This is the accessible text file for GAO report number GAO-06-326 
entitled 'Long-Term Care Facilities: Information on Residents Who Are 
Registered Sex Offenders or Are Paroled for Other Crimes' which was 
released on May 8, 2006.

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as part 
of a longer term project to improve GAO products' accessibility. Every 
attempt has been made to maintain the structural and data integrity of 
the original printed product. Accessibility features, such as text 
descriptions of tables, consecutively numbered footnotes placed at the 
end of the file, and the text of agency comment letters, are provided 
but may not exactly duplicate the presentation or format of the printed 
version. The portable document format (PDF) file is an exact electronic 
replica of the printed version. We welcome your feedback. Please E-mail 
your comments regarding the contents or accessibility features of this 
document to Webmaster@gao.gov.

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

United States Government Accountability Office:

GAO:

Report to Congressional Requesters:

March 2006:

Long-Term Care Facilities:

Information on Residents Who Are Registered Sex Offenders or Are 
Paroled for Other Crimes:

GAO-06-326:

GAO Highlights: 

Highlights of GAO-06-326, a report to congressional requesters.

Why GAO Did This Study: 

Approximately 23,000 nursing homes and intermediate care facilities for 
people with mental retardation (ICF-MR) receive federal Medicare and 
Medicaid funding. Media reports have cited examples of convicted sex 
offenders residing in long-term care facilities and, in some cases, 
allegedly abusing other residents. Given concerns about resident 
safety, GAO was asked to assess (1) the prevalence of sex offenders and 
others on parole for non-sex offenses living in long-term care 
facilities and the extent of any abuse they may have caused, 
(2) the legal requirements for notifying facilities and others when 
offenders are residents, and (3) the extent to which facilities have 
different supervision and separation requirements for offenders. GAO 
analyzed a national database for sex offenders and analyzed state 
databases in a sample of eight states for sex offenders and parolees.  

What GAO Found: 

By analyzing the FBI’s NSOR, which is a compilation of sex offender 
registries submitted by all states, GAO identified about 700 registered 
sex offenders living in nursing homes or ICFs-MR during 2005. Most 
identified sex offenders were male, under age 65, and living in nursing 
homes, and represented 0.05 percent of the 1.5 million residents of 
nursing homes and ICFs-MR. About 3 percent of nursing homes and 0.7 
percent of ICFs-MR housed at least 1 identified sex offender during 
2005. However, these estimates are understated due to data limitations. 
For example, because of a lack of resources or an inability to comply 
with certain FBI reporting requirements, states have had varying 
degrees of difficulty submitting their full state registries to the 
NSOR. While the FBI does not track NSOR submission rates, GAO compared 
sex offender registry data from seven of the eight states reviewed to 
NSOR data and found that the NSOR data included about 57 percent of sex 
offenders registered in these states, with submission rates ranging 
from 1 percent to 83 percent. Because a national data source on 
parolees that included address information was not available, GAO also 
obtained parolee databases from the eight reviewed states and 
identified 204 offenders on parole for non-sex offenses living in long-
term care facilities. GAO could not determine the overall risk that 
registered sex offenders and parolees pose to other residents in long-
term care facilities because offender status is not tracked with abuse 
reporting. Facility administrators expressed greater concern over the 
risk posed by cognitively impaired or mentally ill residents.

Federal law requires state law enforcement agencies to release relevant 
information about registered sex offenders when necessary to protect 
the public, but GAO did not identify a similar federal law for the 
parolee population. States have broad discretion in how to implement 
the requirement for registered sex offender notification. Therefore, 
the extent to which states’ community notification laws apply to all 
registered sex offenders or explicitly include long-term care 
facilities varies. For example, four of the eight states GAO 
reviewed—California, Illinois, Minnesota, and Oklahoma—had laws that 
specified long-term care facilities as entities to be notified for at 
least some registered sex offenders who entered them. However, some 
facility administrators GAO contacted were uncertain as to whether they 
could share information with staff and others about residents who were 
known offenders in light of the Privacy Rule issued under the Health 
Insurance Portability and Accountability Act of 1996 (HIPAA). 

Long-term care facilities GAO contacted do not routinely impose 
different supervision or separation requirements on residents who are 
offenders based solely on their prior convictions. Instead, these 
facilities base such decisions on the demonstrated behaviors of 
residents. Even if long-term care facilities wanted to impose different 
supervision and separation requirements on offenders, their ability to 
do so is limited because they are not always aware of residents’ prior 
convictions.

What GAO Recommends: 

GAO recommends that the Federal Bureau of Investigation (FBI) assess 
the completeness of the National Sex Offender Registry (NSOR), 
including state submission rates, and evaluate options to increase its 
comprehensiveness.  The Department of Justice (DOJ) commented that 
these recommendations are unnecessary because of efforts already in 
place. GAO recognizes these efforts but maintains that the 
recommendations remain valid. 

[Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-06-326].

To view the full product, including the scope and methodology, click on 
the link above. For more information, contact Kathryn G. Allen at (202) 
512-7118 or allenk@gao.gov. 

[End of section]

Contents:

Letter:

Results in Brief:

Background:

Identification of Offenders Living in Long-Term Care Facilities and 
Assessment of the Risk They Pose Limited by Data Shortcomings:

States Required to Notify Community about Registered Sex Offenders, but 
Extent of Notification Varies:

Supervision and Separation of Long-Term Care Facility Residents Largely 
Based on Behavior, Not Prior Convictions:

Conclusions:

Recommendations for Executive Action:

Agency and State Comments and Our Evaluation:

Appendix I: Scope and Methodology:

Appendix II: Registered Sex Offenders Living in Nursing Homes and ICFs- 
MR:

Appendix III: Comments from the Department of Justice:

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

Appendix V: GAO Contact and Staff Acknowledgments:

Tables:

Table 1: Estimated Submission Rates to NSOR, by States Reviewed, 2005:

Table 2: Parolees Identified as Living in Long-term Care Facilities in 
States Reviewed, 2005:

Table 3: Validity of Offender Address Data by Data Source:

Table 4: Registered Sex Offenders Identified as Living in Nursing Homes 
and ICFs-MR, by State, 2005:

Abbreviations:

CMS: Centers for Medicare & Medicaid Services: 
FBI: Federal Bureau of Investigation: 
HHS: Department of Health and Human Services: 
HHS-OCR Department of Health and Human Services Office for Civil 
Rights: 
HIPAA: Health Insurance Portability and Accountability Act of 1996: 
ICF- MR: intermediate care facility for people with mental retardation: 
NNHS: National Nursing Homes Survey: 
NORS: National Ombudsman Reporting System: 
NSOR: National Sex Offender Registry: 
OSCAR: Online Survey, Certification and Reporting system:

United States Government Accountability Office:
Washington, DC 20548:

March 31, 2006:

The Honorable Charles E. Grassley: 
Chairman: 
Committee on Finance: 
United States Senate:

The Honorable Henry A. Waxman: 
Ranking Minority Member: 
Committee on Government Reform: 
House of Representatives:

The Honorable Betty McCollum: 
House of Representatives:

Approximately 16,000 nursing homes and 6,600 intermediate care 
facilities for people with mental retardation (ICF-MR) participate in 
Medicare, Medicaid, or both, and receive federal funding to care for 
their residents. Federal Medicare and Medicaid funds accounted for a 
significant portion of total funding to these nursing homes and ICFs- 
MR--approximately $43 billion of a total of over $110 billion in 2003, 
with about $37 billion spent in nursing homes and $6 billion in ICFs- 
MR.[Footnote 1] Because of the large amount of federal funding these 
long-term care facilities receive, there is a strong interest in 
ensuring the safety and well-being of their vulnerable residents.

Concerns about the quality of care provided to residents of long-term 
care facilities are long-standing. Since 1998, a number of 
congressional hearings have focused on ensuring the quality of care and 
protecting the safety and rights of nursing home residents. Recently, 
news outlets and others have reported accounts of convicted sex 
offenders residing in nursing homes and, in some cases, allegedly 
abusing other residents. Often, nursing home staff and residents' 
families were not informed when convicted sex offenders were residents. 
The admission to long-term care facilities of individuals on parole for 
non-sex offenses[Footnote 2] raises similar concerns about the 
potential for abuse. Because of your concerns about resident safety, we 
are providing information on (1) the prevalence of offenders living in 
nursing homes and ICFs-MR and the extent of abuse caused by such 
offenders in these long-term care facilities; (2) whether federal and 
state laws provide for notification of staff, residents' families, and 
residents when such offenders live in these long-term care facilities; 
and (3) the extent to which these offenders are subject to supervision 
or separation requirements that differ from those for other residents.

To determine the prevalence of registered sex offenders residing in 
these long-term care facilities nationwide, we obtained the Federal 
Bureau of Investigation's (FBI) National Sex Offender Registry (NSOR), 
a national database utilized by law enforcement that compiles 
information about registered sex offenders submitted by all 50 states 
and the District of Columbia, as of January 3, 2005. We used the NSOR 
to identify the home addresses of registered sex offenders and matched 
them with the addresses of nursing homes and ICFs-MR included in the 
Centers for Medicare & Medicaid Services' (CMS) Online Survey, 
Certification and Reporting system (OSCAR) database, which compiles the 
results of state nursing home surveys. To assess the completeness of 
the NSOR data, we requested complete sex offender registries from eight 
states--California, Florida, Illinois, Minnesota, Ohio, Oklahoma, New 
Jersey, and Utah--which we selected on the basis of a number of 
criteria, including variation in geographic location and in the number 
of registered sex offenders identified as living in these states' long- 
term care facilities based on our preliminary analyses. Seven of the 
eight states provided their sex offender registries, and we compared 
the total number of sex offenders on each of the state registries with 
the total number for each state identified through the NSOR.[Footnote 
3] In the course of comparing the results of the NSOR and state sex 
offender registries analyses, we became aware that the NSOR did not 
include the full sex offender registries of these states. Since no 
national database exists for parolees that includes address 
information, we also obtained parolee databases from each of these 
eight states. We matched parolee addresses with addresses for nursing 
homes and ICFs-MR from OSCAR for the eight states; however, because of 
the limited number of states reviewed, we could not generalize these 
results as representative of all states. We obtained state registries 
and parolee databases from January through September 2005.

To gather information about the administration and content of state sex 
offender registries, including their submission of records to the NSOR, 
we interviewed state registry administrators from all eight states and 
submitted a questionnaire via e-mail to all other states, receiving 
responses from an additional 20 states. To assess the reliability of 
FBI and state data, we discussed data quality control procedures, 
reviewed relevant documentation with officials, and conducted 
electronic data testing. We determined that while the NSOR does not 
include all registered or convicted sex offenders, its records are 
regularly audited and are sufficiently reliable for the purposes of 
this report. The lack of comprehensiveness of the data was evaluated 
and taken into account in our discussion of the results. We determined 
that the OSCAR database and state parolee databases were sufficiently 
reliable for our purposes.

To obtain information about resident abuse perpetrated by sex offenders 
and parolees residing in long-term care facilities, we reviewed 
existing research and GAO reports and interviewed officials of industry 
associations and long-term care ombudsmen. Within the eight states we 
reviewed, we also interviewed state officials responsible for nursing 
home and ICF-MR licensing, as well as administrators at 29 of 32 
judgmentally selected long-term care facilities---4 from each state. We 
also relied on these interviews, interviews with federal officials, and 
a review of federal and state laws to determine whether federal and 
state laws provide for notification of facility staff, residents, and 
residents' families when offenders live in these long-term care 
facilities and the extent to which offenders are supervised and 
separated from other residents. In the eight states we reviewed, we 
also examined states' public sex offender Web site registries to 
determine what information on registered sex offenders is available to 
the public. We conducted our work from September 2004 through February 
2006 in accordance with generally accepted government auditing 
standards. (For additional information on our scope and methodology, 
see app. I.)

