrest information in the criminal history record. When final court dispositions could not be linked to an arrest, 27 States did not enter the unlinked information. Eight States created a "dummy" arrest segment with the information from the court disposition record, and three States entered the court information without any linkage to a prior arrest segment. The remaining 12 States and the District of Columbia followed other procedures. In that same 1-year period, 34 State repositories received information from corrections agencies that could not be linked to arrest information in the criminal history record. When States received correctional information for which there was no underlying court information, 25 States imputed a conviction for the record. Nineteen States and the District of Columbia reported that they did not enter a conviction on the record based on information from corrections agencies. When information from corrections agencies cannot be linked to an arrest, the information is not entered into the system in 14 States and the District of Columbia. Six States create a "dummy" arrest segment with the correctional information, and 15 States enter the correctional information without any linkage to a prior arrest segment. From July 1, 2000, to July 1, 2001, 32 State repositories received information from final prosecutor dispositions that could not be linked to arrest information in the criminal history record. When prosecutor information could not be linked to an arrest, 24 States and the District of Columbia did not enter the unlinked information. Six States created a "dummy" arrest segment with the unlinked prosecutor information, and two States entered the prosecutor information without any linkage to a prior arrest segment. Common identifiers absent in criminal history records The States use a wide variety of identifiers to link disposition information with the arrest or charging information, and no single identifier is used by all 50 States. Forty States and the District of Columbia link dispositions by using a unique arrest event identifier; 28 States and the District of Columbia use a unique tracking number for an individual suspect; 22 States and the District of Columbia link dispositions by using the arrestee's name; and 16 States and the District of Columbia link dispositions by using a name and a reporting agency case number. Certain States use other identifiers, such as arrest date, date of birth, or Social Security number of the arrestee, court docket number, or charge number. None of the States require positive identification that would occur with a biometric identifier like fingerprints. Many States unable to link final dispositions to specific charges Disposition information in a criminal record is often not linked to specific charges or counts. In 33 States and the District of Columbia, the method used by the repository to link disposition information to the arrest/charge information allows the link to be made to specific charges or counts, an arrest cycle. State criminal history record systems can be compared only with difficulty if some systems are appreciably more sophisticated and complete than others. Dispositions should be linked to specific charges in all States. Accurate linkage of dispositions to criminal records results from proper use by courts of relevant arrest transaction numbers and the courts' direct submission of a livescan with the associated disposition of the arrest. Courts have indicated a growing interest in utilizing NCHIP funds to purchase livescan equipment. BJS sponsors National Evaluation of NCHIP In response to the Attorney General's June 2001 directive, BJS commissioned a 3-year evaluation of the National Criminal History Improvement Program (NCHIP). The two purposes of the evaluation were -- (1) to measure the performance of State criminal history records systems (2) to identify improvement activities that would be most effective in improving record completeness. The evaluation ensured that the NCHIP grant program more efficiently addressed strengths and weaknesses of State criminal history record systems. Future NCHIP grants will be targeted to the specific record system deficiencies identified for each State. The Records Quality Index(RQI)to measure NCHIP performance In 2001 BJS began development of a new criminal records performance measurement system. The system will better guide NCHIP provision of funds and technical assistance to the States. The records quality index (RQI) collects and summarizes performance measures that objectively assess the relative capabilities of each State's criminal history record system. The RQI assigns a system an overall rating based on a wide variety of performance measures directly related to NCHIP objectives. A State's RQI can be compared to the RQI of other States to determine the relative strength of the State's criminal history record system. The RQI will also allow for more specific analysis, permitting BJS to track the progress of a State by the score received on each of its performance measures. The overall RQI a State receives can be compared to a national average to express whether the overall efficiency of the State's criminal history record system meets, exceeds, or is below the impartial summary of all States. After complete data are received, the RQI will become an invaluable tool for identifying the strengths and weaknesses of each State's criminal history record system. Using RQI information, NCHIP will target funds to activities that will most significantly improve a State's RQI, thereby improving national background check systems more directly and quickly. As of March 2004, 34 States have completed all RQI data submissions. Another 10 States have made partial submissions and have demonstrated substantial progress toward supplying all RQI data. Elements of the RQI offer distinct vantage points The evaluation for the RQI will break down a State's criminal history record system into a wide variety of performance measures that describe the system's components, overall structure, and efficiency. One measure of the efficiency of a State criminal history record system is the average time that elapses before an arrest report is posted to the criminal history repository. Other ways a system's utility can also be assessed are by the percentage of arrest records that are linked to dispositions and by whether State's criminal justice agencies and the State's judicial system use common identifiers. A State will receive a high score on one component of the RQI if all of its criminal history records are automated and will receive a high score on another component if law enforcement agencies are connected to the courts electronically. The RQI consists of three sets of performance measures: input measures, process measures, and outcome measures. Input measures describe the components of a record system, including the following: State laws requiring the fingerprinting of prisoners and transmission of the fingerprints to the State repository; use of unique tracking numbers in the records for individual subjects; and requirements that prosecutors submit declinations to the repository. Process measures reflect system performance, including the following: procedures followed by the repository when a link can not be made between a disposition and an arrest; strategies employed to ensure accuracy of arrest data, fingerprints, and final court dispositions received at the repository; the average time between the occurrence of a final felony court disposition and the entry of data into the criminal history database; and the automation and auditing of certain procedures. Outcome measures reflect the impact of the State's system and whether the records are accessible and complete. Outcome measures include the following: the percentage of dispositions that cannot be linked to an arrest; the number of III records available; backlogs in entering information into the database; and repository notification of prosecutor declinations. Criminal histories integrated with NIBRS reveal key crime factors NIBRS Status of the States, August 2003 NIBRS certified Idaho Utah Colorado Texas Kansas Nebraska South Dakota North Dakota Iowa Arkansas Louisiana Tennessee Kentucky Wisconsin Michigan Ohio West Virginia Virginia South Carolina Connecticut Massachusetts Rhode Island Vermont New Hampshire Testing NIBRS Maine Delaware New York New Jersey Pennsylvania Indiana North Carolina Oklahoma New Mexico Montana Washington Oregon Developing NIBRS California Arizona Minnesota Illinois Missouri Alabama Hawaii Maryland No formal plans regarding NIBRS Nevada Wyoming Mississippi Florida Georgia Alaska Source: FBI, Criminal Justice Information Services, Division, Program Support Section, Greg Swanson, NIBRS Implementation Coordinator The National Incident-Based Reporting System (NIBRS) appreciably advances efforts to learn important details about crimes in the United States. Beyond the summary counts of eight offenses presently collected for the FBI Uniform Crime Reports (UCR), NIBRS represents a more sophisticated data collection system. NIBRS enables State and local jurisdictions to capture detailed offense, offender, victim, property, and arrest information concerning each crime incident. NIBRS goals are -- * to enhance the quantity, quality, and timeliness of crime data collected by law enforcement * to improve the methodology used in compiling, analyzing, auditing, and publishing collected crime statistics. The broader scope of offense and offender characteristics also permits the development of policy-relevant findings. These findings can describe the nature and characteristics of crimes such as gun assaults, domestic violence, hate crimes, and crimes against children. Through NIBRS, law enforcement officials, analysts, and the public can better understand trends in more types of crime and in the amount of harm to victims. Jurisdictions that have incident-level reporting systems are more capable of using crime data for strategic planning, tactical crime analysis, and manpower deployment. This ability should result in more effective use of limited resources and better crime control strategies. Incident-level data, such as NIBRS, provide detailed information on the contingencies of a crime, including victim characteristics and consequences for victims. Such data, when linked to criminal records, would permit a background check to determine whether elements of the offense meet specific prohibitions for firearm transfer or other factors of concern. Incident reports would contain the victim's characteristics, the victim/ offender relationship, and the offender's use of a firearm -- establishing whether the victim was a child or whether the offense involved domestic violence. The FBI and BJS have been active in assisting the States in meeting the NIBRS standards and becoming certified for NIBRS data submission. In 2001 BJS distributed $13 million to 26 States through the NIBRS Implementation Program to develop and operate NIBRS-compatible systems. As of August 2003, 24 States had their State UCR programs approved by the NIBRS Certification Board, according to the FBI's NIBRS Implementation Coordinator (map). Twelve States were testing their systems to determine if they will soon meet certification standards. Nine States were developing a NIBRS system, and six (Alaska, Georgia, Florida, Mississippi, Nevada, and Wyoming) had no formal plans to provide NIBRS data to the FBI. Opportunities for improving background checks Percentage of criminal history records accessible through the Interstate Identification Index (III), by State, December 2001 States with at least 75% of records III-accessible Arizona California Colorado Florida Georgia Idaho Iowa Maryland Michigan Minnesota Mississippi Montana Nebraska New Jersey New Mexico Nevada North Carolina South Carolina South Dakota Virginia West Virginia Wyoming States with 50-74% of records III-accessible Alaska Arkansas Delaware Illinois Indiana Missouri New York Ohio Oklahoma Oregon Pennsylvania Tennessee Utah Washington Wisconsin States with less than 50% of records III-accessible Alabama Connecticut District of Columbia Hawaii Kansas Kentucky Louisiana Maine Massachusetts New Hampshire North Dakota Rhode Island Texas Vermont Since 1995 NCHIP has operated under a broad mandate to improve public