Results in Brief:

Using the FBI's NSOR, we identified about 700 registered sex offenders 
living in long-term care facilities during 2005, representing 0.05 
percent of the 1.5 million residents of these facilities. About 3 
percent of nursing homes and 0.7 percent of ICFs-MR housed a registered 
sex offender during 2005. Almost 90 percent of registered sex offenders 
we identified lived in nursing homes and were considerably younger than 
the general nursing home population, with 57 percent under age 65 
compared to about 10 percent of all nursing home residents. However, 
our count is understated because of limitations in data availability. 
For example, while the NSOR is a national database that compiles 
information about registered sex offenders submitted by all 50 states 
and the District of Columbia, it does not include convicted sex 
offenders who are not on state registries, including those who were 
convicted or released from prison before registration requirements went 
into effect and those who are required by law to register but do not 
comply. While noncompliance is difficult to track, four of the eight 
states we reviewed estimated noncompliance rates that ranged from 4.5 
percent to 25 percent of all sex offenders required to register. In 
addition, because of a lack of resources or an inability to comply with 
certain FBI reporting requirements, states have had varying degrees of 
difficulty submitting their full state registries to the NSOR. While 
the FBI does not track NSOR submission rates, our analyses of the 
registries obtained from seven of the eight states we reviewed 
indicated NSOR submission rates averaging about 57 percent, ranging 
from 1 percent to 83 percent, while the 20 other states that responded 
to our e-mail questionnaire reported NSOR submission rates ranging from 
46 percent to 100 percent. Because there is no national data source on 
parolees that includes address information, we obtained parolee 
databases from all eight states we reviewed and identified 204 
offenders on parole for non-sex offenses living in long-term care 
facilities. We could not estimate the overall risk these registered sex 
offenders and parolees pose to other residents in long-term care 
facilities because data are not available in the reviewed states on 
abuse perpetrated specifically by residents who have prior convictions. 
Facility administrators we interviewed more frequently expressed 
concern over the potential for abuse by residents with cognitive 
impairments or mental illness than those with prior convictions.

Federal law requires state law enforcement agencies to release relevant 
information about registered sex offenders when necessary to protect 
the public, but we did not identify a similar federal law for the 
parolee population. States have broad discretion on how to implement 
this requirement for registered sex offenders and often do so through a 
process known as community notification. Consequently, the extent to 
which states' community notification laws apply to all registered sex 
offenders or explicitly include long-term care facilities varies, and 
this variation was evident in the eight states we reviewed. For 
example, while two states we reviewed apply uniform community 
notification requirements to all registered sex offenders, the six 
remaining states vary these notification requirements depending on the 
crime committed by the registered sex offender or risk to re-offend. 
Similarly, only half of the states we reviewed specify that long-term 
care facilities be notified when at least some registered sex offenders 
are residents. Long-term care facilities in the remaining four states, 
or in states where community notification of these facilities is not 
required for all registered sex offenders, may not be aware of 
residents who are offenders or must rely on other methods, such as 
publicly available state Web site registries, to identify such 
individuals. While we identified no federal law that requires community 
notification for parolees when they enter long-term care facilities, 
three of the eight states we reviewed require community notification 
for all or a subset of parolees. When long-term care facility residents 
are known offenders, opinions differ among state and long-term care 
officials we interviewed as to whether sharing information with other 
residents and staff about such offenders' prior convictions violates 
the Privacy Rule issued under the Health Insurance Portability and 
Accountability Act of 1996 (HIPAA). For example, while some state 
agency officials indicated that the HIPAA Privacy Rule only applied to 
individuals' health information and not their prior convictions, some 
long-term care facility officials thought the HIPAA Privacy Rule 
prohibited the sharing of any such information in most cases. Despite 
concerns that they may violate the HIPAA Privacy Rule by disclosing 
information about the prior convictions of certain residents, some long-
term care facility officials said that they would still notify staff if 
they became aware of such residents. We brought the issue of long-term 
care facilities' uncertainty regarding the applicability of the HIPAA 
Privacy Rule to the attention of the Department of Health and Human 
Services (HHS) in the event that this issue gains more prominence in 
the future.

Having a prior conviction typically is not sufficient to subject 
offenders to supervision or separation requirements that differ from 
those for other residents, according to long-term care facility 
officials we interviewed. Instead, these facilities generally base 
supervision and separation requirements for residents on behavioral 
issues as they arise. Several long-term care ombudsmen, industry 
association officials, and facility officials in the states we reviewed 
indicated that the residents they are most concerned about in terms of 
behavioral problems are those with mental illness, particularly 
dementia, for which behaviors are apt to change as the disease 
progresses. Officials at only 2 of the 29 long-term care facilities we 
contacted said they have a specific policy to separate offenders from 
other residents based solely on their prior convictions. Even if long- 
term care facility officials wanted to impose different separation and 
supervision requirements on offenders, their ability to do so may be 
limited. Not only are long-term care facilities not always notified 
when individuals with prior convictions enter them, the assessment 
tools they use to determine the health care needs of residents 
typically do not gather information about prior convictions. In 
addition, in the event that a facility obtained such information, 
federal and state laws that we reviewed generally do not provide for 
specific supervision or separation requirements for facility residents 
with prior convictions.

While it was not part of our original objectives to fully evaluate the 
NSOR, in the course of our work using the NSOR to identify registered 
sex offenders residing in long-term care facilities, we became aware 
that the database was incomplete for the seven states we reviewed for 
this purpose. Therefore, to ensure that NSOR fulfills its potential as 
a national database on registered sex offenders, we recommend that the 
Attorney General direct the FBI to assess the completeness of the NSOR, 
including state submission rates, and to evaluate options for making it 
a more comprehensive national database of registered sex offenders.

In commenting on a draft of this report, the Department of Justice 
(DOJ) said it believes the recommendations are unnecessary because the 
FBI already performs assessments of the NSOR and explores options for 
improvement. While we acknowledge the states' and FBI's efforts and 
progress to date, we maintain that the intent of the recommendations 
remains valid based on our analysis of a sample of states that 
indicates some states are not submitting a significant percentage of 
registered sex offender records to the NSOR. However, to be more 
specific regarding the need to assess the completeness of NSOR, we 
revised the recommendations to clarify that the FBI should assess state 
submission rates. HHS commented that the report will help to resolve 
much of the uncertainty about the application of the HIPAA Privacy Rule 
to the disclosure of conviction information by a facility, including 
clarifying that information could be used for activities necessary for 
the safe operation of the facility or disclosures that are required by 
state laws.

Background:

Nursing homes provide a residential setting and a range of health care 
services for individuals who can no longer care for themselves because 
of physical or mental limitations. According to the most recent 
National Nursing Homes Survey (NNHS), approximately 90 percent of 
nursing home residents were age 65 and older and more than two-thirds 
were female.[Footnote 4] ICFs-MR are intended to provide a residential 
setting for treatment, rehabilitation, and supervision of people who 
have mental retardation or other disabilities, such as seizure 
disorders or behavior problems. In 2005, approximately 85 percent of 
ICF-MR residents were from 22 to 65; only 7 percent of the total 
resident population was over 65 years of age. In addition, unlike the 
nursing home population, the majority of ICF-MR residents were male.

Approximately 1.5 million individuals lived in Medicaid-and Medicare- 
certified nursing homes and ICFs-MR in 2005.[Footnote 5] Federal 
Medicaid and Medicare funds accounted for approximately 33 percent of 
total spending on nursing homes, and the remaining funds were from a 
combination of state, local, and private sources in 2003. In the same 
year, ICFs-MR, which are funded almost exclusively by Medicaid, 
received about 58 percent of their total funding from federal Medicaid 
funds and the remainder from state Medicaid dollars. Medicaid, a joint 
federal-state program that finances health care coverage for certain 
categories of low-income individuals, is the primary payment source for 
long-term care services for older people with low incomes and limited 
assets. Medicaid pays for an array of long-term care services, 
including services to assist people with activities of daily living 
like eating, dressing, bathing, and using the bathroom. In contrast, 
Medicare, which covers a variety of health care services and items for 
individuals who are 65 or older, have end-stage renal disease, or are 
disabled, does not pay for most long-term care services. Medicare 
covers short-term skilled nursing care following a hospital stay.

To qualify for Medicare or Medicaid funding, these long-term care 
facilities must meet certain federal requirements. For example, they 
are required to conduct resident assessments that examine areas such as 
demographic information, cognition, mood and behavior, psychosocial 
well-being, health conditions, and physical functioning. For example, 
the Preadmission Screening and Resident Review, which is required by 
federal law to determine whether the potential resident needs nursing 
home care, includes an assessment of mental capacity. Although federal 
regulations require that a resident assessment be conducted prior to 
admission to ICFs-MR, there is no standardized assessment tool and 
admission can be based on a prior assessment by an outside source. 
Individuals being admitted to an ICF-MR generally meet certain 
criteria, including having an intellectual functioning level below 70 
to 75 and significant limitations in two or more adaptive skill 
areas.[Footnote 6] In addition, at a minimum, resident assessments are 
conducted annually by nursing home and ICF-MR facility staff after 
admission in order to continually address a resident's needs. For each 
resident for whom they receive Medicare or Medicaid funding, these long-
term care facilities are also required to develop a plan of care that 
addresses the resident's medical, social, and other needs, as 
determined by the resident assessment. Long-term care facilities are 
also required to protect residents' rights and privacy.[Footnote 7] In 
addition, the Privacy Rule issued under HIPAA provides individuals with 
protections regarding the confidentiality of their health information 
and restricts the use and disclosure of individuals' health information 
by health care providers, including nursing homes and ICFs-MR.[Footnote 
8]

As a condition of Medicare or Medicaid participation, long-term care 
facilities must report incidents of abuse according to state 
requirements. CMS defines abuse as the willful infliction of injury, 
unreasonable confinement, intimidation, or punishment with resulting 
physical harm, pain, or mental anguish. Physical abuse generally 
includes hitting, slapping, pushing, and sexual abuse, which is 
nonconsensual sexual contact or nonconsensual sexual involvement of any 
kind. Although the commission of a sexual offense may result in an 
incident of abuse, a uniform definition of sexual offense does not 
exist, and states define sexual offenses in their respective criminal 
codes. Some examples of sexual offenses include rape, sexual assault, 
and incest.[Footnote 9] In some states, related sexual offenses include 
child pornography and willful indecent exposure in public.

Federal statute established the Jacob Wetterling Crimes Against 
Children and Sex Offender Registration Program in 1994.[Footnote 10] 
The statute required every state to have a program to register sex 
offenders by September 1997, and required the Attorney General to 
provide states with guidelines for developing their programs.[Footnote 
11] At a minimum, an individual convicted of a criminal offense against 
a minor or of a sexually violent offense must register a current 
address for 10 years following his/her release from prison or placement 
on parole, supervised release, or probation. In addition, an individual 
who has one or more prior sexual offense convictions, has been 
convicted of an aggravated offense, or is determined to be a sexually 
violent predator must register a current address for life.[Footnote 12] 
States may impose more stringent registration requirements on a broader 
class of offenders than required by federal law. The law also mandates 
that registered sex offenders verify their addresses at least annually 
and that registered offenders classified as sexually violent predators 
verify their addresses quarterly. Registered sex offenders must notify 
local law enforcement officials within their state of address changes, 
and those who move to a different state must comply with registration 
requirements in the new state. States that do not comply with the 
Wetterling Program requirements are subject to a 10 percent reduction 
in their Byrne Formula Grant law enforcement funding.[Footnote 13]

The statute establishing the Wetterling Program was amended twice in 
1996. The first amendment, Megan's Law, required states to release 
information about registered sex offenders when necessary to protect 
the public, but this law did not specify how states must give 
notification.[Footnote 14] The second amendment, the Pam Lychner Sexual 
Offender Tracking and Identification Act of 1996,[Footnote 15] mandated 
the FBI's creation of a national database now known as the 
NSOR.[Footnote 16] According to the FBI, this national database 
combines sex offender registries from all of the states to help law 
enforcement officials track sex offenders on a national level.

Research on sex offender recidivism suggests that the majority of 
individuals previously convicted of sex offenses do not commit 
additional sex offenses, with one such study estimating that about 14 
percent had a new sex offense charge or conviction within 5 years of 
their release from prison, increasing to 27 percent after 20 years. At 
the same time, however, research also indicates that sex offenses are 
underreported. While it is difficult to predict re-offense for any 
individual, certain factors such as sexual deviancy, antisocial 
orientation, and an adverse family environment may contribute to a 
higher likelihood of a re-offense. Those who have strong social 
supports, such as a supportive family and a stable job, may be less 
likely to re-offend. In addition, the likelihood of re-offending may 
diminish as the sex offender ages.

Federal law requires that registered sex offenders be tracked on a 
national and state level; however, parolees are generally monitored and 
supervised by each state. Individuals released from prison prior to the 
completion of their sentences may be subject to certain conditions and 
supervised as parolees for a specified period. Typically the length of 
time states set for parole is 1 to 3 years, although certain crimes and 
sentencing situations may require more or less time. An individual can 
be convicted of a range of crimes from fraud or forgery to murder and 
be eligible for parole. As of December 2003, about 775,000 adults were 
on parole from federal and state prisons nationwide.