safety by enhancing the quality, completeness, and accessibility of the Nation's criminal history record information systems. NCHIP has funded State and local initiatives designed to improve criminal records and to allow these records to be used for criminal justice and authorized noncriminal justice purposes. Achievements of NCHIP were highlighted on pages 1 to 6. A BJS review of the NCHIP program has concluded that the program's resources must be more aggressively targeted toward bringing the States into compliance with the Brady Act's provisions. NCHIP must also become fully responsive to evolving needs for more extensive and thorough background checks to protect homeland security. To reflect these NCHIP management objectives, five recommended action items are to be incorporated into NCHIP: * obtaining the full participation of the States in the FBI's Interstate Identification Index (III) * improving State contributions to the FBI's national databases of prohibited purchasers * formally establishing a National Technical Assistance Center supported by BJS * promoting the adoption of a uniform RAP sheet to standardize records among the States * periodic BJS surveys of State criminal record holdings. Action item 1: Obtain full State participation in the FBI's Interstate Identification Index (III) Serving as the backbone of the NICS system, the Interstate Identification Index (III) is a decentralized index-pointer system maintained by the FBI for exchanging criminal history records. The III connects the FBI's computerized files with State-level computerized files. The III contains personal identifiers for more than 48 million offenders. Federal, State, and local criminal justice agencies may access it to determine whether an individual has a criminal record anywhere in the country. The III's "pointer" specifies the Federal or State source that maintains any relevant record(s). Once a hit is made against the Index, the subject's data are automatically retrieved from each repository and forwarded to the requesting agency. The FBI estimates that the III database contains 94% of the records checked by the NICS; therefore, it is critical that all States enter full partnership and contribute their identifiers for inclusion in III. States generally enter participation in the III system in two phases, defined by the nature of the records that they agree to provide. For the second phase, that of full participation, a State agrees to make its III-indexed records available to Federal and State agencies for both criminal justice and noncriminal justice purposes. In the second phase the FBI ceases to maintain duplicate criminal records for persons arrested and prosecuted in fully participating States. All users of criminal history records -- for criminal justice purposes as well as authorized noncriminal justice purposes -- will obtain records directly from States' central computerized files or from the FBI for Federal records. Fully participating States must meet minimum standards that relate to several factors, including the completeness and maintenance of the records, the presence of fingerprints for each record, and procedures to respond to record requests. At yearend 2001 a quarter or more of the criminal history records in a majority of States could not meet those standards. At the end of 2002, 22 States had at least 75% of their records accessible through III (map). Thirteen States and the District of Columbia had less than half of their records accessible through III. The States and the Federal Government will benefit in several ways when all States become full III participants. First, the duplicate maintenance of criminal history records by the States and the FBI will be eliminated, resulting in significant cost savings. No longer will States have to submit to the FBI fingerprints and charge/disposition data for all felony and serious misdemeanor arrests. Instead, States will submit only fingerprints and textual identification data for each person's first arrest to update the III and the National Fingerprint File (NFF). No longer will the FBI keep records on State offenders and process fingerprint cards for all State arrests. Instead, the FBI will maintain the III, the NFF, and full criminal records of Federal offenders. Second, records made available on an interstate basis will be more complete for both criminal justice and noncriminal justice purposes, including the screening of applicants for certain sensitive positions. State repository records are more current than FBI files. In addition, many States maintain misdemeanor records that have not been submitted to the FBI. These records will become available through the III system for NICS and other authorized uses. Third, certain noncriminal justice users will enjoy faster response times because they will be receiving electronic responses rather than mailed record responses from the FBI. These noncriminal justice users may also benefit from increased efficiency at the State level resulting from increased automation and system improvements preceding III participation. Action item 2: Encourage States to contribute records to the national databases of persons prohibited from purchasing firearms: the NICS Index and National Crime Information Center files A background check for a firearm transfer involves the use of records available through the III and two other databases maintained by the FBI: (1) the NICS Prohibited Persons Index (NICS Index), containing records of persons prohibited by the Gun Control Act to purchase a firearm -- those dishonorably discharged from the armed services, persons who have renounced U.S. citizenship, mental defectives, controlled substance abusers, and illegal/unlawful aliens (2) the National Crime Information Center (NCIC) Files, containing records on sex offenders, wanted persons, convicted persons on supervised release, and subjects of protective/ restraining orders. FBI statistics show that States have contributed infrequently to the NICS Index and NCIC databases. Protection of U.S. residents depends on these databases of prohibited purchasers being complete and current. The NICS Index The NICS Index is divided into six databases of persons disqualified under Federal law from receiving firearms: (1) the Denied Persons File (2) the Mental Defectives/Commitments File (3) the Controlled Substance Abusers File (4) the Illegal/Unlawful Aliens File (5) the Citizenship Renunciates File (6) the Dishonorable Discharges File. State law enforcement officials contribute to only the first four of the six databases. In the following summary a State is listed as having contributed to a database as of February 1, 2003, without regard to the number of records involved. Several contributing States provided fewer than 10 active records. ---------------------------------------------- NICS Index databases show minimal State contributions The NICS Index increased from 3,222 State records in February 1999 to 88,696 in February 2003. Nearly 90% of this increase can be attributed to two States. In February 2002 Vermont added 22,205 records to the Denied Persons File. In November 2002 Michigan added 52,698 records to the Mental Defectives File. As of February 2003 few States had contributed records to the FBI's NICS Index. The Denied Persons File had the largest number of contributing States (12), followed by the Mental Defectives File (7), the Controlled Substance Abusers File (3), and the Illegal/Unlawful Aliens File (2). Since November 1999 State contributions to the NICS Index files have increased for all databases. NICS database status as of February 1, 2003: Denied Persons File contains information on persons who have been denied the purchase of a firearm under Federal law but for whom disqualifying information is not contained in any other file accessed by the NICS. Federal and State law enforcement officials are eligible to enter records. The Denied Persons File contained 34,904 records contributed by 12 States. There were 7,514 Federal records in this file. Mental Defectives/Commitments File contains information from State law enforcement officials, the Department of Defense, and the Department of Veterans Affairs. The file includes persons who have been adjudicated as mentally defective or have been committed to a mental hospital. The Mental Defectives/Commitments File contained 53,551 records contributed by 7 States. Federal agencies have contributed 90,111 files. Controlled Substance Abusers File contains information on persons who are unlawful users of or addicted to any controlled substance. State law enforcement officials, the U.S. Coast Guard, and the Department of Defense are eligible to enter records into this file. The Controlled Substance Abusers File contained 90 records contributed by 3 States and 76 Federal records. Illegal/Unlawful Aliens File contains information provided by the INS and State law enforcement officials on illegal or unlawful aliens. The Illegal/Unlawful Aliens File contained 151 records from 2 States. Federal agencies have contributed 2,683,910 records. --------------------------------------------- State law enforcement contributes to four databases of National Crime Information Center(NCIC) File The NCIC system is a nationwide, computerized database of criminal justice information. The NCIC File is divided into eight databases: (1) the Protection Order File (2) the Convicted Persons on Supervised Release File (3) the Convicted Sex Offender Registry File (4) the Wanted Persons File (5) the Deported Felon File (6) the SENTRY file (7) the U.S. Secret Service Protective File (8) the Foreign Fugitive File. State law enforcement officials contribute to only the first four databases. From January 2000 to February 2003, the number of State and Federal records submitted to the NCIC database nearly tripled from 849,610 to 2,289,320. The increase in the number of NCIC offender records available for background checks can be largely attributed to the increases in the contributions to the National Sex Offender Registry (450% increase) and the National Protection Order File (220% increase). Since January 2002 the number of records in each NCIC file profiled in this report has increased. NCIC record status as of February 2003 Protection Order File contains court orders that are issued to prevent acts of domestic violence against a person or to prevent a person from stalking, intimidating, or harassing another person. The Protection Order File contains 754,033 records from 43 States. Federal agencies have contributed 56 protection orders to this file. Convicted Persons on Supervised Release File contains records on subjects who are put under specific restrictions during their probation, parole, or supervised release sentence following imprisonment. The Convicted Persons on Supervised Release File contained 260,556 records from 15 States. There were three Federal records in this file. Convicted Sexual Offender Registry File contains records on individuals who have been convicted of a criminal offense against a minor, who have been convicted of a sexually violent offense, or who are sexually violent predators. The Convicted Sexual Offender Registry File contained 280,329 records from all States and the District of Columbia. This file contained 10 records from Federal agencies. Wanted Person File contains records on persons for whom a federal warrant, a felony warrant, or a serious misdemeanor warrant is outstanding. The Wanted Person File contained 857,835 records from all States and the District of Columbia. Federal agencies have contributed 42,076 records on wanted persons. Action item 3: Establish a National Technical Assistance Center A National Technical Assistance Center (NTAC) will ensure that all State agencies that develop and use criminal history records are technologically advanced with compatible information technology systems. The importance of system designs which ensure compatibility with other systems to facilitate criminal and noncriminal background checks has never been more acute. The NTAC will assist States to have the capabilities and integration necessary to meet homeland security concerns -- especially in regard to communications and access to and security of information. NTAC will help States to determine whether proposed systems comply with FBI/NICS participation requirements and meet other needs for background checks. BJS surveys conducted in response to the Attorney General's directive demonstrate a clear need for the NTAC. The surveys found that agencies in 46 States rely on manual rather than automated procedures while researching missing dispositions. Of the 46 States, 34 use only manual systems to locate missing final dispositions. Twelve States rely on both manual and automated procedures, and 4 States rely completely on automated means to obtain missing final dispositions. A difficulty that besets State system development is the continuous need to replace both hardware and software as new technology becomes available. Statewide plans less than a year old are frequently determined to be obsolete when the systems become operational. The NTAC will provide guidance to ensure that information technology systems are appropriate, integrated, and efficient in the short and long term. Action item 4: Implement uniform RAP sheets Because Federal and State authorities, as well as private entities, increasingly utilize criminal history records for homeland security background checks, criminal justice purposes, and noncriminal justice purposes, it is more important than ever to make those records as uniform and readable as possible. The interstate exchange and interpretation of criminal history information are needlessly complicated because the States and the FBI use different formats and codes. State criminal history records often contain symbols, or notations, commonly referred to as "flags." Many flags are based on requirements under State law. When shown on interstate RAP sheets, undefined flags can confuse readers. BJS has long encouraged States to convert their criminal history records to the FBI standard interstate RAP sheet format.