Identification of Offenders Living in Long-Term Care Facilities and 
Assessment of the Risk They Pose Limited by Data Shortcomings:

Using the NSOR, we identified 683 registered sex offenders living in 
long-term care facilities during 2005. However, this understates the 
national prevalence of convicted sex offenders residing in long-term 
care facilities for a number of reasons. While the NSOR is a national 
database that compiles information about registered sex offenders 
submitted by all 50 states and the District of Columbia, it does not 
include convicted sex offenders who are not on state registries, 
including those sex offenders who are required by law to register but 
choose not to comply. It also does not include all registered sex 
offenders, as states have had varying degrees of difficulty submitting 
their records to the NSOR because of technical problems, lack of 
resources, or inability to provide the required FBI number for certain 
offenders. Because there is no national data source on parolees that 
includes address information, we also obtained parolee databases from 
the eight states we reviewed and identified 204 offenders on parole for 
non-sex offenses living in long-term care facilities. The risk of abuse 
within nursing homes or ICFs-MR by residents with prior convictions is 
unclear because states we reviewed do not report the prior convictions 
of residents who commit abuse; however, facility administrators we 
interviewed more frequently expressed concern about the potential for 
abuse by residents with cognitive impairments or mental illness than by 
residents with prior convictions.

Most Sex Offenders Identified Were Male, Were under Age 65, and Resided 
in a Small Number of Nursing Homes and ICFs-MR:

Using the NSOR, we identified 683 registered sex offenders living in 
long-term care facilities during 2005, representing about 0.05 percent 
of the total 1.5 million residents of nursing homes and ICFs-MR. (See 
app. II.) Of the approximately 16,000 nursing homes and 6,600 ICFs-MR 
that participate in Medicare or Medicaid, we identified 3 percent of 
nursing homes (470) and 0.7 percent of ICFs-MR (46) as housing at least 
1 registered sex offender during 2005.

About 88 percent of the registered sex offenders we identified resided 
in nursing homes, while the remaining 12 percent resided in ICFs-MR. 
Sex offenders living in nursing homes were younger than the general 
nursing home population, while those in ICFs-MR had a similar age 
distribution as the general ICF-MR population. About 57 percent of 
registered sex offenders we identified as living in nursing homes were 
under age 65, compared to about 10 percent of the general nursing home 
population, and 30 percent were under age 50. Most sex offenders--95 
percent--identified as living in ICFs-MR were under age 65, which is 
similar to the age distribution in the general population of these 
facilities. Similarly, nearly all--99 percent--registered sex offenders 
we identified as residing in long-term care facilities were male, which 
is consistent with the gender of registered sex offenders overall. 
Among registered sex offenders for whom we had information on the 
nature of their crimes, the majority of convictions were for rape and 
sexual assault of adults and minors.

Number of Offenders Identified as Living in Long-term Care Facilities 
Is Understated:

The number of offenders that we identified as living in long-term care 
facilities is understated because of shortcomings in the data. 
Specifically, although national in scope, the NSOR does not include 
certain convicted sex offenders who are not on state registries because 
the registries did not exist at the time they were convicted or 
released from prison or because their registration period has expired. 
The NSOR also does not include all of the records of sex offenders who 
are registered in the states' registries because some states have had 
difficulty submitting their records to the NSOR. NSOR records for 
convicted sex offenders who chose not to comply with registration 
requirements may be incomplete or missing. In addition, since no 
national data source for parolees exists that includes parolee 
residence information, our data only include numbers of parolees from 
the eight states we reviewed.

State Registries Do Not Include All Convicted Sex Offenders:

While some states already had sex offender registries in place, the 
Wetterling Program statute mandated that all states implement a 
registry by September 1997.[Footnote 17] Most state registries only 
include those sex offenders convicted or released from prison after a 
specified date, generally after 1990.[Footnote 18] Consequently, those 
convicted or released before the specified date were not required to 
register and therefore are not included in our analysis. This 
limitation may help explain the age distribution of registered sex 
offenders we identified as living in nursing homes. While the majority 
of offenders identified in nursing homes were under the age of 65, this 
could be a consequence of the limited period that sex offender 
registries have existed rather than an accurate reflection of the age 
distribution of convicted sex offenders living in nursing homes, since 
many elderly sex offenders would not be registered if their convictions 
predated the implementation of their state's registry.

The nearly 700 registered sex offenders we identified through the NSOR 
database as living in long-term care facilities also do not include 
convicted sex offenders whose registration period expired or whose 
information was missing because they did not comply with registration 
requirements. While noncompliance is difficult to track, four of the 
reviewed states provided us with estimated noncompliance rates ranging 
from 4.5 percent to 25 percent. Similarly, the advocacy organization, 
Parents for Megan's Law, released estimates in 2003 that 24 percent of 
sex offenders nationally fail to comply with registration requirements. 
Sex offenders may fail to comply for several reasons, including a lack 
of understanding about registration requirements or to avoid the 
possible negative consequences experienced by some registered sex 
offenders, such as the loss of a job, harassment, social 
stigmatization, or physical assault.

State Submissions to the NSOR Do Not Include All Registered Sex 
Offenders:

We found a range of submission rates by state registries to the NSOR, 
which suggests that the NSOR may be missing a portion of sex offenders 
who are registered in states. Registry administrators from the 20 
states that responded to our e-mail questionnaire estimated their 
submission rates to be from 46 percent to 100 percent of the total 
number of records in their state registries. Most reported that at 
least 80 percent of their records were submitted, while 2 states 
reported that they were only able to submit about half of their 
records. We also compared the total number of sex offenders included in 
the state registries to the number included in the NSOR for 7 of the 8 
states we reviewed. (See table 1.) The NSOR included about 57 percent 
of sex offenders registered in these states, with submission rates 
ranging from 1 to 83 percent. For example, Utah had submitted about 1 
percent of its registry to the NSOR. While the state intends to fully 
submit its registry to the NSOR in the future, it currently lacks the 
resources to do so, according to a state official. However, the FBI 
considers state participation in the national database to be in 
compliance with federal requirements if a state has submitted at least 
one record to the NSOR.[Footnote 19] A DOJ official confirmed that all 
states have been determined to be in compliance with NSOR submission 
requirements, based on FBI notifications regarding each state's 
participation in the NSOR, and was not aware of any state that had been 
penalized with the loss of Byrne Formula Grant law enforcement funding 
solely on the basis of the extent of state NSOR participation.

Table 1: Estimated Submission Rates to NSOR, by States Reviewed, 2005:

State: Florida; 
Number of sex offenders listed on state registry: 34,810; 
Number of sex offenders listed on NSOR, January 3, 2005: 25,494; 
State's submission rate to NSOR (percentage): 73%.

State: Illinois; 
Number of sex offenders listed on state registry: 20,690; 
Number of sex offenders listed on NSOR, January 3, 2005: 13,349; 
State's submission rate to NSOR (percentage): 65%.

State: Minnesota; 
Number of sex offenders listed on state registry: [A]; 
Number of sex offenders listed on NSOR, January 3, 2005: 9,769; 
State's submission rate to NSOR (percentage): [A].

State: New Jersey;
Number of sex offenders listed on state registry: 11,382; 
Number of sex offenders listed on NSOR, January 3, 2005: 9,454; 
State's submission rate to NSOR (percentage): 83%.

State: Ohio[B]; 
Number of sex offenders listed on state registry: 16,864; 
Number of sex offenders listed on NSOR, January 3, 2005: 2,409; 
State's submission rate to NSOR (percentage): 14%.

State: Oklahoma; 
Number of sex offenders listed on state registry: 5,235; 
Number of sex offenders listed on NSOR, January 3, 2005: 4,234; 
State's submission rate to NSOR (percentage): 81%.

State: Utah; 
Number of sex offenders listed on state registry: 7,409; 
Number of sex offenders listed on NSOR, January 3, 2005: 49; 
State's submission rate to NSOR (percentage): 1%.

Average: State's submission rate to NSOR (percentage): 57%. 

Sources: GAO analysis of the FBI's NSOR as of January 3, 2005; state 
sex offender registries as submitted by states to GAO from January 
through August 2005.

[A] Minnesota initially submitted to GAO only those offenders residing 
in Minnesota who could potentially be identified as nursing home 
residents, which was the purpose of our request for the data, and 
therefore excluded offenders who were listed as out of state, deported, 
homeless, civilly committed, in the witness protection program, or 
address unknown. Consequently, we could not determine an overall 
submission rate. As of February 2006, Minnesota reported that 73 
percent of its active registrants had been accepted by the NSOR.

[B] As of January 3, 2005, Ohio had not submitted the majority of its 
sex offender registry because of technical problems, but a state 
official reported that the state submitted registry data on computer 
disk in August 2005.

[End of table]

Registry administrators from among the 8 states we reviewed and the 20 
additional states that responded to our e-mail questionnaire reported 
that several factors complicate their efforts to submit complete sex 
offender registries to NSOR. For example, registry administrators 
frequently responded that they were not able to submit records of 
registered sex offenders who did not have FBI numbers. FBI numbers are 
required by the FBI for all records submitted to the NSOR to ensure 
positive identification of individuals for the purposes of employment 
background checks.[Footnote 20] States may lack FBI numbers for several 
types of offenders, such as juvenile sex offenders who do not receive 
FBI numbers or sex offenders from other states. If a sex offender comes 
from out of state, his/her FBI number can be obtained from the state 
where the conviction occurred, but it can be labor-intensive if the 
other state does not cooperate or never submitted fingerprint 
information to establish the offender's FBI number.[Footnote 21] 
Registry administrators in two of the states we reviewed estimated that 
in recent years about 30 percent of the records they submitted to the 
NSOR were rejected as incomplete.[Footnote 22] In addition, states are 
required to verify information, including home address, for each 
registered offender at least annually and quarterly for registered 
offenders classified as sexually violent predators, a process that can 
also be labor-intensive.[Footnote 23] If states are unable to verify an 
offender's address information, the offender should be considered 
noncompliant, and the NSOR record will not be up-to-date nor reflect 
current address information. Some states have also experienced 
technical difficulties submitting their registry records to the NSOR. 
An FBI official told us that states that had registries prior to the 
creation of the NSOR had difficulty reprogramming their registry 
databases to conform to the NSOR formats. One of the states we reviewed 
did not realize until 2005 that only a fraction of its records were 
being submitted to the NSOR because of a technical problem, and it is 
currently submitting records on computer disks while making plans to 
implement a system for automatic electronic submission of its full sex 
offender registry to the NSOR.

Although the FBI does not track states' submission rates to the NSOR, 
it does periodically assess state participation in the NSOR and 
provides assistance to help states improve the comprehensiveness and 
accuracy of their registries. In addition to the requirement that 
states annually validate registry records, we were informed that the 
FBI conducts triennial audits of states' participation in the NSOR. 
During fiscal year 2005, the FBI also conducted a fiscal audit, 
assessed states' level of participation in the NSOR and requested 
information from states about what assistance they need to improve 
their participation. DOJ provides grants to help states improve their 
law enforcement information systems, which states have utilized for 
enhancements to their sex offender registries such as enabling the 
automatic transmission of records to the NSOR and for monitoring data 
accuracy. DOJ informed us that it also provides training and technical 
assistance to states, and that the FBI has an advisory group that is 
reviewing issues such as state submission of data to the NSOR and the 
process for the verification and validation of NSOR records.

Approximately 200 Parolees Identified as Living in Long-term Care 
Facilities in Eight States:

Using data provided by each of the eight states we reviewed, we 
identified 204 parolees as residents of long-term care facilities. (See 
table 2.) Because there is no national source of data on parolees that 
includes their home address information, our numbers are limited to the 
eight states and cannot be generalized as representative of all states. 
Among parolees for whom we had information on the nature of their 
crimes, the convictions were most commonly for burglary, assault, 
murder, or drug-related offenses.

Table 2: Parolees Identified as Living in Long-term Care Facilities in 
States Reviewed, 2005:

State: California; 
Nursing homes: 63; 
ICFs-MR: 4; 
Total[A]: 67.

State: Florida; 
Nursing homes: 7; 
ICFs-MR: 0; 
Total[A]: 7.

State: Illinois; 
Nursing homes: 70; 
ICFs-MR: 4; 
Total[A]: 74.

State: Minnesota; 
Nursing homes: 2; 
ICFs-MR: 0; 
Total[A]: 2.

State: New Jersey; 
Nursing homes: 2; 
ICFs-MR: 1; 
Total[A]: 3.

State: Ohio; 
Nursing homes: 42; 
ICFs-MR: 1; 
Total[A]: 43.

State: Oklahoma; 
Nursing homes: 6; 
ICFs-MR: 0; 
Total[A]: 6.

State: Utah; 
Nursing homes: 2; 
ICFs-MR: 0; 
Total[A]: 2.

Total: Nursing homes: 194; 
ICFs-MR: 10; 
Total[A]: 204. 

Sources: GAO analysis of parolee databases for eight reviewed states, 
January through September 2005; CMS's OSCAR database, 2004.

[A] Results do not include parolees who were also listed on state sex 
offender registries or the NSOR.

[End of table]

Abuse by Offenders Who Live in Long-term Care Facilities Not Tracked, 
but Risk May Not Be Widespread:

Long-term care facilities participating in Medicare or Medicaid are 
required to report all allegations of abuse and neglect to officials in 
accordance with applicable state law and, in the case of nursing homes, 
this includes reporting to the state.[Footnote 24] This requirement 
would encompass the reporting of abuse committed by staff or residents. 
In the eight states we reviewed, long-term care facilities do stratify 
reported abuse into categories, such as physical, sexual, financial, or 
resident-to-resident abuse; however, they do not report information on 
whether residents alleged to have caused abuse have prior convictions. 
The National Ombudsman Reporting System (NORS) also collects nursing 
home abuse data on a national level and includes various categories of 
abuse, such as incidents that occur between residents and incidents 
perpetrated by nursing home staff. Similar to the states we reviewed, 
NORS does not track whether residents alleged to have abused other 
residents have prior convictions.

Because data are not available nationally or in our reviewed states on 
abuse perpetrated specifically by residents who have prior convictions, 
the potential risk for abuse by offenders residing in long-term care 
facilities cannot be accurately estimated. However, based on a number 
of factors, including the small percentage of facilities identified as 
housing offenders, the risk may not be widespread. For example, 
offenders residing in nursing homes or ICFs-MR who have significant 
physical limitations may be unable to commit abuse against other 
residents. In addition, research on recidivism by sex offenders also 
suggests that most do not re-offend and that the risk of re-offending 
may decline with age.

In our interviews with officials of long-term care facilities, state 
nursing home associations, and state ombudsmen for long-term care, 
concern was more frequently expressed about the behavior and potential 
for abuse by cognitively impaired and mentally ill residents than about 
abuse by residents with prior convictions. Several of those interviewed 
mentioned they were concerned about the potential for abuse by 
residents with Alzheimer's disease or dementia, a disease for which 
their behavior may change significantly after their admission and 
original assessment. The administrator of a facility in Ohio that 
specializes in residents with behavioral issues and that has housed 
multiple offenders said that he has had fewer problems with his 
residents who are identified sex offenders than with other residents 
who have behavioral problems. Several sources, including ombudsmen, a 
researcher, and a nursing home advocate, suggested that a resident's 
behavioral issues are sometimes not fully disclosed to a nursing home 
upon admission or that some nursing homes with low occupancy may be 
more likely than others to accept mentally ill patients in order to 
increase their occupancy levels.

Long-term care facility officials we interviewed, some of whom knew 
they have had offenders as residents and some of whom spoke 
hypothetically, said they would use their judgment to determine whether 
a registered sex offender or parolee could appropriately be cared for 
in their facilities. Several long-term care facility administrators 
told us that if they discovered a resident was an offender, they would 
evaluate the potential risk posed by that individual on a case-by-case 
basis. For example, the facility administrator may determine the degree 
of safety risk on the basis of whether the offender's health status is 
such that the individual cannot move independently. If the 
administrator determines that the risk is greater than the long-term 
care facility can manage, the facility may choose not to admit the 
offender.

States Required to Notify Community about Registered Sex Offenders, but 
Extent of Notification Varies:

Federal law requires state law enforcement agencies to release relevant 
information about registered sex offenders when necessary to protect 
the public, but we did not identify a similar federal requirement 
pertaining to the parolee population. The federal requirement for 
registered sex offender notification allows states to implement this 
requirement at their discretion, within broad federal guidelines. 
Consequently, the extent to which states' community notification laws 
apply to all registered sex offenders or explicitly include nursing 
homes and ICFs-MR varies. Absent direct notification, these facilities 
may not know they house offenders or may only become aware of offenders 
through other means. For example, in the case of registered sex 
offenders, facilities may identify some offenders by reviewing publicly 
available Web sites, while for parolees, they may become aware of the 
person's criminal background from a parole officer. When facility 
residents are known offenders, differing interpretations exist among 
states, industry, and long-term care facility officials as to whether 
sharing information about their prior convictions may violate the 
Privacy Rule issued by HHS under HIPAA.

Federal Law Requires States to Provide Community Notification for 
Registered Sex Offenders, but Direct Notification to Long-term Care 
Facilities Varied:

Megan's Law, a 1996 amendment to the Wetterling Program statute, 
required each state to release information about registered sex 
offenders when necessary to protect the public. The law applied 
specifically to registered sex offenders and not to convicted sex 
offenders who were not obligated to register.[Footnote 25] Although 
Megan's Law stipulated that information about the victims of registered 
sex offenders was not to be released, it otherwise did not specify the 
information to be disseminated about registered sex offenders, did not 
mandate that community notification be uniform for all registered sex 
offenders, and did not specify how states were to release information.

Consequently, states' community notification laws vary, particularly in 
terms of the extent to which notification by law enforcement entities 
applies to all registered sex offenders. Such variation was evident in 
the notification laws of the eight states we reviewed. While two states 
we reviewed--Illinois and Utah--apply community notification 
requirements to all registered sex offenders uniformly in each 
state,[Footnote 26] the community notification requirements in the 
remaining six states--California, Florida, Minnesota, New Jersey, Ohio, 
and Oklahoma--vary depending on the crime committed by the registered 
sex offender.[Footnote 27] For example, New Jersey classifies its 
registered sex offenders into three categories based on their assessed 
risk of re-offending. For sex offenders determined to be lowest risk, 
state law requires notification of law enforcement agencies. In 
contrast, for the highest risk sex offenders, the law requires 
notification of additional entities, including schools, religious and 
youth organizations, and those likely to encounter the offender. 
Similarly, Florida's law explicitly requires broad community 
notification when individuals designated to be sexual predators reside 
in the community, but it does not require broad notification for other 
sex offenders.

Variation also exists in the extent to which state community 
notification laws explicitly require the notification of long-term care 
facilities. Four states we reviewed--California, Illinois, Minnesota, 
and Oklahoma--passed laws in summer and fall 2005 that specified long- 
term care facilities as entities to be notified for at least some 
registered sex offenders who entered them.[Footnote 28],[Footnote 29] 
Notification in these states is conducted by individual facility 
officials, state or law enforcement officials, or registered sex 
offenders themselves. For example, Illinois' law requires long-term 
care facilities to determine whether each resident or potential 
resident is a registered sex offender and to notify staff, residents or 
their legal guardians, and facility visitors when offenders are 
residents.[Footnote 30] Similarly, Oklahoma's law requires notification 
of these facilities by several methods. For example, the Department of 
Corrections must notify the Department of Health when any person in its 
custody seeks placement in these facilities, and the Department of 
Health must then notify the facility of the potential for the placement 
of a registered sex offender. When residents are determined to be 
registered sex offenders, information about them must be displayed in 
the facility in an area that is accessible to staff, visitors, and 
residents. The law in California also requires state officials to 
notify long-term care facilities when registered sex offenders are 
released to them from the Department of Corrections and Rehabilitation, 
the State Department of Mental Health, or other state- operated places 
of confinement. The law does not provide for such notification when sex 
offenders enter long-term care facilities from the community. Unlike 
other states we reviewed, Minnesota's law requires registered sex 
offenders to disclose their status if seeking admission to long-term 
care facilities. Upon receiving such notification from certain 
registered sex offenders, long-term care facilities are responsible for 
sharing this information with other residents or their legal guardians. 
Minnesota also requires law enforcement officials to notify health care 
facilities if they become aware that a registered sex offender has been 
admitted for care.

The other four states we reviewed--Florida, New Jersey, Ohio, and Utah-
-do not specifically require the notification of long-term care 
facilities when registered sex offenders enter them. Long-term care 
facilities in these states, or in states where community notification 
of such facilities is not required for all registered sex offenders, 
may not be aware of residents who are offenders or must rely on other 
methods to identify such individuals. For instance, administrators we 
interviewed at 8 of the 29 long-term care facilities indicated that one 
or more registered sex offenders had lived in their facilities for some 
period. Each of these 8 long-term care facilities was notified about 
the registered sex offenders, although the method of notification 
varied. For example, while 4 facilities were notified before the 
offenders entered them, either by offenders' family members or the 
state department of corrections, the 4 remaining facilities were 
notified after the registered sex offenders were admitted, either by 
local law enforcement officials who were verifying sex offenders' 
residential addresses or by an advocacy group conducting research on 
registered sex offenders living in certain long-term care facilities.

Long-term care facilities may access states' publicly available sex 
offender registry Web sites to determine where registered sex offenders 
reside. A 2003 amendment to the Wetterling Program statute required 
states to maintain a publicly available Web site with information about 
registered sex offenders.[Footnote 31] The law did not provide 
instruction on how these Web sites should be designed or what specific 
information should be included.[Footnote 32] Depending on the state, 
these Web sites provide varying amounts of information to the public 
about registered sex offenders. For example, the Web site registry in 
each of the eight states we reviewed included some address information 
for all or a portion of the state's adult registered sex offenders. 
Five states we reviewed--Florida, Illinois, Ohio, Oklahoma, and Utah-- 
provided the full residential address of all the state's adult 
registered sex offenders, while three others--California, Minnesota, 
and New Jersey--included certain registered sex offenders on their Web 
sites and in some cases did not always list their full addresses. For 
example, Minnesota separates offenders into three levels and includes 
Level 3 offenders--those deemed predatory or most likely to re-offend-
-on its Web site. Approximately 6 percent of the registered sex 
offenders in this state who are living in the community are assigned 
the highest risk level. Similarly, New Jersey includes certain moderate 
and all high-risk registered sex offenders on its Web site, which, 
according to a state official, represents about 16 percent of all 
registered sex offenders in the state. In California, a state official 
told us that its Web site registry includes at least some address 
information for approximately 74 percent of the state's registered sex 
offenders, including full address information for about 57 percent who 
committed crimes considered to be the most serious. The remaining 
approximately 26 percent of the state's registered sex offenders are 
not posted on the Web site because they committed less severe offenses 
or are excluded from the Web site for various reasons, such as not 
being designated sexually violent predators. In addition, for the 
registered sex offenders listed on the Web sites of the eight states we 
reviewed, information is included about the crimes registered sex 
offenders committed; their names, nicknames, or aliases, when 
applicable; date of birth or age; and race or ethnicity.

While the NSOR database is not directly accessible by the general 
public, long-term care facilities can access the recently developed 
National Sex Offender Public Registry maintained by the DOJ.[Footnote 
33] This Web site, which was first launched in May 2005, seeks to 
compile public sex offender registry information available through 
state Web sites, and as of January 2006, it included public registry 
data from all but two states. Although this Web site provides the 
public with one-stop access to states' online sex offender registries, 
it may be of limited usefulness because states' sex offender registry 
Web sites, as described above, do not always include a comprehensive 
list of registered sex offenders.

Although Community Notification of Parolees Not Uniformly Required, 
Parolees in Long-term Care Facilities Often Identified by Law 
Enforcement:

We did not identify a federal law specifying community notification 
requirements for law enforcement when parolees enter the community that 
was similar to the federal law for registered sex offenders. However, 
three of the eight states we reviewed--Illinois, Minnesota, and 
Oklahoma--passed laws in summer 2005 that require community 
notification for offenders who have committed crimes other than sex 
offenses, including some offenders who are parolees.[Footnote 34] 
Illinois' law requires the state Department of Corrections to give some 
information to certain long-term care facilities when parolees or 
certain other offenders become residents. In addition, these long-term 
care facilities are required to notify the other residents when 
parolees reside in their facilities. In Minnesota and Oklahoma, long- 
term care facilities receive community notification for some 
individuals convicted of non-sex offenses, including some parolees, 
under the same requirements as those for registered sex offenders. 
Minnesota's law applies to individuals convicted of some crimes, 
including murder or kidnapping. Oklahoma's law requires notification 
for individuals who are required to register under the Mary Rippy 
Violent Crime Offenders Registration Act, which includes individuals 
convicted of crimes such as murder or manslaughter in the first degree.

Department of Corrections' officials or other authorities in each of 
the eight states we reviewed stated that as a matter of practice, they 
generally notified long-term care facilities when individuals released 
from prison, including parolees, are placed in such facilities. For 
example, according to officials in Ohio's Department of Rehabilitation 
and Corrections, when an inmate who needs long-term care is paroled, a 
parole officer works with the facility to ensure that medical records 
are transferred and that a plan of care is established to meet the 
needs of the parolee.