***Footnote 9: See, Increasing the Utility of the Criminal History Record: Report of the National Task Force, NCJ 156922, December 1995.*** More recently, a national task force, the Joint Task Force on RAP Sheet Standardization, determined that adoption of a uniform RAP sheet by all States and the FBI would benefit users of criminal history records. Information in those records would be more easily understood, more complete, and more accurate. Adoption of a uniform RAP sheet would also provide decision makers with the ability to combine or electronically collate criminal records from multiple sources into a single chronology of events. The set of data elements for the uniform RAP sheet is intended to include all pertinent information without overburdening systems with the collection of extraneous information. NCHIP funds should be used to assist the States in adopting uniform RAP sheets that contain the following five elements: (1) a clearly defined link between an arrest and the subsequent activities and dispositions (2) sufficient information to allow users to identify each agency that contributed particular information (3) a notice of the existence of sealed information that some users may be authorized to obtain (4) computer interfaces to enhance the quality and efficacy of the criminal history record (5) a standard transmission format to ensure that information from a State can be easily formatted into the model presentation format, even if the contributing State has not adopted the model format. Action item 5: Survey State criminal history record holdings periodically The efficiency of the NICS depends entirely on the levels of completeness, accuracy, and accessibility of State criminal history records. Through the NCHIP program BJS observes the performance of State criminal history record systems. The monitoring identifies the critical areas of improvement that will have the largest impact on improving the Nation's record systems. BJS monitors State record systems in two ways: (1) Record Quality Index The RQI will be used to critically evaluate the strengths and weaknesses of a State's criminal history record system. The RQI will measure improvements over time based on a several key performance measures. NCHIP will use the RQI results to work with the States and ensure that funds are directed to the areas of greatest need in each State. The RQI will ensure accurate performance measurement as well as responsive and timely grant administration. (2) Survey of State Criminal History Information Systems Since 1989 BJS has conducted a biennial data collection that examines State criminal history record systems. This survey provides important details by State on the level of automation of State record systems, the level of disposition reporting and felony flagging, the timeliness of criminal record processing, measures of criminal history record completeness, and State procedures to ensure the accuracy and security of criminal records. Latest survey findings are available in Survey of State Criminal History Information Systems, 2001 Previous surveys were completed in 1989, 1992, 1993, 1995, 1997, and 1999. Appendix I Identifying prohibited purchasers: The mentally ill The Gun Control Act as amended prohibits the sale of a firearm to several categories of potential purchasers, including persons who have been adjudicated as a mental defective or have been committed to a mental institution. Regulations of the Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) define this mental health category as -- * individuals who have been adjudicated as a mental defective, meaning -- "(a) A determination by a court, board, commission, or other lawful authority that a person, as a result of marked subnormal intelligence, or mental illness, incompetency, condition, or disease: (1) Is a danger to himself or to others; or (2) Lacks the mental capacity to contract or manage his own affairs. (b) The term shall include -- (1) A finding of insanity by a court in a criminal case; and (2) Those persons found incompetent to stand trial or found not guilty by reason of lack of mental responsibility pursuant to articles 50a and 72b of the Uniform Code of Military Justice."***Footnote 11: See 27 CFR 178.10.*** * persons who have been committed to a mental institution, meaning -- "formal commitment of a person to a mental institution by a court, board, commission, or other lawful authority. The term includes a commitment to a mental institution involuntarily. The term includes commitment for mental defectiveness or mental illness. It also includes commitments for other reasons, such as for drug use. The term does not include a person in a mental institution for observation or a voluntary admission to a mental institution."***Footnote 11: See 27 CFR 178.10.*** This appendix examines -- * the number of States that as part of a background check are unable to regularly access court orders or involuntary commitments for mental illness * the principal factors that prohibit mental health information from being regularly accessed during background checks* BJS initiatives to improve access to State mental health records* the results of a 2002 BJS survey that covered the practices of the 14 States that consult mental health records during background checks. Mental health information rarely included in NICS background checks Impediments to the timely identification of categories of prohibited purchasers are significant and have been well documented in recent reports by BJS and the General Accounting Office(GAO).***Footnote 11: Continuing Criminal History Records Improvement Evaluation: Final 1994-98 Report, BJS report, NCJ 179768, February 2000. (All BJS reports are available at .) Survey of State Procedures Related to Firearms Sales, Midyear 2002, BJS report, NCJ 198830, April 2003. Use and Management of Criminal History Record Information: A Comprehensive Report, 2001 Update, BJS report, NCJ 187670, December 2001. Opportunities to Close Loopholes in the National Instant Background Check System, General Accounting Office, GAO-02-720, July 12, 2002. Improving the National Instant Criminal Background Check System, General Accounting Office, GAO-T-GGD-00-163, June 21, 2000. Gun Control: Options for Improving the National Instant Criminal Background Check System, General Accounting Office, GAO/GGD-00-56, April 12, 2000.*** In March 2002 BJS surveyed all State criminal history repository directors to determine whether mental health records were included in the State's background checks for firearm sales. Twenty-five States and the District of Columbia do not serve as points of contact for background checks, referring all background checks to the NICS system. None had contributed mental health records or any related mental health information to the FBI for use during presale firearm checks. Of the remaining 25 States that serve as points of contact, 11 do not check mental health records while conducting presale firearms checks, and 14 do (Map). Few States make their mental health records available to the FBI's Mental Defectives File. As of February 1, 2003, the NICS Index contained records of the mentally ill or incompetent from seven States. Michigan provided over 50,000 records, but no other State contributed more than 30 records to the file. To protect the privacy of the mentally ill, a number of States have opted to provide records on individuals prohibited from possessing a firearm for mental health reasons to the FBI's Denied Persons File instead of the Mental Defectives File. Several factors prohibit complete mental health information from being accessed during background checks Point of contact States that check mental health records California Oregon Washington Utah Nebraska Wisconsin Illinois Michigan Pennsylvania Virginia Connecticut New Jersey Georgia Point of contact States that do not check mental health records Nevada Arizona Colorado Iowa Indiana New York Vermont Maryland North Carolina Tennessee Florida Not point of contact States. They do not provide mental health information to the FBI's NICS Index. Idaho Montana Wyoming North Dakota South Dakota Kansas New Mexico Oklahoma Texas Minnesota Missouri Arkansas Louisiana Misssissippi Alabama Kentucky South Carolina West Virginia Massachusetts New Hampshire Maine Delaware Rhode Island State mental health records incomplete and infrequently automated A 2002 BJS survey of the States revealed that the condition of the mental health record systems often renders those records unusable for background checks. Findings include the following: * Nearly 40% of the States (18) reported the lack of an interface between the State's mental health database and the criminal history repository. * For 12% of the States (6), mental health records were incomplete. * Nearly a third of the States (15) reported that their mental health records were not fully automated. * In 16 States officials reported that mental health records lacked biometric identifiers like fingerprints to positively identify an individual. Condition of State mental health records also precludes their transmission to the NICS Index. The reasons given for not sharing the mental health information with the FBI included: * Incomplete records (8 States) * Automation issues (30 States) * No biometric identifiers (15 States). State privacy laws protect mental health information When State officials were surveyed as to the factors that explain why complete mental health information could not be accessed during background checks, they cited privacy laws more frequently than any other factor. Nearly 40% of the States reported that State privacy laws prevented access to complete mental health information during background checks.***Footnote 12: Citations to privacy laws were provided by officials from Alaska (Alaska Stat. sec 40.25.120), Colorado (Colo. Rev. Stat. sec 25-1-1202), New Hampshire (N.H. Rev. Stat. sec 151:13), North Carolina (N.C. Gen. Stat. sec 122C-52), Wisconsin (Wis. Stat. sec 51.30), and Wyoming (Wyo. Stat. sec 25-10-122).