Officials Uncertain about Ability of Long-term Care Facilities to 
Disclose Offender Information under the HIPAA Privacy Rule:

While the HIPAA Privacy Rule applies to individually identifiable 
health information, differing interpretations exist among state, 
industry, and long-term care facility officials we interviewed in the 
eight states regarding the applicability of the rule to facilities' 
efforts to notify others about residents who have prior convictions, 
such as those who are registered sex offenders or parolees. These 
difficulties existed regardless of whether this information was 
obtained from a medical record or in another way, such as from a law 
enforcement official. For instance, long-term care agency officials 
from three states we reviewed indicated that protection of health 
information under the HIPAA Privacy Rule did not extend to information 
on prior convictions. In addition, long-term care facility and other 
agency officials from these and three other states we reviewed 
maintained that it was permissible to disclose information about a 
resident's prior convictions to employees in a long-term care facility 
who needed to know in order to provide care for the resident. Yet other 
officials in six of the eight states we reviewed told us they were 
either unsure whether the HIPAA Privacy Rule would be violated by 
sharing information about the prior convictions of any offender living 
in a facility or that they believed the HIPAA Privacy Rule did not 
apply to disclosing such information about residents who are offenders. 
Officials at 11 of the 29 long-term care facilities we interviewed in 
eight states said that they were concerned they would violate the HIPAA 
Privacy Rule if they disclosed information about the prior convictions 
of offenders living in their respective facilities, but indicated that 
they would notify staff if they became aware of such residents.

We brought the issue of long-term care facilities' uncertainty 
regarding the applicability of the HIPAA Privacy Rule to the attention 
of an official of the Department of Health and Human Services Office 
for Civil Rights (HHS-OCR), the federal entity responsible for 
implementing and enforcing the HIPAA Privacy Rule. The official 
indicated that HHS-OCR has not published regulations or other guidance 
specifically regarding the applicability of the HIPAA Privacy Rule to 
the disclosure of information related to prior convictions of long-term 
care facility residents. However, the official stated that to the 
extent that such information is maintained by long-term care facilities 
as protected health information under the HIPAA Privacy Rule, such 
information could be used or disclosed for specifically permitted 
purposes, such as when necessary to run the health care operations of a 
facility or required by another federal or state law. In addition, the 
HHS-OCR official indicated that affected entities, such as long-term 
care facilities, would need to make the determination on a case-by-case 
basis as to whether the information is protected health information, 
and if so, whether its intended use or disclosure is permitted by the 
HIPAA Privacy Rule. The official added that long-term care facilities 
should consult their legal counsel if they have questions in making 
this determination. Although HHS-OCR does maintain a list of answers to 
frequently asked questions about the HIPAA Privacy Rule on its Web 
site, it does not cover this specific issue. In commenting on a draft 
of this report, Department of Corrections officials from one state we 
reviewed stated that it would be helpful for HHS-OCR to describe some 
situations in which it believes HIPAA would not be applicable with 
regard to the disclosure of information about offenders admitted to 
health care facilities. They stated that HHS-OCR's direction to 
approach each case individually is not very helpful and that additional 
guidance would be very useful.

Supervision and Separation of Long-Term Care Facility Residents Largely 
Based on Behavior, Not Prior Convictions:

Residents' prior convictions alone would not be sufficient in most 
cases to subject them to supervision or separation requirements that 
differed from other residents, according to facility officials we 
interviewed. Administrators at only 2 of the 29 long-term care 
facilities we contacted indicated that they have a specific policy to 
separate offenders from other residents based solely on their prior 
convictions. Instead, long-term care facilities in the eight states we 
reviewed typically base supervision and separation decisions on 
behavioral issues that arise. For example, in the states we reviewed, 
several long-term care ombudsmen, industry association officials, and 
facility officials we interviewed indicated that the residents they 
have particular concerns about, in terms of behavioral problems, are 
those with mental illness, such as dementia, for which behaviors are 
apt to change as the disease progresses.

Although most officials we spoke with at long-term care facilities in 
the eight states we reviewed do not supervise or separate offenders 
based solely on their prior convictions, some officials indicated the 
potential for a future need for residential facilities separate from 
long-term care facilities exclusively for certain offenders. For 
instance, Minnesota state officials said that some long-term care 
facilities may be hesitant to accept sex offenders as residents in the 
future, believing that certain sex offenders pose a risk to the safety 
of other residents. Therefore, a state commission has recommended the 
development of secure health care settings that would serve people who 
have committed certain sex offenses and who may not otherwise have 
access to services. In order to establish this facility, state 
officials are working with federal officials to resolve issues related 
to balancing resident rights with the safety interests of the larger 
community.

Even if long-term care facility officials wanted to impose different 
supervision and separation requirements on offenders, numerous factors 
could affect their ability to do so. For example, as previously noted, 
long-term care facilities were not always notified when individuals 
with prior convictions entered them. Federal laws we reviewed do not 
require long-term care facilities to obtain information about prior 
convictions, and among the eight states we reviewed, only Illinois had 
such a requirement.[Footnote 35] In addition, assessment tools long- 
term care facilities in these eight states use to determine the health 
care needs of residents usually are not designed to gather information 
about prior convictions. Even if facilities obtained such information, 
federal and state laws that we reviewed generally do not provide for 
specific supervision or separation practices for facility residents 
with prior convictions.

Conclusions:

Each incident of resident abuse committed by offenders living in 
nursing homes--even if isolated or infrequent--is of concern. However, 
while long-term care facilities may learn that certain of their 
residents are sex offenders or parolees through required community 
notification or through other means, our findings did not indicate that 
residents with prior convictions are more likely than other residents 
to commit abuse within these facilities. Absent such evidence, it may 
be more appropriate to focus on residents' behaviors versus their prior 
convictions when assessing the potential for committing abuse. Facility 
officials we interviewed more frequently expressed concerns about the 
behavior and potential for abuse by cognitively impaired and mentally 
ill residents than by offenders who may have no behavioral issues. 
Facilities already document problematic behaviors and assess the risk 
of individuals through resident assessments and care planning 
procedures, and when they accept residents with behavioral issues or 
such issues arise after admission, they must appropriately address 
those behaviors through the care planning for these individuals or 
transfer them to facilities better equipped to handle such residents. 
In addition, focusing on prior convictions alone can be problematic in 
that some offenders, such as those with certain physical impairments, 
likely do not pose a risk to other residents. Nonetheless, in the 
interest of identifying potential risks and taking precautionary 
measures, four states we reviewed--California, Illinois, Oklahoma, and 
Minnesota--enacted measures in 2005 to require notification to long- 
term care facilities when offenders are residents. Assessing their 
experiences as they implement these measures over time, including any 
negative impact on offenders' access to long-term care, may be 
instructive for other states with similar concerns.

While it was not part of our original objectives to fully evaluate the 
NSOR, it was our primary data source for identifying registered sex 
offenders residing in long-term care facilities. In the course of our 
analysis, we became aware that the FBI's NSOR, which links states' sex 
offender registration programs so that law enforcement agencies can 
identify sex offenders regardless of which state maintains their 
registration, was incomplete for the seven states we reviewed for this 
purpose. States face various barriers to fully submitting their 
registry records to the NSOR, including difficulties such as obtaining 
the required FBI number for each offender and a lack of staff 
resources. While the FBI has been reviewing issues related to states' 
submission of records to the NSOR, it currently does not track 
submission rates, so the proportion of state records missing from the 
NSOR is not precisely known. Continued improvements in the 
comprehensiveness of the NSOR can enhance the ability of local law 
enforcement agencies to identify offenders and notify the community, 
including long-term care facilities, where appropriate.

Recommendations for Executive Action:

We recommend that the Attorney General direct the FBI to take the 
following two actions:

* assess the completeness of the NSOR, including state submission 
rates, and:

* evaluate options for making it a more a comprehensive national 
database of registered sex offenders.

Agency and State Comments and Our Evaluation:

We provided copies of a draft of this report for comment to DOJ; HHS; 
and the eight states we reviewed: California, Florida, Illinois, 
Minnesota, New Jersey, Ohio, Oklahoma, and Utah. We received written 
responses from DOJ and HHS, which are included in this report as 
appendixes III and IV, respectively. We also received comments from 
California, Florida, Illinois, Minnesota, New Jersey, and Oklahoma. 
These agency and state comments and our evaluation follow.

DOJ commented that the recommendations are unnecessary because the FBI 
already performs assessments of the NSOR and explores options for 
improvement. For example, DOJ said that the FBI conducts triennial 
audits of states' NSOR participation, provides training and technical 
assistance to states, and seeks input from states about what assistance 
they need to improve their level of participation in the NSOR. DOJ 
characterized our evaluation as incomplete because we did not ask for 
information about the entire NSOR program or include a more extensive 
discussion in the draft report of their efforts to improve the NSOR. We 
obtained information about these efforts over the course of our work 
through interviews with FBI staff, documents available on their Web 
site, and through state officials. Because a comprehensive evaluation 
of the NSOR was not one of our reporting objectives, we did not include 
a complete listing of the FBI's assistance to states in our draft 
report. To respond to DOJ's comments, we revised the report to include 
additional information about the FBI's initiatives to assist states in 
data submission and to assess the accuracy of NSOR records. Including 
this additional information, however, does not alter our overall 
finding concerning the discrepancy between state sex offender 
registries and states' NSOR submissions.

We acknowledge, as DOJ pointed out in its comments, that there may be 
valid reasons for a certain amount of discrepancy between state 
registries and their NSOR submissions, such as if a state chooses not 
to submit the records of sex offenders still incarcerated since their 
whereabouts do not need to be tracked by the NSOR until their release. 
We also acknowledge the challenge states face in maintaining current 
and accurate information about registered sex offenders. However, we 
continue to believe that the intent of the recommendations remains 
valid because of the evidence we analyzed for a sample of states that a 
significant percentage of registered sex offender records are not being 
successfully submitted by some states to the NSOR, despite the states' 
and FBI's efforts to date. We believe the FBI needs to track state 
submission rates to the NSOR as a measure of comprehensiveness that can 
quantify the remaining gap as well as improvements over time. We 
therefore revised the first recommendation to specify that we are 
recommending that the FBI assess state submission rates as a means of 
assessing the completeness of NSOR.

DOJ commented on three additional issues:

* The risk posed by offenders residing in long-term care facilities. 
DOJ suggested that GAO discounted the risk posed by sex offenders 
residing in long-term care facilities based on insufficient evidence. 
We agree that the placement of a sex offender into a long-term care 
facility requires careful evaluation, particularly as the often-frail 
condition of long-term care residents makes them vulnerable to 
victimization. Based on our research and interviews with administrators 
of long-term care facilities, it is our view that the risk posed by 
offenders should be considered on a case-by-case basis. The presumption 
that offenders pose a threat to other residents could lead facilities 
to unnecessarily deny admission to low-risk offenders or unnecessarily 
seclude them from other residents. DOJ did not provide any new evidence 
to support its assertion that sex offenders pose a greater threat than 
the analysis we presented in the report.

* The likelihood that convicted sex offenders will commit additional 
sex offenses after their release from prison. DOJ objected to our 
citation of sex offender recidivism rates of 14 percent because they 
were based on only a 5-year post-incarceration period, saying the 
period was too short to be the basis of inferences about the likelihood 
that a sex offender will commit additional sex offenses, and because of 
evidence that sex offenses are underreported. We revised the report to 
clarify that the same research also cites 20-year sex offender 
recidivism rates of 27 percent.

* The usefulness of the NSOR in assisting law enforcement to identify 
sex offenders residing in long-term care facilities. DOJ questioned 
GAO's assertion that improvements in the comprehensiveness of the NSOR 
would improve the ability of local law enforcement to identify sex 
offenders residing in nursing homes, commenting that offenders would 
either already be on the state registry and thus identifiable or they 
would not be registered and therefore not included in the NSOR. We 
believe that a more comprehensive NSOR would improve the tracking of 
sex offenders who enter long-term care facilities in the same way it 
improves the tracking of sex offenders generally. If offenders are 
registered in one state but move to another state and fail to register, 
their records could be in the NSOR from the original state but not on 
the registry of the second state. A more comprehensive NSOR thus better 
ensures the national tracking of sex offenders who may choose to cross 
state lines.

HHS commented that this report brought to its attention the uncertainty 
that some long-term care facility officials have about the application 
of the HIPAA Privacy Rule to the disclosure of conviction information, 
as well as the issue that future guidance may be needed. HHS commented 
that the report will help to resolve the uncertainty about the HIPAA 
Privacy Rule, including clarifying that disclosures could be allowed 
for activities necessary for the safe operation of the facility or as 
required by state laws.

DOJ, HHS, and the states also provided technical comments, which we 
incorporated as appropriate.

As arranged with your offices, unless you publicly announce the 
contents of this report earlier, we plan no further distribution of it 
until 30 days after its issue date. At that time, we will send copies 
to the Attorney General, the Secretary of Health and Human Services, 
and other interested parties. We will also make copies available to 
others upon request. This report is also available at no charge on 
GAO's Web site at [Hyperlink, http://www.gao.gov].

If you or your staffs have any questions about this report, please 
contact me at (202) 512-7118 or allenk@gao.gov. Contact points for our 
Offices of Congressional Relations and Public Affairs may be found on 
the last page of this report. GAO staff who made major contributions to 
this report are listed in appendix V. 

Signed By:

Kathryn G. Allen: 
Director: 
Health Care: 

[End of section]

Appendix I: Scope and Methodology:

To determine the prevalence of registered sex offenders residing in 
long-term care facilities nationwide, we matched the addresses of 
registered sex offenders listed in the Federal Bureau of 
Investigation's (FBI) National Sex Offender Registry (NSOR) as of 
January 3, 2005, with the addresses of nursing homes and intermediate 
care facilities for people with mental retardation (ICF-MR) listed in 
the Centers for Medicare & Medicaid Services' (CMS) Online Survey, 
Certification and Reporting system (OSCAR) database. After 
standardizing address spellings and abbreviations, we used SAS, a 
statistical analysis program, to compare registered sex offender and 
long-term care facility addresses. Using a SAS function that quantifies 
the magnitude of difference between two text variables, we identified 
exact matches as well as near matches where the addresses differed 
slightly. We manually reviewed the addresses that differed slightly to 
determine if they were the same address.

To evaluate the comprehensiveness of the NSOR, we requested the full 
state sex offender registries from 8 states--California, Florida, 
Illinois, Minnesota, Ohio, Oklahoma, New Jersey, and Utah--in order to 
compare the number of records in each registry to the number of records 
in the NSOR for that state. We chose these 8 states on the basis of a 
number of criteria, including variation in geographic location and in 
the number of registered sex offenders identified as living in long- 
term care facilities based on our preliminary analyses. California 
state officials did not provide us with the state's sex offender 
registry in view of their concerns with state privacy laws. We also 
interviewed FBI staff about the management of the NSOR database. To 
obtain information about the administration and content of state 
registries, including their submission of records to the NSOR, we 
interviewed state registry administrators from the 8 states we reviewed 
and submitted a questionnaire via e-mail to the remaining 42 states, 
receiving responses from 20 of them.

Since no national data source on parolees that includes address 
information exists, we obtained parolee databases from each of the 
eight states we reviewed. We matched parolee addresses to nursing homes 
and ICFs-MR in OSCAR using the same methods we used for our analysis of 
NSOR and state sex offender registries.

We excluded some records from our analysis because there was no valid 
domestic address for the offender. Table 3 shows the number of records 
we analyzed from all data sources for both registered sex offenders and 
parolees, and the number of records excluded from each source because 
of missing, invalid, or otherwise unusable address information.

Table 3: Validity of Offender Address Data by Data Source:

Validity of offender address data : State abbreviation does not match 
one of 50 states or Washington, D.C; 
NSOR: Number of records: 27,141; 
NSOR: Share of total: 7%; 
Parolee data from eight reviewed states: Number of records: N/A; 
Parolee data from eight reviewed states: Share of total: N/A.

Validity of offender address data : Offender incarcerated or deported; 
NSOR: Number of records: 23,863; 
NSOR: Share of total: 6%; 
Parolee data from eight reviewed states: Number of records: 529; 
Parolee data from eight reviewed states: Share of total: 0%.

Validity of offender address data : Offender transient, homeless, or 
address unknown; 
NSOR: Number of records: 5,936; 
NSOR: Share of total: 2%; 
Parolee data from eight reviewed states: Number of records: 1,653; 
Parolee data from eight reviewed states: Share of total: 1%.

Validity of offender address data : Address listed is invalid; 
NSOR: Number of records: 8,775; 
NSOR: Share of total: 2%; 
Parolee data from eight reviewed states: Number of records: 11,879; 
Parolee data from eight reviewed states: Share of total: 5%.

Validity of offender address data : City/state does not match zip code; 
NSOR: Number of records: 327; 
NSOR: Share of total: 0%; 
Parolee data from eight reviewed states: Number of records: 12; 
Parolee data from eight reviewed states: Share of total: 0%.

Validity of offender address data : Valid address; 
NSOR: Number of records: 304,489; 
NSOR: Share of total: 82%; 
Parolee data from eight reviewed states: Number of records: 248,290; 
Parolee data from eight reviewed states: Share of total: 95%.

Validity of offender address data : All; 
NSOR: Number of records: 370,531; 
NSOR: Share of total: 100%; 
Parolee data from eight reviewed states: Number of records: 262,363; 
Parolee data from eight reviewed states: Share of total: 100%. 

Sources: GAO analysis of the NSOR, 2005; GAO analysis of parolee 
databases for eight reviewed states obtained March through September 
2005.

[End of table]

To obtain information about resident abuse perpetrated by registered 
sex offenders and parolees, we reviewed existing research and prior GAO 
reports. We also interviewed long-term care facility administrators in 
the eight states we reviewed, including administrators at facilities 
with registered sex offenders as residents, as well as state department 
of health and industry association officials and ombudsmen. To identify 
facilities for administrator interviews, we initially chose four long- 
term care facilities in each of the eight states we reviewed. These 
facilities were chosen from two groups of facilities based on our 
initial analysis of NSOR and OSCAR data. One group comprised facilities 
with registered sex offender matches and the other group did not have 
any such matches, and when possible, we selected two facilities from 
each grouping. If a selected facility refused our request for an 
interview, we selected another facility as a replacement from the same 
group. If a state did not have enough facilities with or without sex 
offenders to complete two interviews from each group of facilities, we 
used facilities from the other group. In all, we interviewed 
administrators at 29 long-term care facilities, 11 with registered sex 
offender matches and 18 without matches. We achieved a 91 percent 
response rate for the facility interviews.

To determine whether federal laws provide for notification of facility 
staff, residents, and residents' families when sex offenders or 
parolees live in long-term care facilities or for the supervision and 
separation of sex offenders and parolees living in these facilities, we 
reviewed federal laws and interviewed Department of Justice and CMS 
officials. We also interviewed Department of Health and Human Services 
Office for Civil Rights officials about the applicability of the Health 
Insurance Portability and Accountability Act of 1996 Privacy Rule to 
the notification of facilities about residents who are sex offenders or 
parolees.

To determine whether states we reviewed have laws or long-term care 
facilities have practices that provide for notification of these 
individuals and to determine the extent to which these individuals are 
subject to supervision and separation requirements that differ from 
those for other residents, we reviewed laws and interviewed state 
officials responsible for long-term care facility licensing, industry 
officials, long-term care ombudsmen, and the administrators at 29 long- 
term care facilities, which were chosen based on the criteria discussed 
above. We also interviewed Department of Corrections' officials 
regarding their efforts to inform facilities about their placement of 
parolees in them. To determine what information on sex offenders is 
available to the public, we also reviewed state sex offender Web site 
registries available in the states we reviewed.

The key sources used to identify registered sex offenders and parolees 
living in long-term care facilities included CMS's OSCAR database, the 
NSOR, and parolee databases from selected states. To assess the 
reliability of these data, we conducted electronic data testing, 
reviewed relevant documentation, and interviewed knowledgeable agency 
officials about the data quality control procedures. We determined that 
while the NSOR does not include all registered or convicted sex 
offenders, its records are regularly audited and are sufficiently 
reliable for the purposes of this report. The lack of comprehensiveness 
of the data was evaluated and taken into account in our discussion of 
the results. The OSCAR database and state parolee databases were also 
found to be sufficiently reliable for our purposes.

We conducted our work from September 2004 through February 2006 in 
accordance with generally accepted government auditing standards.

[End of section]

Appendix II: Registered Sex Offenders Living in Nursing Homes and ICFs- 
MR:

To determine the prevalence of registered sex offenders residing in 
long-term care facilities nationwide, we matched the addresses of 
registered sex offenders listed in the NSOR as of January 3, 2005, with 
the addresses of nursing homes and ICFs-MR listed in CMS's OSCAR 
database. Using this methodology we identified 683 registered sex 
offenders living in long-term care facilities. The number of registered 
sex offenders identified as residing in long-term care facilities in 
each state varied considerably, ranging from 0 to 144, as demonstrated 
in table 4.

Table 4: Registered Sex Offenders Identified as Living in Nursing Homes 
and ICFs-MR, by State, 2005:

State: Alabama; 
Nursing homes: 1; 
ICFs-MR: 0; 
Total: 1.

State: Alaska; 
Nursing homes: 3; 
ICFs-MR: 0; 
Total: 3.

State: Arizona; 
Nursing homes: 15; 
ICFs-MR: 0; 
Total: 15.

State: Arkansas; 
Nursing homes: 10; 
ICFs-MR: 1; 
Total: 11.

State: California; 
Nursing homes: 141; 
ICFs-MR: 3; 
Total: 144.

State: Colorado; 
Nursing homes: 9; 
ICFs-MR: 6; 
Total: 15.

State: Connecticut; 
Nursing homes: 9; 
ICFs-MR: 2; 
Total: 11.

State: Delaware; 
Nursing homes: 6; 
ICFs-MR: 0; 
Total: 6.

State: District of Columbia; 
Nursing homes: 0; 
ICFs-MR: 0; 
Total: 0.

State: Florida; 
Nursing homes: 31; 
ICFs-MR: 4; 
Total: 35.

State: Georgia; 
Nursing homes: 14; 
ICFs-MR: 0; 
Total: 14.

State: Hawaii; 
Nursing homes: 0; 
ICFs-MR: 0; 
Total: 0.

State: Idaho; 
Nursing homes: 5; 
ICFs-MR: 7; 
Total: 12.

State: Illinois; 
Nursing homes: 78; 
ICFs-MR: 2; 
Total: 80.

State: Indiana; 
Nursing homes: 3; 
ICFs-MR: 1; 
Total: 4.

State: Iowa; 
Nursing homes: 9; 
ICFs-MR: 5; 
Total: 14.

State: Kansas; 
Nursing homes: 9; 
ICFs-MR: 1; 
Total: 10.

State: Kentucky; 
Nursing homes: 5; 
ICFs-MR: 0; 
Total: 5.

State: Louisiana; 
Nursing homes: 11; 
ICFs-MR: 2; 
Total: 13.

State: Maine; 
Nursing homes: 3; 
ICFs-MR: 0; 
Total: 3.

State: Maryland; 
Nursing homes: 2; 
ICFs-MR: 0; 
Total: 2.

State: Massachusetts; 
Nursing homes: 0; 
ICFs-MR: 0; 
Total: 0.

State: Michigan; 
Nursing homes: 18; 
ICFs-MR: 1; 
Total: 19.

State: Minnesota; 
Nursing homes: 22; 
ICFs-MR: 3; 
Total: 25.

State: Mississippi; 
Nursing homes: 0; 
ICFs-MR: 0; 
Total: 0.

State: Missouri; 
Nursing homes: 19; 
ICFs-MR: 2; 
Total: 21.

State: Montana; 
Nursing homes: 1; 
ICFs-MR: 0; 
Total: 1.

State: Nebraska; 
Nursing homes: 0; 
ICFs-MR: 0; 
Total: 0.

State: Nevada; 
Nursing homes: 0; 
ICFs-MR: 0; 
Total: 0.

State: New Hampshire; 
Nursing homes: 4; 
ICFs-MR: 0; 
Total: 4.

State: New Jersey; 
Nursing homes: 12; 
ICFs-MR: 0; 
Total: 12.

State: New Mexico; 
Nursing homes: 13; 
ICFs-MR: 2; 
Total: 15.

State: New York; 
Nursing homes: 10; 
ICFs-MR: 23; 
Total: 33.

State: North Carolina; 
Nursing homes: 10; 
ICFs-MR: 0; 
Total: 10.

State: North Dakota; 
Nursing homes: 0; 
ICFs-MR: 0; 
Total: 0.

State: Ohio; 
Nursing homes: 3; 
ICFs-MR: 0; 
Total: 3.

State: Oklahoma; 
Nursing homes: 10; 
ICFs-MR: 1; 
Total: 11.

State: Oregon; 
Nursing homes: 6; 
ICFs-MR: 0; 
Total: 6.

State: Pennsylvania; 
Nursing homes: 0; 
ICFs-MR: 0; 
Total: 0.

State: Rhode Island; 
Nursing homes: 1; 
ICFs-MR: 0; 
Total: 1.