*** Officials from 30 States reported that their privacy laws also prevented the State from providing complete mental health information to the FBI for inclusion in the NICS Index. Agencies lack time and resources to access mental health records In 2002 just over a quarter of the States reported that they did not have sufficient time or resources to access mental health information during background checks. Just under a quarter cited insufficient resources as one factor preventing them from sharing mental health records with the FBI. BJS develops 6 initiatives to widen access to mental health records BJS has developed six new initiatives to assist the States in upgrading their ability to consult mental health information during background checks as required by the Gun Control Act. * BJS will formulate an initiative to write a Model Data Sharing Agreement. This collaborative effort among the States and BJS will develop State-specific proposals for making State mental health records accessible for firearm background checks. * Under the Best Practices Initiative, BJS will work with selected States to develop a best practices guide. This guide will highlight critical issues that must be addressed before mental health records can be included in database searches for disqualifying information. The guide will describe State practices in use and will follow up promising practices that emerge after the publication of the guide. * The Mental Health Courts Initiative will be a collaboration with the Bureau of Justice Assistance (BJA) to improve the criminal justice system's supervision of mentally ill offenders. BJS will work with BJA and their grant recipients to improve disposition reporting for mentally ill offenders. * BJS will work with the States to improve reporting to the FBI so that Federal databases will reflect more complete information on individuals prohibited from purchasing firearms for mental health reasons. * BJS will expand two data collections to provide information on the ability of the States to examine records of those adjudicated as a mental defective and of those committed to a mental institution. * BJS will more effectively target efforts to provide NCHIP funds for the improvement of access to mental health records. 1. Model Data Sharing Agreement will be developed In response to a recent BJS survey, officials from 11 point-of-contact States reported that they do not consult mental health records as part of background checks for firearm purchases. In nearly two-thirds of these States, officials indicated that records are not consulted because State law or regulation prohibits access. One of BJS' statutory mandates is to "provide for the development and enhancement of State and local criminal justice information systems."***Footnote 13: 42 USC sec 3732 (18).*** As part of that mandate, BJS will consult with the appropriate legal, mental health, and criminal justice officials to develop a Model Data Sharing Agreement. Such an agreement would make State mental health records accessible for firearms background checks under the Federal Gun Control Act. One promising tactical option for the Model Data Sharing Agreement is to specify a reporting arrangement with appropriate State authorities. The State could submit only the names of those with prohibiting mental health backgrounds to the FBI's Denied Persons File rather than to the FBI's database of mental defectives. This is one of many options that simultaneously protects the privacy of records of persons with mental health histories and enables the States to ensure that firearms purchases comply with Federal law. 2. Best Practices series initiated Because State agencies vary in their ability to access mental health records for presale background checks, BJS will undertake a Best Practices Initiative to highlight promising activities in an ongoing series of reports. The Best Practices series will present profiles of States that can quickly consult mental health records in response to a NICS inquiry. Findings will highlight several critical issues, among them -- (1) the legal, policy, or regulatory basis for consulting mental health records (2) types of mental health records consulted (3) whether a centralized State database exists and which agency maintains it (4) which agencies (public, private, or both) are responsible for contributing mental health information (5) procedures that are followed when a background check is carried out (6)unexpected problems that arose after implementation of the system (7) suggested reforms to improve the system (8) contact information for a State official with responsibility for the system (9) how issues relating to patient confidentiality were addressed. BJS will continue to actively report on the progress made by all States in accessing mental health information for firearm presale background checks. 3. Mental Health Courts Initiative undertaken In 2000 Congress enacted the "America's Law Enforcement and Mental Health Project" to establish demonstration mental health courts. These courts exist for improving supervision of and delivery of services to mentally ill offenders.***Footnote 14: America's Law Enforcement and Mental Health Project, Pub. L. No. 106-515, 114 Stat. 2399 (2000).*** BJA is administering the Mental Health Courts Program in collaboration with the Substance Abuse and Mental Health Services Administration in the Department of Health and Human Services. This program supports "mental health court demonstration projects that mobilize communities to implement innovative, collaborative efforts to bring systemwide improvements in the way they address adult offenders with mental disabilities or illnesses."***Footnote 15: BJA, Mental Health Courts Project Program Competitive Grant Announcement, 2002 at 4. See also, BJA, Emerging Judicial Strategies for the Mentally Ill in the Criminal Caseload: Mental Health Courts in Fort Lauderdale, Seattle, San Bernardino, and Anchorage. (NCJ 182504) April 2000, reprinted February 2001.)*** In FY 2003 BJA provided over $2 million to establish mental health courts that provide continuous juridical oversight and intensive case management. BJS will work with BJA to ensure that recipients of Mental Health Court Program grants focus on disposition reporting. State and Federal authorities should have accurate and complete records on persons who fall within the purview of the Gun Control Act if they are processed through the mental health courts. 4. Improved State reporting to the FBI encouraged In 2002, 2 of the 14 point-of-contact States that check mental health records prior to the sale of a firearm forward denial information to the FBI for the Denied Persons database in the NICS Index. State officials from the 12 States not submitting denial information to the FBI provided reasons that fall into three categories: State law or regulations prohibit the transmission of this information to the FBI (6 States): State officials from California, Connecticut, Nevada, Oregon, Pennsylvania, and Wisconsin reported prohibition of transmitting mental health data. Oregon officials cited State confidentiality agreements with mental health agencies. Although Illinois law does not prohibit disclosure, officials reported that they do not provide these denials to the FBI because State law does not require them to do so. Requisite technology is not in place to reliably submit denial information based on mental health records to the FBI (3 States): Respondents from Washington, Wisconsin, and Hawaii noted that they currently do not have the technology to reliably submit denials based on mental health records to the FBI. States need administrative reforms and improved interagency coordination (3 States): Nebraska and Washington officials indicated that the current fragmented nature of conducting firearms purchases and issuing firearms permits prevents State officials from providing denial information from mental health records to the FBI. Utah officials stated that defendants who are involuntarily committed generally have their commitments overturned after responding to treatment, and these changes in status are not reliably reported to other State agencies or the FBI. To address these problems, BJS will request that State repositories and their data suppliers code court orders which commit offenders to mental institutions as dispositions of arrests on the offenders' RAP sheets. Those dispositions are to be maintained on the III- accessible record as well. BJS will also continue to encourage States to transmit information on firearm purchase rejections or commitments to mental institutions to the NICS Denied Persons File for inclusion in the national NICS checks. This will remain a high priority for the FY 2004 NCHIP. 5. BJS improves data collection concerning the mentally ill Two BJS surveys obtain information on States' ability to examine records of persons adjudicated as mental defectives or committed to mental institutions. In response to Attorney General Ashcroft's June 2001 directive to BJS, BJS followed up on the Survey of State Procedures Relating to Firearms Sales and the Survey of State Criminal History Information Systems. These improvements will permit BJS to better inform policymakers about State efforts to examine mental health records during background checks. The Survey of State Procedures Related to Firearm Sales is an annual national survey produced by the BJS Firearm Inquiry Statistics (FIST) project, a component of NCHIP. The FIST project was established in 1995 to collect data that describe the scope and impact of firearm background checks required under Federal or State law. Survey reports have been published annually since 1996. Information is collected from hundreds of Federal, State, and local agencies, including law enforcement organizations, statistical analysis centers, and legislative research bureaus. The current survey asks point-of-contact States to indicate whether they access a statewide mental health database when conducting a background check for a firearms transfer. The follow-up survey will examine the complicated issue of how a State handles information about persons adjudicated as mental defectives or committed to a mental institution. The follow-up survey will -- * Gather more detailed information about procedures used to check mental health records, the level of automation of these records, the nature of the impediments to checking mental health records, and initiatives planned or underway to consult these records. * Add questions about the nature of the database of those States checking mental health records from a centralized database. The Survey of State Criminal History Information Systems is a biennial survey of all directors of State criminal history repositories. The initial survey in this series was administered in 1989. The survey is the primary source of information on the size, level of automation, and shareability of criminal history record holdings in each State. It provides details of the procedures by which State repositories receive and process criminal history information including arrests, correctional data, and disposition data. Survey enhancements will include the following: * For those States that serve as points of contact for background checks, information will be obtained on the centralized State databases that are utilized to consult records of those adjudicated as a mental defective and those committed to a mental institution. Analysis of results will address the size, level of completeness, and accuracy of these databases. If databases are described as incomplete, follow-up questions will ascertain the reasons for the incompleteness and the remedial measures being undertaken, if any. * To assess the accuracy of these databases, the survey will examine how often the databases are audited to remove items such as overturned judicial findings of mental illness. For States that do not serve as points of contact for background checks, the follow-up will examine which State agencies maintain mental health information and the impediments to transmittal of this information to the NICS Index. 6. Technical assistance for improving access to mental health records provided to States A longstanding priority of NCHIP has been establishing the infrastructure to support full implementation of the NICS. BJS has strongly encouraged States to improve access to mental health records, stressing its importance through its annual NCHIP announcements and communications with the States.***Footnote 16: National Criminal History Improvement Program Fiscal Year 2002 Program Annoucement, NCJ 192811, February 2002.*** BJS will make NCHIP funds available for three related purposes: (1) To support of the Best Practices Initiative for the replication of projects featured in the Best Practices report and the bulletins describing promising activities. (2) To support the Initiative to Improve State Reporting to the FBI to remove the technological impediments to report denials based on mental health disqualifiers. (3) To directly assist the States in designing and implementing an array of new programs, policies, and technologies to upgrade capabilities to search mental health records. BJS will also encourage States that have few automated mental health records to use NCHIP funds for the automation of mental health records. States will be encouraged to establish processes that enable courts to automatically update a defendant's criminal record if the offender is adjudicated mentally incompetent or not guilty by reason of insanity. BJS surveyed States that search mental health records in background checks In 2002 BJS conducted an in-depth survey of all 14 States that include mental health records in database searches before transfers of firearms. This survey examined a number of issues relating to these States' procedures for checking mental health records, including the following: * how the authority to check mental health records was initially obtained * which agencies maintain the information * which entities provide it * whether the mental health records are maintained in a central State database * whether an agreement exists between the State repository and the mental health care providers * information concerning the types of mental health information checked * costs of establishing and maintaining the system * improvements that are needed to make the system more effective * other related issues. 14 point-of-contact States check mental health records Of the 25 States that serve as points of contact (POC's) for firearm purchases, the BJS survey found that 14 include State mental health records in the search prior to the transfer of a firearm.***Footnote 17: For this survey, the designation "POC State" covers States conducting presale checks for all firearms or for only certain firearms that are sold in the State.*** The POC States that include mental health records in their presale firearms checks are California, Illinois, Michigan, Pennsylvania, New Jersey, Virginia, Connecticut, Georgia, Hawaii, Nebraska, Oregon, Utah, Washington, and Wisconsin. The 11 POC States that do not check mental health records are New York, Arizona, Colorado, Florida, Indiana, Iowa, Maryland, Nevada, New Hampshire, North Carolina, and Tennessee. The remaining 25 States and the District of Columbia do not serve as POC's. None of these jurisdictions contributes mental health records or any related mental health information to the FBI for use during presale firearm checks. On average, the 14 States that examine mental health records prior to sales of a firearm have over 7 years of experience in conducting these checks. The States with the most experience are Hawaii (18 years), California (11 years), Illinois (10 years), Virginia (9 years), and Washington and Wisconsin (8 years each). The BJS survey asked the State repository directors to characterize their State's mental health records on the basis of -- (1) accuracy (2) timeliness (3) accessibility for a firearms check (4) completeness. Most State repositories rated their records as "very good" or "excellent" in each of the four criteria. The California, Georgia, Illinois, and New Jersey repository directors rated their mental health records as generally "excellent." Repository directors for Connecticut and Utah rated their mental health records as generally "very good." Three of the State repository directors rated their mental health records as "poor" on one of the criteria. Directors in Oregon and Nebraska characterized the timeliness of their mental health records as poor. Virginia reported the accessibility for a firearms check to be poor. The survey asked State officials to provide a comprehensive assessment of the State's program to allow mental health information to be checked prior to the sale of a firearm. Respondents from Michigan and Nebraska commented that the process is burdensome on personnel and costly in terms of human resources and technology. In contrast, survey respondents from California, Georgia, and New Jersey indicated that their systems work well and perform a valuable public service. In California from 1991 through February 2002, more than 3,100 individuals were denied the purchase of a firearm due to a pre-existing mental health diagnosis that the individual was a danger to himself or others. Repository directors recommended reforms for checks of mental health records The survey asked the 14 State repository directors what could be done to make their checks of mental health records more efficient. Recommendations included -- * greater automation of records so that checks may be conducted online * permit courts, public and private mental health facilities, and other mandatory reporters to submit data about prohibited persons in Internet-based real time reporting to the State database * permit interstate access to the mental health information by State officials * broaden the scope of mental health databases to include private mental health facilities * standardize the process for obtaining and maintaining records, removing inconsistent requirements among States * add positive identifications of mental health patients, such as a fingerprint or other biometric identifier. States check different types of mental health information Each of the 14 POC States that consult mental health records conducts a check of adjudicated commitments, in which individuals were committed after a hearing by a judge or a specially appointed officer or board. Apart from the category of adjudicated commitments, the mental health information that is checked during investigations of firearm purchasers varies among the States. California included in its check of mental health records information on individuals who have communicated serious threats of violence to a licensed therapist. States vary approaches for checking mental health records Ten of the 14 States maintained a central database of mental health records. Agencies supporting or managing these databases are the State police (Michigan, Pennsylvania, and Virginia), State mental health department or department of health (Connecticut, Hawaii, New Jersey, and Washington), criminal justice agencies (California and Georgia), and the Department of Human Services (Illinois). Some State databases collect information on mental health records only from publicly-funded facilities while other States include private facilities as contributors. In Hawaii only the publicly-funded mental health facilities contribute information to the State database. Other States collect from public and private mental health facilities (New Jersey and Washington) as well as public mental health care psychiatrists, psychologists, and counselors (California, Oregon, and Illinois). Connecticut, Georgia, Michigan, and Virginia receive mental health information only through the courts. Under Pennsylvania law mandatory reporters include judges of the Courts of Common Pleas, mental health administrators, and mental health review officers. Several States noted that 100% of the individuals within the State database would be prohibited from purchasing a firearm in the State. However, most States were unable to estimate what percentage of the State's mental health patients are included in the central database. State databases of mental health information are updated frequently. Only one State (Oregon) updates its database in real time. Eight States update their mental health databases daily (California, Connecticut, Georgia, Hawaii, Illinois, Michigan, Pennsylvania, and Virginia). New Jersey updates its mental health database on a weekly basis. Of the four States without a centralized database, Utah and Wisconsin generally maintain their mental health information in the courts, while Nebraska uses the courts as well as State commissions and individual mental health facilities. Oregon relies on several databases maintained by various State agencies including the State mental health department. Informal relationships link State criminal record repositories and mental health agencies A formal agreement, such as a memorandum of understanding signed by a State's criminal history record repository and mental health providers, can govern submission of mental health information for determining eligibility to purchase a firearm. In only Georgia, New Jersey, and Oregon does such an agreement exist, the result of negotiations between the affected agencies. The remaining 11 States reported no formal agreement. Seven States indicated that State administrative regulations or internal policies governed the transfer of mental health information. Four of the 14 States (Illinois, Michigan, Nebraska, and Wisconsin) responded that no such administrative regulations existed. Most States can restore right to possess firearms to individuals disqualified for mental health reasons In 8 of the 14 States, a prohibition against a firearm purchase based on a mental health disqualification may be lifted once a court or other adjudicatory body restores a "non-mental defective" status to the individual. These States are California, Connecticut, Michigan, New Jersey, Pennsylvania, Utah, Virginia, and Wisconsin. In California certain individuals are prohibited from purchasing firearms for life pursuant to Federal law. For other individuals the prohibition may be lifted under certain circumstances, such as release from a mental health treatment facility or successful completion of the court's restoration process. The majority of prohibitions in California's central database last 5 years. In Georgia and Illinois the prohibition against purchasing firearms lasts for 5 years after discharge from a treatment facility. In Virginia the prohibition may be lifted upon an individual's release from a mental health facility. Nebraska and Oregon are the only States to report that the prohibition against purchasing firearms is permanent. Legislation served as the primary source of authority for examining mental health information In almost all of the States (12 of the 14), the authority to use mental health records as disqualifying information came from the State legislature. Only Nebraska and New Jersey cited other sources of authority. In half the States the criminal record repository was involved in drafting the legislation. The legislation generally came about from a gradual recognition of the need to examine mental health records. Only Connecticut reported that State legislation was enacted as a result of a single high-profile incident. Most mental health information mailed to the repository Five of the 14 States transmit mental health information to the criminal history repository or firearms processing agency online or by some other automated method (Connecticut, Michigan, Oregon, Utah, and Wisconsin). Ten of the States mail hard copies or tapes of the mental health information to the repository at regular time intervals (California, Georgia, Illinois, Michigan, Nebraska, New Jersey, Pennsylvania, Virginia, Washington, and Wisconsin). Although a few States utilize more than one mode of transmission, most do not. All State repositories have adopted security measures to protect mental health information All State repositories reported security measures to ensure that only personnel determining firearm eligibility have access to the mental health information in the repository. Two States provide for broader access. In California law enforcement personnel are granted access to the information under statutorily specified conditions limited to criminal investigations that involve the carrying or use of a firearm. In Pennsylvania the subject of the record and law enforcement officials who enforce State firearm laws are provided access to the mental health records in the repository. In addition to personnel restrictions, each State repository reported system security measures such as firewalls, password-protected information, and employee identification checks to prevent unauthorized parties from accessing the mental health information. State repositories often use quality control measures to ensure the accuracy, completeness, and timeliness of the mental health information received. Most of the States (11 of the 14) conduct a manual review of the incoming data or a manual review of the data prior to making an eligibility determination. Five States (California, Hawaii, Illinois, Michigan, and Oregon) rely on computers to edit and verify the information. The Hawaii and Illinois repositories conduct regular audits of the system to ensure quality control. In the event of a breach of quality control, such as a complaint concerning the unauthorized release of information or mistaken identity, most of the States (10 of 14) provide a mechanism by which the complainant can formally appeal the error. These complaints are handled through an appeals process in California, Connecticut, Georgia, Hawaii, Illinois, New Jersey, Oregon, Utah, Virginia, and Wisconsin. There is no appeals process for quality control breaches in Michigan, Nebraska, and Pennsylvania. States overlap concerning the identifiers used with mental health records. Nearly every State used the subject's name, date of birth, Social Security number/driver's license number, race, and gender. Other identifiers include the adjudication date (Virginia), admission date (New Jersey), date of commitment (Pennsylvania), case number (Michigan), patient identification number (Illinois), address (Illinois), facility name/number (Illinois and New Jersey), physical description (California, Virginia, and Wisconsin), and charge code(Michigan). 