State: South Carolina; 
Nursing homes: 1; 
ICFs-MR: 3; 
Total: 4.

State: South Dakota; 
Nursing homes: 11; 
ICFs-MR: 0; 
Total: 11.

State: Tennessee; 
Nursing homes: 1; 
ICFs-MR: 0; 
Total: 1.

State: Texas; 
Nursing homes: 61; 
ICFs-MR: 8; 
Total: 69.

State: Utah; 
Nursing homes: 0; 
ICFs-MR: 0; 
Total: 0.

State: Vermont; 
Nursing homes: 0; 
ICFs-MR: 0; 
Total: 0.

State: Virginia; 
Nursing homes: 7; 
ICFs-MR: 2; 
Total: 9.

State: Washington; 
Nursing homes: 8; 
ICFs-MR: 0; 
Total: 8.

State: West Virginia; 
Nursing homes: 1; 
ICFs-MR: 0; 
Total: 1.

State: Wisconsin; 
Nursing homes: 20; 
ICFs-MR: 1; 
Total: 21.

State: Wyoming; 
Nursing homes: 0; 
ICFs-MR: 0; 
Total: 0.

Total: Nursing homes: 603; 
ICFs-MR: 80; 
Total: 683. 

Sources: GAO analysis of the FBI's NSOR as of January 3, 2005; CMS's 
OSCAR database, 2004.

[End of table]

[End of section]

Appendix III: Comments from the Department of Justice: 

U.S. Department of Justice:

Washington, D.C. 20530:

March 9, 2006:

Ms. Kathryn G. Allen: 
Director:
Health Care:
United States Government Accountability Office: 
Washington, D.C. 20548:

Re: GAO REPORT 06-326:

Dear Ms. Allen:

Thank you for the opportunity to review the Government Accountability 
Office (GAO) draft report entitled "LONG-TERM CARE FACILITIES: 
Information on Residents Who Are Registered Sex Offenders or Paroled 
for Other Crimes." The draft report has been reviewed by various 
components of the Department of Justice (Department), including the 
Federal Bureau of Investigation (FBI) and its Criminal Justice 
Information Services (CJIS) Division. This letter serves as the formal 
comments of the Department, and it is requested that this letter be 
included by the GAO in its final report.

The Department set out its comments in two sections. The first section 
addresses the recommendation relating to the National Sex Offender 
Registry (NSOR). The second section discusses the treatment of issues 
of sex offender recidivism and potential danger.

Support for the Recommendation and the Completeness of Information 
About the NSOR:

The draft GAO report includes a two part recommendation that "the 
Attorney General direct the FBI to take the following two actions: (1) 
assess the completeness of the NSOR and (2) evaluate options for making 
it a more comprehensive national database of registered sex offenders." 
[Numbering added for clarification.]

We believe the recommendation is unnecessary. Had the GAO conducted a 
comprehensive review of the NSOR program it most likely would had 
discovered that the FBI already performs assessments and explores 
options to improve the NSOR.

The CJIS Division regularly assesses the competencies of NSOR and 
suggests improvements for its operations that would increase the 
registry's value to the end users. For example, during fiscal year 
2005, CJIS assessed the states' level of participation in the NSOR and 
requested specific information as to what assistance was needed to 
improve their level of participation. While the majority of the states 
responded to the request for information, a side-by-side comparison of 
record counts (state registries against the NSOR, the methodology of 
the GAO) may not accurately identify the level of participation. For 
example, many states chose not to include all of their records in the 
NSOR, such as those regarding incarcerated offenders and those included 
in the NSOR by new state of registry, stating they would be of no value 
at the national level. As such, when these states responded to NSOR 
they included only those released and registered. Consequently, it is 
understandable that a discrepancy exists between the states' numbers 
and those in NSOR. Further, 11 states indicated that NSOR's database 
requirements limited the states' ability to include all of their sex 
offenders in the NSOR. CJIS has been advised by these states that they 
need additional funding for training and personnel before they can 
improve their level of participation.

In addition to the fiscal 2005 audit, the CJIS Division conducts 
triennial audits of the states' participation in the NSOR. These audits 
include evaluations of the accuracy, completeness, and timeliness of 
the information in the NSOR.

Regarding the second part of the recommendation, CJIS already evaluates 
options for making NSOR a more comprehensive national database. The 
CJIS Division operates under a shared management concept and an 
Advisory Policy Board (APB) that is formed under the Federal Advisory 
Committee Act. The CJIS APB makes recommendations regarding general 
policy to the FBI Director with respect to CJIS philosophy, concept, 
and operational principles. Further, the CJIS APB has created an ad hoc 
task force to address the states' concerns with the NSOR. As such, the 
CJIS APB already provides a forum to discuss and develop any changes to 
the NSOR and all 50 states have representation in this APB process.

The GAO may wish to undertake additional information gathering and 
analyses of the entire NSOR program. The GAO draft report reaches 
conclusions and makes findings that reflect an imperfect understanding 
of the NSOR program. This may be the case because the GAO based its 
findings on information that could have been far more complete. The GAO 
obtained information from select state governments and from the CJIS 
Division. There is nothing in the report to suggest that the GAO 
collected and evaluated additional information. While the Department 
cannot be sure what information the states provided, the Department 
knows that the CJIS Division supplied the GAO with legislative 
background, database records from the NSOR, and information regarding 
technical functionality of the data, and nothing more. This is all the 
GAO requested. The GAO did not ask for information about the entire 
NSOR program. The Department believes the GAO failed to collect enough 
data to allow it to prepare a thorough analysis of the NSOR program 
and, therefore, an accurate report.

For example, the GAO proceeded without asking for information about the 
relationship between the CJIS Division and the state sex offender 
registries. This may explain why the draft is silent about the training 
and technical assistance that the FBI provides to the states for the 
purpose of helping them include their information in the NSOR. The 
absence of any discussion about the assistance provided by the 
Department suggests that the Department has failed to provide such 
assistance. In fact, such assistance is provided. Moreover, as a result 
of working with the states to facilitate their data reporting, the 
Department has developed a wealth of information about NSOR. The 
Department would have given such information to the GAO had it asked.

Sex Offenders, Recidivism, and Potential Danger:

The report acknowledges an inability to obtain any definite information 
about the likelihood that registered sex offenders or other convicts 
will engage in abuse of other residents in long-term care facilities, 
because the available information about incidents of abuse does not 
show whether the perpetrators have prior criminal convictions. See 
"Highlights" summary; pp. 5, 11, 17. Nevertheless, the discussion and 
conclusions in the report seem to discount on inferential grounds any 
risk such offenders may pose to other residents. The report also 
attempts to tie in its findings and recommendations concerning the NSOR 
to the original subject of the report by stating that "improvements in 
the comprehensiveness of the NSOR can enhance the ability of local law 
enforcement agencies to identify offenders and notify the community, 
including long-term care facilities, where appropriate." Page 28. 
Neither conclusion is warranted by the information in the report.

Regarding the suggested low-risk status of convicted sex offenders, the 
report relies in part on empirical study which it sees as suggesting 
that convicted sex offenders "typically" do not commit more sex 
offenses.

By way of illustration, a study is cited in which it was estimated that 
"about 14 percent [of sex offenders] had a new sex offense charge or 
conviction within 5 years of their release from prison." Page 10. 
Whether a 14 percent re-offense rate could properly be regarded as a 
small concern in relation to offenses as serious as sexual assaults is 
certainly a debatable matter and, in any event, the information 
provided by studies of this type is limited by the duration of the 
follow-up period for which recidivism is reckoned - in this case, 5 
years. For example, assume for the sake of discussion that a sex 
offender lives on average about 35 years following his release from 
prison. That would be 7 times as long as the 5 year follow-up period 
considered in the study. Multiplying the study's recidivism figure of 
14 percent over a 5-year period by 7 would then yield a lifetime 
recidivism rate for sex offenders of virtually 100 percent (7 times 14 
percent equals 98 percent). Such an inference would, of course, be 
invalid because (among other reasons) the likelihood of sex offenders 
committing more sex offenses later in their lives may not be the same 
as their likelihood of doing so within a few years of their release 
from imprisonment. But the example illustrates the hazards of making 
inferences about the overall risk of recidivism based on recidivism 
figures reckoned for relatively short periods. In inferring from the 
cited data that sex offenders "typically" do not reoffend, the report 
implicitly makes a contrary assumption - that a sex offender who does 
not commit more sex offenses within a few years of release probably 
will never do so. The report does not substantiate this assumption.

A second limitation of the cited data is that, as the report itself 
notes, "research also indicates that sex offenses are underreported." 
Page 10. Reliance on re-arrest figures shows only the number of cases 
in which the offenses committed by sex offenders are reported or 
otherwise detected by the authorities, and the case is successfully 
investigated, resulting in the identification of the offender and his 
being charged with the crime. Researchers have attempted to go beyond 
this type of incomplete information through such methodologies as self-
reports by sex offenders under assurances of confidentiality, and 
comparison of reported crime figures with data from victimization 
surveys. Studies of this type indicate that a large proportion of sex 
offenses never come to light, or do not result in the arrest of the 
offender. There is a comparable problem in the draft report's inference 
of a relatively low recidivism rate for sex offenders without any 
serious effort to determine how much of their actual recidivism is not 
reflected in re-arrest figures and the like.

The report also downplays the risk convicted offenders may pose to 
other residents of long-term care facilities by noting that such 
offenders may be incapable of committing crimes because of physical 
infirmity or impairment. Pages 17-18, 27. This notion may be undercut 
to some extent by the report's observation that registered sex 
offenders identified as living in nursing homes "were considerably 
younger than the general nursing home population, with 57 percent under 
age 65 compared to about 10 percent of all nursing home residents." 
Pages 4-5; but cf. p. 13 (discussing possible reason). In any event, in 
assessing the risk of victimization, the condition of potential victims 
- not just the condition of potential offenders - needs to be taken 
into account. The resident population in long-term care facilities must 
include many individuals who are especially vulnerable to 
victimization, and who may be attractive targets for sexually violent 
criminals present in such facilities, because of mental and/or physical 
frailty or impairment. Hence, any reduced risk resulting from infirmity 
of potential offenders must be balanced against any increased risk 
resulting from infirmity of potential victims.

A final point made by the report on this question is that the facility 
administrators who were interviewed more frequently expressed concern 
about residents with presently observable cognitive impairments or 
mental illness (such as dementia), as opposed to residents with past 
criminal convictions. But this information as well would not support 
reliable inferences about the actual risk posed by residents with 
criminal histories. One reason such residents may not rank high among 
administrators' concerns is that the administrators do not know that 
they have such residents. As the report recounts, facility 
administrators may not have this information because the state in which 
the facility is located does not give notice concerning registered sex 
offenders to long-term care facilities, or because the state notifies 
such facilities only concerning a subclass of sex offenders, or because 
the sex offenders in such facilities are not registered since their 
convictions occurred prior to the states' establishment of sex offender 
registration programs. Pages 2, 6-7, 13, 19-22. Also, even if there is 
awareness of the presence of such offenders in a facility, an 
administrator may be unable to judge to what extent the conviction 
entails a present risk to others, and it would be understandable if 
such residents were accordingly viewed as a lesser concern than 
residents with manifest current conditions that dispose them to violent 
behavior. But neither of these reasons would imply anything about the 
actual degree of danger to other residents posed by the presence of 
such offenders in long-term care facilities, or that they are probably 
not enough of a danger to worry about much, as the report seems to 
suggest. See page 27. The report turned up no information sufficient to 
support policy conclusions on this issue, and it would be more 
appropriate to acknowledge this point straightforwardly.

Finally, there is the report's assertion that "[c]ontinued improvements 
in the comprehensiveness of the NSOR can enhance the ability of local 
law enforcement agencies to identify offenders and notify the 
community, including long-term care facilities, where appropriate." 
Without questioning the importance of ensuring that the information in 
NSOR is as comprehensive as possible, it is difficult to see how that 
could have the particular significance asserted in this statement. If a 
sex offender residing in a long-terns care facility registers as 
required with the state in which the facility is located, then local 
law enforcement agencies (or other agencies responsible for sex 
offender notification) have access to that information through the 
state system, and it makes no difference for this purpose whether the 
information is also in NSOR. The other possibility is that a sex 
offender in such a facility will fail to register his current address 
with the state, in which case NSOR cannot have the information either, 
because the information in NSOR comes from the state systems. So the 
report should either explain the logical justification for this 
statement, or delete it.

Thank you for the opportunity to comment on your report.