11 States provide appellate review of denials of firearm purchase due to mental health history Eleven of the 14 States have established an appeals process for prospective firearms purchasers who are initially denied access to a firearm because of a disqualifying mental health history. These States are California, Georgia, Hawaii, Illinois, New Jersey, Oregon, Pennsylvania, Utah, Virginia, Washington, and Wisconsin. The three that do not provide appellate review of these denials are Connecticut, Michigan, and Nebraska. Costs for accessing State mental health information varies across States The costs of developing and maintaining access to State mental health information depend on a number of factors including the following: * the State's regulations concerning access to and use of mental health records * the size of the State's mental health population * the presence or absence of procedures to lift the firearms prohibition based on a mental health disqualification * the presence or absence of interagency agreements * the level of automation of the State's mental health records. Cost estimates to develop systems to access mental health information range from "minimal cost" (Wisconsin) to $800,000 (California). Six States provided cost estimates between these two. Systems development cost $150,000 in Connecticut, $80,000 in Oregon, $54,000 in New Jersey, $25,000 in Nebraska, $10,000 in Virginia, and $7,000 in Utah. Startup costs were higher than anticipated in Connecticut, Hawaii, Nebraska, and Pennsylvania. They were lower than anticipated in New Jersey, Oregon, Utah, and Virginia. Similar variation occurred among State estimates of annual costs to maintain these systems. New Jersey, Nebraska, Virginia, and Wisconsin reported that annual maintenance costs were negligible while California estimated the annual cost to be $408,000. Annual maintenance costs are estimated to be $30,000 in Hawaii, $10,000 in Connecticut, $3,000 in Oregon, and $1,500 in Utah. The nine States of Connecticut, Georgia, Hawaii, Illinois, Michigan, Pennsylvania, Utah, Virginia, and Wisconsin reported that State funds support the systems by which mental health information is examined during a firearms background check. These States rely on both long-term and annual appropriations by the State legislature. New Jersey and Nebraska reported relying on Federal funds, while California, Pennsylvania, Oregon, and Wisconsin finance all or part of their systems through fees. State systems that access mental health information report success Most States report no significant operational changes in the systems that provide the State repositories with mental health information. Illinois improved their system by incorporating more professional mental health consultants. Utah officials added an audit feature to permit the courts and the Department of Public Safety to compare and verify information on a weekly basis. California officials reported that corrective legislation was required to eliminate duplicative reporting. Appendix II Identifying prohibited purchasers: Domestic violence misdemeanants and subjects of restraining orders The Gun Control Act as amended prohibits the possession of a firearm by individuals who have been convicted of a misdemeanor crime of domestic violence. According to ATF regulations, misdemeanor crime of domestic violence means -- "a Federal, State or local offense that: * Is a misdemeanor under Federal or State law or, in States which do not classify offenses as misdemeanors, is an offense punishable by imprisonment for a term of one year or less, and includes offenses that are punishable only by a fine. (This is true whether or not the State statute specifically defines the offense as a 'misdemeanor' or as a 'misdemeanor crime of domestic violence.' The term includes all such misdemeanor convictions in Indian Courts established pursuant to 25 CFR part 11.) * Has, as an element, the use or attempted use of physical force (e.g., assault and battery), or the threatened use of a deadly weapon * Was committed by a current or former spouse, parent, or guardian of the victim, by a person with whom the victim shares a child in common, by a person who is cohabiting with or has cohabited with the victim as a spouse, parent, or guardian,(e.g., the equivalent of a 'common law' marriage even if such relationship is not recognized under the law), or a person similarly situated to a spouse, parent, or guardian of the victim (e.g., two persons who are residing at the same location in an intimate relationship with the intent to make that place their home would be similarly situated to a spouse). "A person shall not be considered to have been convicted of such an offense for purposes of this part unless: * The person is considered to have been convicted by the jurisdiction in which the proceedings were held. * The person was represented by counsel in the case, or knowingly and intelligently waived the right to counsel in the case; and * In the case of a prosecution for which a person was entitled to a jury trial in the jurisdiction in which the case was tried, either (i) The case was tried by a jury, or (ii) The person knowingly and intelligently waived the right to have the case tried by a jury, by guilty plea or otherwise. "A person shall not be considered to have been convicted of such an offense for purposes of this part if the conviction has been expunged or set aside, or is an offense for which the person has been pardoned or has had civil rights restored (if the law of the jurisdiction in which the proceedings were held provides for the loss of civil rights upon conviction for such an offense) unless the pardon, expunction, or restoration of civil rights expressly provides that the person may not ship, transport, possess, or receive firearms, and the person is not otherwise prohibited by the law of the jurisdiction in which the proceedings were held from receiving or possessing any firearms."***Footnote 18: See 27 CFR 178.11. *** The Act as amended also prohibits possession of firearms by individuals who are the subject of a restraining order. Federal law defines this prohibition as a person who -- "is subject to a court order that restrains such person from harassing, stalking, or threatening an intimate partner of such person or child of such intimate partner or person, or engaging in other conduct that would place an intimate partner in reasonable fear of bodily injury to the partner or child, except that this paragraph shall only apply to a court order that -- (a) was issued after a hearing of which such person received actual notice, and at which such person had the opportunity to participate; and (b)(i) includes a finding that such person represents a credible threat to the physical safety of such intimate partner or child; or (ii) by its terms explicitly prohibits the use, attempted use, or threatened use of physical force against such intimate partner or child that would reasonably be expected to cause bodily injury."***Footnote 19: 18 USC sec 922.*** Several recent BJS and GAO reports make clear that domestic violence misdemeanor conviction records and restraining orders continue to pose significant challenges to the NICS system: -- Use and Management of Criminal History Record Information: A Comprehensive Report, 2001 Update, (NCJ 187670) December 2001.-- Continuing Criminal History Records Improvement Evaluation: Final 1994-98 Report, (NCJ-179768) February 2000. -- Survey of State Procedures Related to Firearms Sales, Midyear 2002, (NCJ-198830) April 2003. -- Opportunities to Close Loopholes in the National Instant Background Check System, General Accounting Office, (GAO-02-720) July 12, 2002. -- Improving the National Instant Criminal Background Check System, General Accounting Office, (GAO-T-GGD-00-163) June 21, 2000. -- Gun Control: Options for Improving the National Instant Criminal Background Check System, General Accounting Office, (GAO/GGD-00-56) April 12, 2000. The first section of this appendix examines the principal impediments to accessing complete domestic violence misdemeanor records and restraining orders, including -- * Nonfelon information being generally unavailable to State repositories * Unautomated conviction records for domestic violence misdemeanors and unautomated restraining orders * Incomplete records for domestic violence misdemeanor convictions and incomplete restraining orders * Insufficient resources to examine domestic violence records and restraining orders * Problematic case management for domestic violence * Complicated Federal definition of "misdemeanor crime of domestic violence." The final section of Appendix II presents five new BJS initiatives to improve access to these records. States improve access to domestic violence records Since 1996 BJS has encouraged States to use NCHIP funds to flag records to designate misdemeanor convictions for domestic violence, stalking, protection orders, and crimes against children, the elderly, and the disabled. Between 1995 and 2002, BJS awarded over $18.5 million, or 5% of all available NCHIP funds, to 45 States and the District of Columbia for the automating, upgrading, and flagging of domestic violence records and protection order files. States have also used NCHIP funds to develop registries for protection orders and restraining orders. Improving records for domestic violence offenders will remain one of BJS' most important priorities for two reasons. First, domestic violence misdemeanants represent a disproportionately large percentage of firearm retrieval actions referred to the Bureau of Alcohol, Tobacco, Firearms, and Explosives (ATF). Across more than 34 months, 1998-2001, ATF data indicate that domestic violence misdemeanants accounted for 14% of all FBI NICS denials but nearly 26% of firearm retrieval referrals.***Footnote 20: Opportunities to Close Loopholes, pp. 18-20.***The percentage of retrieval actions referred to ATF for subjects of a protection order was not disproportionate. Individuals subject to a protection order comprised nearly 4% of NICS denials and 1% of firearm retrieval referrals. Second, domestic violence caseloads have increased in recent years, strongly suggesting an increase in the number of domestic violence convictions. According to the National Center for State Courts, domestic violence filings increased 12% between 1996 and 2000.***Footnote 21: Examining the Work of State Courts, 2001: A National Perspective from the Court Statistics Project, National Center for State Courts, 2001.*** Ten States had increases of 10% or more in the number of domestic violence filings between 1998 and 2000: South Dakota (34%), Ohio (27%), Illinois (21%), Indiana (21%), Tennessee (19%), Montana (18%), Michigan (15%), North Dakota (15%), West Virginia (12%), and Virginia (11%).***Footnote 22:Examining the Work of State Courts, 2001, p. 42. *** Without better procedures to identify domestic violence offenders and persons with protection orders against them, the threat to public safety will become more severe as the number of domestic violence offenders increases. States hampered in checks for domestic violence misdemeanants and subjects of protection orders Repositories miss domestic violence misdemeanor records As of midyear 2002, 36 State criminal history repositories maintained data on domestic violence misdemeanor convictions, and 43 repositories contained information on restraining orders.*** Footnote 23: survey of State Procedures Related to Firearm Sales, Midyear 2002 (NCJ 198830) April 2003.*** How complete these databases are or how frequently they are updated or audited is unknown. The lack of comprehensive misdemeanor arrest and disposition data has been identified as a major deficiency in State criminal history record systems.***Footnote 24: Use and Management of Criminal History Record Information: A Comprehensive Report, 2001 Update(NCJ 187670) December 2001.*** In many States only suspects charged with felonies are routinely fingerprinted, and only their arrests are reported to the repository for inclusion in a suspect's RAP sheet. Automation remains a problem for some domestic violence misdemeanor records and protection orders In 2002 four States indicated that the lack of automated misdemeanor records prevented officials from fully checking domestic violence convictions. Seven States reported in 2002 that State officials could not retrieve accurate protection order information because the records were not fully automated. The lack of automation prevents many States from forwarding these records to the FBI. Eight States reported that lack of automated records prevented them from sending misdemeanor domestic violence information to the FBI. In 11 States, officials indicated that they were prevented from forwarding protection order information to the FBI because the information was not automated. Records incomplete for domestic violence misdemeanors and protection orders In response to a 2002 survey, 20 States reported that incomplete records prevented the State from retrieving accurate domestic violence misdemeanor information for background checks. These records were incomplete for several reasons, including the following: * absence of a flag denoting domestic violence misdemeanor * the age of the record * the lack of a final disposition * the lack of information about the victim-offender relationship * the failure of arresting agencies and courts to report the information promptly to the repository * agency procedures that do not require fingerprinting misdemeanants. Twelve States also reported in 2002 that incomplete restraining order information made it difficult to conduct background checks. Most of these States indicated that restraining order records were incomplete due to the courts' inability to provide the repository with the necessary information. Other factors included the failure to have proper identifiers (such as date of birth) in the record, lack of automation, and insufficient resources to process the information accurately and promptly. Incomplete records not only complicate background checks, they also prevent States from submitting material to FBI databases, such as the National Protection Order File. Problems from incomplete records prevented 15 States from providing domestic violence misdemeanor information to the FBI and prevented 13 States from providing protection orders. Nineteen States reported that the absence of a flag in the record prevented the State from sending domestic violence misdemeanor records to the FBI. State checking agencies lack time and resources to access domestic violence records or restraining orders In 2002 eight States reported that they did not have sufficient time or resources to access complete domestic violence misdemeanor information for background checks. Four states reported that additional time or resources were needed before they could retrieve complete restraining order information for background checks. Officials from five States reported that insufficient resources were a factor that prevented them from providing the FBI with complete domestic violence misdemeanor information. In 10 States officials reported that they did not have sufficient resources to provide complete protection order information to the FBI. Tracking of domestic violence cases often difficult Handling of domestic violence-related cases varies within and between States. Cases involving domestic violence may arise from a variety of divisions within a State judicial system, such as civil, criminal, juvenile, and family divisions, as well as from specialized domestic violence courts. Case management systems vary in their ability to track case filings, the issuance of protection orders, and documenting case outcomes. The lack of data on domestic violence is also related to the legal definitions of offenses. Recorded offenses may not include information about the victim. That information is essential to determine whether an assault, for example, occurred between intimate partners, acquaintances, or strangers. To better understand how the courts handle domestic violence cases, in 2002 BJS initiated a study of case-processing flows through the criminal justice system for persons charged with domestic violence offenses. Representative data from the 75 largest counties will be collected. An estimated 16,000 domestic violence cases will be tracked from arrest, through prosecution, to conviction and sentencing. Information will be obtained on the types of sentences imposed for domestic violence, including supervision in the community under sentences to probation. Factors affecting the likelihood of conviction will be documented. Apart from legal provisions, these factors may include the following: the victim-offender relationship, criminal histories of victim and offender, policies guiding issuance and enforcement of restraining orders, prosecutorial policies regarding domestic violence cases, and pretrial release and sentencing practices. The study's objective is to determine whether domestic violence cases are handled differently from other types of assault filings. Reporting to criminal record repositories may be one of the differences. Definition of "misdemeanor crime of domestic violence" complicated The GAO identified another impediment to domestic violence misdemeanor reporting.***Footnote 25: Opportunities to Close Loopholes in the National Instant Background Check System, pp. 21, 67.*** The GAO noted that the Federal definition of "misdemeanor crime of domestic violence" presents a unique challenge for States for at least two reasons: (1) the law is retroactive in scope (2) the definition requires information about court proceedings that may be available only through manual research. For a Federal firearms eligibility determination to be made, several elements must be known about the misdemeanor domestic violence conviction, including the following: (1) whether an offender was represented by counsel or knowingly and intelligently waived the right to counsel (2) whether there was a jury trial if the offender was entitled to one, or whether this right was intelligently waived (3) whether a conviction that meets these criteria was later expunged or set aside or was for an offense for which the offender was pardoned or had his/her civil rights restored. According to a 2002 BJS survey, seven States reported that the complicated nature of the Federal definition of a "misdemeanor crime of domestic violence" prevented the State from retrieving complete and accurate domestic violence misdemeanor information. Four States reported that the cumbersome Federal definition of a "misdemeanor crime of domestic violence" prevented the State from sending complete and accurate domestic violence misdemeanor information to the FBI. BJS has developed 5 initiatives to improve access to misdemeanor domestic violence records and to protection orders: 1. NCHIP will focus on improving the flagging of domestic violence misdemeanor records Since 1996, when a conviction for a domestic violence misdemeanor was added to the list of Federal firearm prohibitors, NCHIP has supported State efforts to ensure that records of domestic violence related offenses are included or "flagged" in criminal history records. Allowable costs under NCHIP have included the costs of flagging or algorithms used for flagging as well as those associated with the identification of records to be flagged. Since 1996 NCHIP funds have enabled certain States to build centralized statewide domestic violence databases and to upgrade their records; however, not all States have done so. This new initiative will encourage State officials to place a higher priority on meeting the challenges posed by domestic violence misdemeanor convictions. NCHIP funds will be made available to the States to allow them to implement procedures that ensure that misdemeanor convictions involving domestic violence are systemically incorporated into RAP sheets and made available for use during a NICS check. The BJS evaluation of the NCHIP program will gather systematic data on techniques used across States to overcome impediments to identifying prohibited firearm purchasers. The evaluation will document changes in the ability of States to provide such information to the NICS Index. 2. NCHIP will assist States in obtaining NIBRS certification BJS has provided funds under CITA to the States to promote NIBRS, an emerging FBI data system designed to capture detailed offense, offender, victim, property, and arrest data on each crime incident. BJS strongly encourages States to link NIBRS data on offenders to RAP sheet arrest and disposition transactions. The victim and incident characteristics, such as victim-offender relationship, can thereby be used in the State's background check. The NIBRS system can improve State criminal records' identification of domestic violence offenders. For example, a criminal record that shows an assault conviction might be linked with NIBRS information to identify the relationship between the victim and offender, including relationships within the family (such as spouse, common-law spouse, and sibling) and outside the family (such as acquaintance, ex-spouse, and boy/girlfriend). This additional information may replace the tedious manual research currently required to determine whether a conviction meets the Federal definition for a domestic violence disqualifier. In 2001 BJS distributed $13 million to 26 States through the NIBRS Implementation Program to develop and operate NIBRS-compatible systems. As of August 2003, 24 States had their State UCR programs approved by the NIBRS Certification Board. BJS remains committed to assisting States in attaining NIBRS certification. BJS will place a higher priority on its collaboration with the FBI to reach the goal of having all States become NIBRS certified as soon as practicable. This collaboration will focus on the 14 States testing their systems to determine whether they may soon meet certification standards, on the 9 States developing their NIBRS systems, and on the 6 States with no formal plans to provide NIBRS data. 3. BJS will improve its data collections domestic violence misdemeanors and protection orders Two BJS surveys examine States' ability to use data on domestic violence misdemeanor convictions and protection orders: the Survey of State Procedures Relating to Firearms Sales and the Survey of State Criminal History Information Systems. This initiative will require BJS to expand upon these surveys, learning more about the problems that the NICS faces in checking these records. In response to the Attorney General's 2001 directive to BJS, BJS has followed up on the Survey of State Procedures Relating to Firearms Sales and the Survey of State Criminal History Information Systems. The Survey of State Procedures Related to Firearm Sales an annual national survey, describes the scope and impact of firearm background checks required under Federal or State law. The current survey asks States to indicate whether they access a statewide database for domestic violence misdemeanor convictions when conducting a background check for a firearms transfer. A separate question asks whether a statewide database for restraining orders is accessed during background checks. BJS has recently supplemented these questions to gain in-depth information on-- * State procedures for checking domestic violence misdemeanor records and protection orders * impediments to checking these records * whether this information is forwarded to the FBI * impediments to transmittal of this information to the FBI * whether improvements are underway or planned to improve the State's ability to access or use domestic violence misdemeanor records and protection orders. The Survey of State Criminal History Information Systems is the primary source of information on the size and level of automation of criminal history record holdings by State. For the States that serve as points of contact for background checks, BJS will follow up on the survey to learn more about the centralized State databases used for domestic violence misdemeanor records and protection orders. BJS will also determine the size, level of completeness, and accuracy of these databases. If databases containing domestic violence information are described as incomplete, follow-up questions will seek reasons for the incompleteness and remedial measures being undertaken. To examine the accuracy of these databases, BJS will determine whether and how frequently the databases are audited to remove items such as expired protection orders. For those States that do not serve as points of contact for background checks, the followup survey will examine which State agencies maintain information concerning domestic violence misdemeanor convictions and the issuance of protection orders. Survey questions will also examine the impediments that prevent the transmittal of this information to the FBI's NICS Index. 4. NCHIP will work to improve misdemeanor domestic violence conviction reporting from the courts The Federal prohibition against possessing firearms attaches upon conviction in any court of a misdemeanor crime of domestic violence. However, current court reporting practices do not ensure that misdemeanor domestic violence convictions are consistently reported to a State's criminal history repository. BJS will encourage States to expend NCHIP funds to improve accessibility of domestic violence misdemeanor records and to improve the reporting of those records by courts to the repository and the NICS Index. Several States have utilized NCHIP funds for these purposes in recent years. BJS will also work with all States to automate case management systems. In addition to the provision of NCHIP funds, BJS will continue its funding of the Court Statistics Project through the National Center for State Courts. The National Center for State Courts recently developed a Family Violence Data Reporting Prototype to streamline the process of classifying, counting, and reporting family violence cases for all types of courts. This prototype assists the court in accurately reporting whether domestic violence related elements were present in a case, along with trial outcome, case disposition, and sentencing information. Recent tests of the Family Violence Data Reporting Prototype revealed impediments such as courts that lacked integrated data systems and jurisdictional problems with the test cases. Despite these flaws, most demonstration courts found the prototype to be a viable tool. Case reporting tools such as the Family Violence Data Reporting Prototype will lead to the greater availability and comparability of information on domestic violence misdemeanors. 5. BJS will encourage States to increase submissions to the National Protection Order Database The contribution of State protection order records to the FBI's National Protection Order file is designed to permit interstate enforcement of protection orders and allow for more thorough NICS background checks. A 2002 report from the Department of Justice's Office for Victims of Crime estimated that in 1998 the FBI's national database contained less than 5% of the 2 million protection orders believed to qualify for entry into the file.***Footnote 26: Office for Victims of Crime, Enforcement of Protection Orders, OVC Legal Series Bulletin No.4, January 2002, pg. 4.*** As of February 2003, the National Protection Order file held no records from Alabama, the District of Columbia, Hawaii, Mississippi, Nevada, New Jersey, Utah, Virginia, or West Virginia. In recent years the database has grown considerably, averaging nearly 200,000 additional records per year. BJS will undertake a concerted effort to target non- participating States for transmittal of protection orders to the FBI for inclusion in the national system. Appendix III BJS surveys in the aftermath of September 11, 2001 Through its recent surveys BJS has learned important details about some States' inability to routinely identify prohibited purchasers of firearms. Combined with the tragic events of September 11, 2001, these findings generated two research questions about the current capacity to protect U.S. residents against foreign threats. How adequate are State procedures in the event that a Federal immigration law violator is detained in the custody of local officials? The FBI arrested suspected conspirator Zacarias Moussaoui on immigration charges on August 17, 2001, and detained him in a Minnesota county jail. Based on this set of events, do States have policies to obtain immigration offenders' fingerprints and promptly provide them to the FBI, INS, or State criminal history repository when Federal officials apprehend the offenders and hold them in a local jail?***Footnote 27: The BJS survey described in this section was administered before the INS was incorporated into the Department of Homeland Security as the Bureau for Citizenship and Immigration Services.*** BJS commissioned a survey in November 2001 that addressed this issue. State procedures vary widely concerning these Federal detainees but generally do not ensure that their fingerprint information reaches the appropriate Federal authorities. The survey found that eight States do not have policies to take fingerprints from a Federal detainee charged with a violation of Federal immigration law and placed in the custody of local authorities. Eight States and the District of Columbia have policies to fingerprint the individual and forward the information to the FBI. Twenty-two States have policies to fingerprint the suspect and send fingerprint information to the INS. Are the Nation's criminal history record systems physically safe from external threats? BJS surveyed the States to learn whether their criminal history records are adequately protected by backup databases remote from the repository. All States reported having backup systems, and 47 States indicated that the backup systems were located offsite from the central repository. States vary in how they process Federal immigration law violators Broad disparities exist among State procedures in processing individuals apprehended by Federal officers and detained locally for violating Federal immigration laws. In the November 2001 survey of State criminal history repositories, only Massachusetts reported that it fingerprints detainees and forwards the prints to the State repository, the FBI, and INS (table). Eight States reported that they do not have a policy to fingerprint the detainees. In one of the eight, Maryland officials indicated that if an individual is detained by State or local law enforcement authorities for an immigration offense, fingerprints are taken, forwarded to the State repository for a criminal records check, and then kept on file. Fewer than half the States forwarding fingerprint information to the INS Twenty-two States operate under a policy to fingerprint locally held detainees suspected of violating Federal immigration laws and forward the information to the INS. Of the 22 States, 9 forward the information to INS but do not conduct an in-State criminal history check or create a permanent record. Fewer than a fifth of the States forwarding fingerprints to the FBI California, Connecticut, Louisiana, Massachusetts, and the District of Columbia reported that they collect fingerprints from detainees held locally for violating Federal immigration laws and routinely send them to the State criminal history repository for trans-mission to the FBI. Mississippi, Nevada, Texas, and Virginia reported sending the fingerprints directly to the FBI. One State in three submitting fingerprints of Federal detainees to a State criminal records check Eighteen States and the District of Columbia collect fingerprints from Federal immigration detainees held by local authorities and then perform a criminal records check. Missouri retains the fingerprint and arrest information in a separate classification created specifically for Federal detainees. Seven States and the District of Columbia have statutory authority to create a criminal history record and to issue a State identification number for individuals arrested in the State for violating Federal law. Two States and the District of Columbia retain the fingerprint and Federal arrest information in a miscellaneous file created to store records not meeting requirements for specific classifications. 1 in 5 States retaining fingerprints of Federal detainees in State AFIS. Twelve States operate under a policy to retain the suspect's fingerprints in the State AFIS for subsequent identification. In these States no new criminal history record is created. The States are Alaska, Arizona, Arkansas, Connecticut, Delaware, Idaho, Indiana, Missouri, Nebraska, New Hampshire, North Carolina, and Vermont. 2 States obtain dispositions of cases of Federal detainees Two States reported that they obtain dispositions for cases involving Federal detainees in their State by following up with other agencies. California reported that the State repository obtains disposition information from the INS and Federal courts. Alabama's repository follows up with the INS to obtain complete disposition information. Although all States were surveyed to determine how they obtained dispositions for cases involving Federal detainees, most States either indicated that dispositions are not obtained (14 States and the District of Columbia) or did not respond (34 States). State criminal history record systems respond to security concerns The September 11th terrorist attacks on the United States have forced Federal and State agencies to re-examine the security of physical and information infrastructures. As part of this effort, BJS surveyed the States to determine whether the Nation's criminal history record systems are physically safe from external threats. Most States with offsite backup systems updated frequently Every State maintains a backup of their criminal history record system. Forty-seven States indicated that their criminal record backups systems were located offsite from the central repository (New Hampshire, New Jersey, and New Mexico did not). Data in the State criminal record repository backup systems are updated frequently, although improvements could be made in at least 15 States. Six State repositories update their backup systems in real time (Alabama, Colorado, Florida, Indiana, New York, and Oklahoma). Nearly half of the State repositories (24) reported that they update their backup systems daily, and 15 repositories, weekly. The content of the backup systems varied among the States Nearly all of the States (49) reported that their backup systems contained automated criminal history records; however, this does not mean that the backup systems contained only automated records. In seven States, the criminal records repository director reported that their backup systems contained only automated criminal history records. Fifteen States reported that their backup systems contained; (1) automated criminal history records, (2) hard copies of criminal history records, and (3) electronic fingerprint images. Twenty five States reported that their backup systems contained automated criminal history records and electronic fingerprint images. One State reported that its backup systems contained automated criminal history records and hard copies of criminal history records. One State reported that its backup system contained only electronic fingerprint images. Among the States that include hard copies of criminal history records in their backup systems, four States update those records by creating an automated version of the record. Six States had policies to manually add updated information to the record. End of file 07/12/05 ih