Sincerely yours, 

Signed By:

Paul R. Corts:
Assistant Attorney General for Administration:

[End of section]

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

Department Of Health & Human Services: 
Office of Inspector General:
Washington, D.C. 20201:

Ms. Kathryn G. Allen: 
Director: 
Health Care: 
U.S. Government Accountability Office: 
Washington, DC 20548:

Dear Ms. Allen:

Enclosed are the Department's comments on the U.S. Government 
Accountability Office's (GAO) draft report entitled, "Long-Term Care 
Facilities: Information on Residents Who Arc Registered Sex Offenders 
or Paroled for Other Crimes" (GAO-06-326). These comments represent the 
tentative position of the Department and are subject to reevaluation 
when the final version of this report is received.

The Department provided several technical comments directly to your 
staff.

The Department appreciates the opportunity to comment on this draft 
report before its publication.

Sincerely, 

Signed By:

Daniel R. Levinson: 
Inspector General:

Enclosure:

Comments Of The Department Of Health And Human Services On The U.S. 
Government Accountability Office's Draft Report Entitled. "Long-Term 
Care Facilities Information On Residents Who Are Registered Sex 
Offenders Or Paroled For Other Crimes" (GAO-06-326):

The Department of Health and Human Services (HHS) appreciates the 
opportunity to comment on the draft report. The comments that follow 
represent HHS's responses to the draft report. We note that the draft 
GAO Report contains no conclusions or recommendations concerning the 
Health Insurance Portability and Accountability Act of 1996 (HlPAA) 
Privacy Rule.

HHS appreciates GAO's bringing to our attention the uncertainty of some 
long-term care facility officials with respect to the application of 
the HIPAA Privacy Rule in the uncommon circumstances outlined in the 
proposed Report. We also appreciate GAO's suggestion that future 
guidance may be needed should this issue gain in prominence.

We believe the Report will contribute to resolving much of the 
uncertainty based on the discussion on page 25 of the relevant Privacy 
Rule provisions, including that, even to the extent conviction 
information is maintained by the long-term care facility as protected 
health information, it could still be used within the facility or 
disclosed to others for specific purposes permitted by the HIPAA 
Privacy Rule. This includes activities necessary for the safe operation 
of the facility or disclosures that are required by other law, such as 
certain of the State laws referenced elsewhere in the draft Report.

However, in the Results in Brief section of the proposed Report, on 
page 6, the statement that some long-term care facility managers would 
notify facility staff of a resident's prior conviction "[d]espite 
concerns that they may violate the HIPAA Privacy Rule" may wrongfully 
suggest that these disclosures would, in all cases, violate the HIPAA 
Privacy Rule. Such a misinterpretation could be avoided by including in 
the sentence that follows a reference to the explanation of the 
application of the HIPAA Privacy Rule that appears later on page 25.

[End of section]

Appendix V: GAO Contact and Staff Acknowledgments:

GAO Contact:

Kathryn G. Allen, (202) 512-7118 or allenk@gao.gov:

Acknowledgments:

In addition to the contact named above, Susan T. Anthony, Assistant 
Director; George Bogart; Katherine Crumley; Michaela M. Monaghan; 
Elizabeth T. Morrison; Sari B. Shuman; and Kara Sokol made key 
contributions to this report. 

(290416): 

[End of section]

FOOTNOTES:

[1] For this report, we refer to nursing homes and ICFs-MR together as 
long-term care facilities.

[2] For this report, we refer to registered sex offenders and other 
offenders on parole for non-sex offenses as offenders. 

[3] California state officials did not provide us with the state's sex 
offender registry in view of their concerns with state privacy laws. 

[4] The NNHS is conducted by the Centers for Disease Control and 
Prevention's National Center for Health Statistics.

[5] Of the 1.5 million individuals living in these long-term care 
facilities in 2005, about 100,000 lived in ICFs-MR.

[6] Adaptive skills include communication, self-care, home living, and 
social skills. The criteria specify that these conditions need to have 
been identified at or before the age of 18. 

[7] CMS sets conditions of participation for facilities that receive 
federal funding. Part of the conditions of participation requires that 
residents have certain rights to personal privacy and the 
confidentiality of personal records.

[8] 45 C.F.R. pts. 160 and 164 (2005).

[9] Rape is defined as forced sexual intercourse with a male or female 
victim. Sexual assault is defined as a variety of victimizations that 
involve unwanted sexual contact. Incest is defined as nonforcible 
sexual intercourse between persons who are related to each other to a 
degree where marriage is prohibited by law. Department of Justice, 
Bureau of Justice Statistics, An Analysis of Data on Rape and Sexual 
Assault: Sex Offenses and Offenders (Washington, D.C.: February 1997).

[10] For this report, we refer to the Jacob Wetterling Crimes Against 
Children and Sex Offender Registration Program as the Wetterling 
Program.

[11] Violent Crime and Law Enforcement Act of 1994, Pub. L. No. 103- 
322, § 170101, 108 Stat. 1796, 2038 (1994) (codified at 42 U.S.C. § 
14071).

[12] Criminal offenses against minors include criminal sexual conduct 
toward a minor and solicitation of a minor to engage in sexual conduct. 
"Sexually violent offenses" include offenses that consist of aggravated 
sexual abuse or sexual abuse. A "sexually violent predator" is defined 
as a person who has been convicted of a sexually violent offense and 
who suffers from a mental abnormality or personality disorder that 
makes the person likely to engage in predatory sexually violent 
offenses. 42 U.S.C. § 14071(a)(3). 

[13] Under the Byrne Formula Grants Program, the DOJ's Bureau of 
Justice Assistance provides federal financial assistance to grantees 
for the purpose of enforcing state and local laws that establish 
offenses similar to offenses established under the Controlled 
Substances Act and to improve the functioning of the criminal justice 
system with emphasis on violent crime and serious offenders. In fiscal 
year 2004, all 50 states, the District of Columbia, and the U.S. 
territories received financial assistance through Byrne Grants.

[14] Pub. L. No. 104-145, § 2, 110 stat. 1345 (1996) (codified at 42 
U.S.C. § 14071).

[15] For this report, we refer to the Pam Lychner Sexual Offender 
Tracking and Identification Act of 1996 as the Lychner Act.

[16] Pub. L. No. 104-236, § 2, 110 Stat. 3093 (1996) (codified at 42 
U.S.C. § 14072). The NSOR is a nationwide system that links the states' 
sex offender registration and notification programs. The system uses a 
person's FBI number to connect the registration information in the 
National Criminal Information Center with the registrant's criminal 
history information that includes his/her fingerprints. According to 
the DOJ the NSOR should be used to enhance a state's ability to locate 
offenders in its jurisdiction who may be violating the law by not 
registering. 

[17] States could apply for a 2-year extension of the statutory 
deadline from the DOJ if they had made good faith efforts to comply but 
were unable to meet the original deadline for implementing a state sex 
offender registry.

[18] Scott Matson and Roxanne Lieb, Sex Offender Registration: A Review 
of State Laws (Olympia, Wash.: Washington State Institute for Public 
Policy, 1996). Department of Justice, Bureau of Justice Statistics, 
Summary of State Sex Offender Registries, 2001 (Washington, D.C.: 
2002). 

[19] An FBI official explained that the submission of one record 
demonstrates that the state has completed the reprogramming of its 
database to conform to NSOR standards.

[20] Under the Lychner Act, NSOR information must also be disclosed for 
employment background checks. To ensure that information released in 
background checks is accurate, a person's identification is verified 
using fingerprints. The FBI number provides the necessary link between 
the sex offender registry record and an offender's fingerprint records 
to technically achieve the inclusion of an offender's NSOR records in 
employment background checks.

[21] States also have the option of searching for an offender's FBI 
number by conducting inquiries of the FBI's Fingerprint Identification 
Record System using name and date of birth or fingerprints.

[22] According to NSOR data documentation provided by the FBI certain 
pieces of information are mandatory for state registry records to be 
accepted into the NSOR, including: offender's name; physical 
characteristics, including gender, height, weight, eye, and hair color; 
race; date of birth; registration beginning and ending dates; FBI 
number; and conviction information. The NSOR also includes other 
information, such as vehicle license plate numbers and home address, 
but will accept records even if this information is not provided.

[23] In addition to annual address verification, states are required to 
validate information in the NSOR on an annual basis to ensure the 
accuracy and completeness of the information. 

[24] 42 C.F.R. §§ 483.13(c)(2), 483.420(d)(2).

[25] The federal law requires the registration of sex offenders 
convicted of criminal offenses against minors or of sexually violent 
offenses or those designated as sexually violent predators.

[26] 730 ILCS Stat. Ann. § 152/120 (West 2005); Utah Code Ann. § 77-27- 
21.5 (2003).

[27] Cal. Penal Code § 290.45 (Dearing 2005); Fla. Stat. Ann. § 944.606 
(West 2005); Minn. Stat. Ann. § 244.052 (West 2003); N.J. Stat. Ann. § 
2C: 7-8 (West 2005); Ohio Rev. Code Ann. §§ 2950.081, 2950.11 (Anderson 
2005); Okla. Stat. Ann. Tit. 57, § 584 (West 2005-2006).

[28] 2005 Cal. Adv. Legis. Serv. c. 466 (Dearing); 2005 Ill. Legis. 
Serv. 94-163 (West); 2005 Minn. Laws c. 243.166; 2005 Okla. Sess. Laws 
Serv. c. 465 (West).

[29] Requirements in state community notification laws specifying that 
nursing homes and ICFs-MR be notified about registered sex offenders 
who were residents appear to be a recent trend. For example, a 2001 
review of state community notification laws by the Bureau of Justice 
Statistics found that states generally did not notify nursing homes or 
ICFs-MR when offenders entered the facilities.

[30] Emergency rules implementing this law require licensed long-term 
care facilities, such as nursing homes and ICFs-MR, to check the 
background of potential residents through the state sex offender 
database. 77 Ill. Reg. § 300.625 (as added for emergency rules 
published on Sept. 2, 2005). These rules expired on December 7, 2005, 
and have not yet been replaced by permanent rules.

[31] The Prosecutorial Remedies and Other Tools to End the Exploitation 
of Children Today Act of 2003, Pub. L. No. 108-21 § 604, 117 Stat. 650, 
688.

[32] DOJ has published proposed guidelines in the Federal Register with 
respect to state Internet sites for sex offender information at 70 Fed. 
Reg. 12721 (2005).

[33] See [Hyperlink, http://www.nsopr.gov/].

[34] 2005 Ill. Legis. Serv. 94-163 (West); 2005 Minn. Laws c. 243.166; 
2005 Okla. Sess. Laws Serv. c. 465 (West).

[35] 77 Ill. Reg. §§ 300.615, 300.625 (as added by emergency rules 
published Sept. 2, 2005). These rules expired on December 7, 2005, and 
have not yet been replaced by permanent rules.

GAO's Mission:

The Government Accountability Office, the investigative arm of 
Congress, exists to support Congress in meeting its constitutional 
responsibilities and to help improve the performance and accountability 
of the federal government for the American people. GAO examines the use 
of public funds; evaluates federal programs and policies; and provides 
analyses, recommendations, and other assistance to help Congress make 
informed oversight, policy, and funding decisions. GAO's commitment to 
good government is reflected in its core values of accountability, 
integrity, and reliability.

Obtaining Copies of GAO Reports and Testimony:

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through the Internet. GAO's Web site ( www.gao.gov ) contains 
abstracts and full-text files of current reports and testimony and an 
expanding archive of older products. The Web site features a search 
engine to help you locate documents using key words and phrases. You 
can print these documents in their entirety, including charts and other 
graphics.

Each day, GAO issues a list of newly released reports, testimony, and 
correspondence. GAO posts this list, known as "Today's Reports," on its 
Web site daily. The list contains links to the full-text document 
files. To have GAO e-mail this list to you every afternoon, go to 
www.gao.gov and select "Subscribe to e-mail alerts" under the "Order 
GAO Products" heading.

Order by Mail or Phone:

The first copy of each printed report is free. Additional copies are $2 
each. A check or money order should be made out to the Superintendent 
of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or 
more copies mailed to a single address are discounted 25 percent. 
Orders should be sent to:

U.S. Government Accountability Office

441 G Street NW, Room LM

Washington, D.C. 20548:

To order by Phone:

Voice: (202) 512-6000:

TDD: (202) 512-2537:

Fax: (202) 512-6061:

To Report Fraud, Waste, and Abuse in Federal Programs:

Contact:

Web site: www.gao.gov/fraudnet/fraudnet.htm

E-mail: fraudnet@gao.gov

Automated answering system: (800) 424-5454 or (202) 512-7470:

Public Affairs:

Jeff Nelligan, managing director,

NelliganJ@gao.gov

(202) 512-4800

U.S. Government Accountability Office,

441 G Street NW, Room 7149

Washington, D.C. 20548: