This is the accessible text file for GAO report number GAO-07-241 
entitled 'Nursing Homes: Efforts to Strengthen Federal Enforcement Have 
Not Deterred Some Homes from Repeatedly Harming Residents' which was 
released on April 23, 2007. 

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. 

Report to the Ranking Minority Member, Committee on Finance, U.S. 
Senate: 

United States Government Accountability Office: 

GAO: 

March 2007: 

Nursing Homes: 

Efforts to Strengthen Federal Enforcement Have Not Deterred Some Homes 
from Repeatedly Harming Residents: 

GAO-07-241: 

GAO Highlights: 

Highlights of GAO-07-241, a report to the Ranking Minority Member, 
Committee on Finance, U.S. Senate 

Why GAO Did This Study: 

In 1998 and 1999 reports, GAO concluded that enforcement actions, known 
as sanctions, were ineffective in encouraging nursing homes to maintain 
compliance with federal quality requirements: sanctions were often 
rescinded before being implemented because homes had a grace period to 
correct deficiencies. In response, the Centers for Medicare & Medicaid 
Services (CMS) began requiring immediate sanctions for homes that 
repeatedly harmed residents. Using CMS enforcement and deficiency data, 
GAO (1) analyzed federal sanctions from fiscal years 2000 through 2005 
against 63 homes previously reviewed and (2) assessed CMS’s overall 
management of enforcement. The 63 homes had a history of harming 
residents and were located in 4 states that account for about 22 
percent of homes nationwide. 

What GAO Found: 

From fiscal years 2000 through 2005, the number of sanctions decreased 
for the 63 nursing homes GAO reviewed that had a history of serious 
quality problems, a decline consistent with nationwide trends. While 
the decline may reflect improved quality or changes to enforcement 
policy, it may also mask survey weaknesses that understate quality 
problems, an issue GAO has reported on since 1998. Although the number 
of sanctions decreased, the homes generally were cited for more 
deficiencies that caused harm to residents than other homes in their 
states. Almost half of the homes reviewed continued to cycle in and out 
of compliance; 19 did so 4 times or more. These homes temporarily 
corrected deficiencies and, even with sanctions, were again found out 
of compliance on subsequent surveys. Several weaknesses appeared to 
undermine the effectiveness of the sanctions implemented against the 
homes reviewed. First, civil money penalties (CMP), which by statute 
are not paid while under appeal—a process that can take years—were 
generally imposed at the lower end of the allowable dollar range. For 
example, the median per day CMP ranged from $350 to $500, significantly 
below the maximum of $3,000 per day. Second, CMS favored the use of 
sanctions that give homes more time to correct deficiencies, increasing 
the likelihood that the sanctions would not be implemented. Thus, more 
than half of the denial of payment for new admissions (DPNA) that CMS 
imposed were the type that give homes 3 months to correct deficiencies 
rather than those that only give homes up to 15 days. Third, there was 
no record of a sanction for about 22 percent of the homes reviewed that 
met CMS’s criteria for immediate sanctions, a problem GAO also 
identified in 2003; moreover, 60 percent of DPNAs imposed as immediate 
sanctions were not implemented until 1 to 2 months after citation of 
the deficiency. Finally, involuntary termination of homes from 
participating in the Medicare or Medicaid programs was rare because of 
concerns about access to other nearby homes and resident transfer 
trauma; 2 of the 63 homes reviewed were involuntarily terminated 
because of quality problems. 

CMS’s management of enforcement is hampered by the complexity of its 
immediate sanctions policy and by its fragmented and incomplete data. 
Its policy allows some homes with the worst compliance histories to 
escape immediate sanctions. For example, a home cited with a serious 
deficiency and that has not yet corrected an earlier serious deficiency 
is spared an immediate sanction. Such rules may in part explain why the 
63 homes reviewed only had 69 instances of immediate sanctions over a 6-
year period despite being cited 444 times for deficiencies that harmed 
residents. Although CMS initiated development of a new enforcement data 
system 6 years ago, it is fragmented and has incomplete national 
reporting capabilities. CMS is taking additional steps to improve 
nursing home enforcement, such as developing guidance to encourage more 
consistency in CMP amounts, but it is not clear whether and when these 
initiatives will address the enforcement weaknesses GAO found. 

What GAO Recommends: 

GAO recommends that the CMS Administrator (1) develop an administrative 
process for collecting civil money penalties more expeditiously and 
seek legislation to implement this process effectively, as appropriate; 
(2) strengthen its immediate sanctions policy; (3) expand its oversight 
of homes with a history of harming residents; and (4) improve the 
effectiveness of its enforcement data systems. CMS generally concurred 
with GAO’s recommendations. 

[Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-07-241]. 

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: 

Number of Sanctions Has Decreased: 

Despite Changes in Federal Enforcement Policy, Many Homes Continued to 
Cycle In and Out of Compliance: 

Complex Immediate Sanctions Policy and Data Limitations Hamper CMS 
Management of Enforcement: 

Conclusions: 

Recommendations for Executive Action: 

Agency and State Comments and Our Evaluation: 

Appendix I: Scope and Methodology: 

Appendix II: Percentage of Nursing Homes Cited for Actual Harm or 
Immediate Jeopardy, by State, Fiscal Years 2000-2005: 

Appendix III: Federal Sanctions for Nursing Homes Reviewed, by State, 
Fiscal Years 2000-2005: 

Appendix IV: Examples of Homes Reviewed That Frequently Cycled In and 
Out of Compliance: 

Appendix V: Number of Days between Survey and Implementation Date of 
DPNA for Homes Reviewed, Fiscal Years 2000-2005: 

Appendix VI: Comments from the Centers for Medicare & Medicaid 
Services: 

Appendix VII: Comments from the California Department of Health 
Services: 

Appendix VIII: Comments from the Michigan Department of Community 
Health: 

Appendix IX: Comments from the Texas Department of Aging and Disability 
Services: 

Appendix X: GAO Contact and Staff Acknowledgments: 

Related GAO Products: 

Tables: 

Table 1: Number of Nursing Homes Reviewed in 1999 That Were Included in 
Our Analysis for This Report: 

Table 2: Scope and Severity of Deficiencies Identified during Nursing 
Home Surveys: 

Table 3: Sanctions Available to Encourage Nursing Home Compliance with 
Requirements: 

Table 4: Sanctions Implemented for Homes Reviewed, Fiscal Years 2000- 
2002 and 2003-2005: 

Table 5: Examples of Homes with Low Implemented CMPs: 

Table 6: Example of a Michigan Nursing Home That Frequently Cycled In 
and Out of Compliance and Was Still Open as of November 2006: 

Table 7: Examples of Homes' Deficiency Histories and Termination 
Actions, Fiscal Years 2000-2005: 

Table 8: Number of Nursing Homes Reviewed in 1999 That Were Included in 
Our Analysis for This Report, by State: 

Table 9: Percentage of Nursing Homes Cited for Actual Harm or Immediate 
Jeopardy during Standard Surveys, Fiscal Years 2000-2005: 

Table 10: Number of Sanctions Implemented Among Homes We Reviewed, 
Fiscal Years 2000-2005: 

Table 11: Examples of Homes that Frequently Cycled In and Out of 
Compliance: 

Figures: 

Figure 1: Federal-State Responsibilities in the Enforcement Process: 

Figure 2: Percentage of Implemented Sanctions for Homes Reviewed Over 
Three Time Periods (July 1995-October 1998, Fiscal Years 2000-2002, and 
Fiscal Years 2003-2005): 

Figure 3: Frequency that Reviewed Homes Cycled In and Out of 
Compliance, Fiscal Years 2000-2005: 

Figure 4: Number of Days between Survey and Implementation of CMPs and 
DPNAs among Homes Reviewed, Fiscal Years 2000-2005: 

Figure 5: Lag Time between Survey and CMP Payment for a Michigan 
Nursing Home: 

Figure 6: Number of Homes with One or More Double Gs, Fiscal Years 2000-
2005: 

Figure 7: Impact of Intervening Periods of Compliance Rule on Immediate 
Sanctions for One Pennsylvania Nursing Home, 2000: 

Figure 8: Impact of Clearing Effect Rule on Immediate Sanctions for One 
Michigan Nursing Home, 2000-2002: 

Abbreviations: 

ACTS: ASPEN Complaints/Incidents Tracking System: 
AEMA: SPEN Enforcement Manager: 
ASPEN: Automated Survey Processing Environment: 
CMP: civil money penalty: 
CMPTS: CMP Tracking System: 
CMS: Centers for Medicare & Medicaid Services: 
DPNA: denial of payment for new admissions: 
LTC: Long Term Care Enforcement Tracking System: 
OBRA 87: Omnibus Budget Reconciliation Act of 1987: 
OSCAR: On-Line Survey, Certification, and Reporting system: 
PDQ: Providing Data Quickly: 
QIO: Quality Improvement Organization: 

United States Government Accountability Office: 
Washington, DC 20548: 

March 26, 2007: 

The Honorable Charles E. Grassley: 
Ranking Minority Member: 
Committee on Finance: 
United States Senate: 

Dear Senator Grassley: 

The nation's 1.5 million nursing home residents are a highly vulnerable 
population of elderly and disabled individuals for whom remaining at 
home is no longer feasible. The federal government plays a key role in 
ensuring that nursing home residents receive appropriate care by 
setting quality requirements that nursing homes must meet to 
participate in the Medicare and Medicaid programs and by contracting 
with states to routinely inspect homes and conduct complaint 
investigations.[Footnote 1] Moreover, to encourage compliance with 
these requirements, Congress has authorized certain enforcement 
actions, known as sanctions, including civil money penalties (CMP) or 
termination from participating in these programs. With the aging of the 
baby boom generation, the number of individuals needing nursing home 
care and the associated costs are expected to increase dramatically. 
Combined Medicare and Medicaid payments for nursing home services were 
about $67 billion in 2004, including a federal share of about $46 
billion.[Footnote 2] 

In 1998 and 1999 reports, we identified significant weaknesses in 
federal and state activities designed to detect and correct quality 
problems at nursing homes.[Footnote 3] A key finding was that sanctions 
imposed on nursing homes, including those that repeatedly harmed 
residents, often did not take effect. Instead, the sanctions were 
rescinded prior to their effective dates because homes had a grace 
period in which they could and often did correct deficiencies. We 
referred to this phenomenon as a "yo-yo" pattern of compliance because 
homes cycled in and out of compliance, harming residents while avoiding 
sanctions. Overall, we concluded that the goal of the enforcement 
process--to help ensure that homes maintain compliance with federal 
quality requirements--was not being realized. In response to our 
recommendations, the Centers for Medicare & Medicaid Services (CMS), 
the federal agency that manages these two public health care programs, 
took several steps, including the introduction of an immediate 
sanctions policy for homes found to repeatedly harm residents and the 
development of a new data system to improve management of the 
enforcement process. Under CMS's immediate sanctions policy, sanctions 
may be imposed without giving homes an opportunity to correct serious 
deficiencies that resulted in actual resident harm or put residents at 
risk of death or serious injury. We also reported that the deterrent 
effect of CMPs can be hampered by a backlog of appeals, which further 
delays payment of CMPs; by statute, CMPs are not paid until appealed 
cases are closed. 

You asked us to assess CMS's progress in improving the enforcement 
process, particularly for homes with a history of harming residents. In 
response to your request, we (1) analyzed federal sanctions from fiscal 
years 2000 through 2005 against 63 homes with a history of harming 
residents as well as nationwide trends in nursing home sanctions for 
the same time period, (2) evaluated the extent to which the homes 
cycled in and out of compliance and the impact of CMS's immediate 
sanctions policy, and (3) assessed CMS's management of enforcement 
activities. The nursing homes were located in California, Michigan, 
Pennsylvania, and Texas and their prior compliance and enforcement 
histories formed the basis for the conclusions in our March 1999 
report.[Footnote 4] These homes were selected for that report because 
of their serious or sustained compliance problems prior to 1999 and are 
not representative of homes in those states or homes 
nationwide.[Footnote 5] The 63 homes we reviewed for this report 
participated in Medicare and Medicaid for at least 6 months during 
fiscal years 2000 through 2005. Table 1 shows the number of homes that 
participated by fiscal year. Changes in the number of homes from year 
to year are a result of homes' closure, termination, or reinstatement 
of participation. For example, the change from 61 homes in fiscal year 
2000 to 59 homes in fiscal year 2001 represents the voluntary closure 
of 2 homes, the involuntary termination of 1, and the reinstatement of 
1, for a net decrease of 2 homes.[Footnote 6] 

Table 1: Number of Nursing Homes Reviewed in 1999 That Were Included in 
Our Analysis for This Report: 

1999 report: 74; 
Current report: Fiscal year: 2000: 61; 
Current report: Fiscal year: 2001: 59; 
Current report: Fiscal year: 2002: 58; 
Current report: Fiscal year: Average 2000-2002: 59; 
Current report: Fiscal year: 2003: 58; 
Current report: Fiscal year: 2004: 54; 
Current report: Fiscal year: 2005: 55; 
Current report: Fiscal year: Average 2003-2005: 56; 
Current report: Fiscal year: Total number of homes with any 
participation in 2000-2005: 63. 

Source: GAO. 

Note: Some of the 63 homes only participated in the Medicare and 
Medicaid programs for a portion of fiscal years 2000 through 2005 
because they either closed permanently or closed temporarily and were 
subsequently reinstated. To be included in our analysis we required 
such homes to have participated for at least 6 months of the fiscal 
year. 

[End of table] 

Our analysis relied primarily on (1) deficiency data from CMS's On-Line 
Survey, Certification, and Reporting system (OSCAR) and the CMS 
Providing Data Quickly (PDQ) Web site; (2) sanctions data from its Long 
Term Care Enforcement Tracking System (LTC) and ASPEN Enforcement 
Manager (AEM);[Footnote 7] and (3) CMP payment information from its CMP 
Tracking System (CMPTS).[Footnote 8] We also examined CMS regional 
office and state enforcement case files for the nursing homes we 
reviewed. We analyzed deficiency and sanctions data to identify the 
number and type of sanctions implemented and their implementation 
rates; the extent to which homes cycled in and out of compliance; the 
use of immediate sanctions for homes that repeatedly harmed residents, 
including their deterrent effect; the use of termination; and 
variability in state approaches to enforcement. To identify trends, we 
compared deficiency and sanctions data across two time periods: fiscal 
years 2000 through 2002 and fiscal years 2003 through 2005.[Footnote 9] 
We focused our analysis on three types of sanctions--CMPs, denial of 
payment for new admissions (DPNA), and terminations--which accounted 
for about 81 percent of all sanctions from fiscal years 2000 through 
2005. Although termination was used infrequently--less than 1 percent 
of all sanctions--we included it in our analysis because it is the most 
severe sanction, resulting in the loss of Medicare and Medicaid 
revenue.[Footnote 10] Based on our assessment of the data from the case 
file review, we determined that the sanctions data were sufficiently 
reliable to assess general nationwide trends in implemented sanctions. 
Because we could not conduct such checks of the data in all 50 states 
and the District of Columbia, we did not analyze trends across the 
individual states.[Footnote 11] We also reviewed CMS enforcement policy 
and guidance and discussed the immediate sanctions policy and data 
reliability issues with CMS and state officials. Finally, we obtained 
perspectives from regional office and state officials on the sanctions 
used for the homes we reviewed. Our findings on sanctions, such as 
implementation rates and use of the available range of sanctions, 
against these homes cannot be generalized to all homes in the 4 states 
or to all nursing homes nationwide. However, we believe that the 
findings are illustrative of the overall adequacy of federal and state 
responses to nursing homes with a history of serious noncompliance with 
federal quality requirements. Appendix I provides a more detailed 
description of our scope and methodology, including steps taken to 
ensure the reliability of the data used in this report. We performed 
our work from January 2005 through January 2007 in accordance with 
generally accepted government auditing standards. 

Results in Brief: 

For the homes we reviewed in four states, the number of sanctions 
implemented as well as the number of serious deficiencies cited 
declined from fiscal years 2000 through 2005--trends that were also 
seen nationwide. While the decline may reflect improved quality or 
changes to enforcement policy, it may also mask survey weaknesses that 
understate quality problems, an issue we have reported on since 1998. 
In general, the homes were cited for more deficiencies that caused harm 
to residents than other homes in their respective states. For example, 
the homes we reviewed in California had three times as many serious 
deficiencies as other homes in the state. We also found differences in 
the implementation rate of various sanctions for the homes we reviewed. 
Comparing results from the baseline period of July 1995 to October 1998 
with the period fiscal years 2003 through 2005, the implementation rate 
of CMPs increased from 32 percent to 86 percent but declined for DPNAs 
by about 20 percent. However, the deterrent effect of CMPs was diluted 
because CMS imposed CMPs at the lower end of the allowable range for 
the homes we reviewed. For example, the median per day CMP amount 
imposed for deficiencies that do not cause immediate jeopardy to 
residents was $500 in fiscal years 2000 through 2002 and $350 in fiscal 
years 2003 through 2005; the allowable range is $50 to $3,000 per day. 
Generally, CMS did not exercise its discretionary authority to impose 
DPNAs and terminations for the homes; rather, it waited until these 
sanctions could be imposed on a mandatory basis, allowing the homes 
more opportunities to escape sanctions prior to implementation. 
Moreover, in some instances, CMS extended the implementation dates of 
imposed terminations, thus allowing homes additional time to avoid 
being terminated by correcting deficiencies. 

Despite changes in federal enforcement policy, almost half of the 63 
homes we reviewed--homes with prior serious quality problems--continued 
to cycle in and out of compliance during fiscal years 2000 through 
2005, causing harm to residents. These homes corrected deficiencies 
only temporarily and, despite having had sanctions implemented, were 
again found to be out of compliance, including 8 homes that cycled in 
and out of compliance 7 or more times. During this same time period, 27 
of the 63 homes were cited 69 times for deficiencies that warranted 
immediate sanctions, but 15 of these cases did not result in immediate 
sanctions. Moreover, the "immediate sanctions" label is misleading 
because CMS's policy requires only that homes be notified immediately 
of CMS's intent to implement sanctions, not that sanctions be 
implemented immediately. When DPNAs are imposed, the lag time between 
the occurrence of a deficiency that results in an immediate sanction 
and the sanction's implementation date provides a de facto grace 
period; if the home is able to correct the deficiency, it can escape 
sanctions. Although the use of CMPs avoids this de facto grace period 
because they can be implemented retroactively, by statute, payment of 
CMPs may be delayed until after exhausting appeals of the underlying 
deficiency, a process that can take years. Nor did CMS's implementation 
of immediate sanctions appear to deter future repeat deficiencies--18 
of the 27 homes with immediate sanctions had multiple instances of such 
sanctions in fiscal years 2000 through 2005. Termination of a home from 
the Medicare and Medicaid programs was infrequent. By the end of fiscal 
year 2005, only 2 homes were terminated involuntarily because of 
quality problems. Another 9 that closed did so voluntarily. In effect, 
these homes picked their own closure dates and may have continued to 
harm residents before closing. For example, 2 such homes were cited for 
harming residents 21 and 26 times, respectively, and had sanctions 
implemented numerous times from fiscal year 2000 until their voluntary 
closures in 2004. 

In general, the effectiveness of CMS's management of nursing home 
enforcement is hampered by the overall complexity of its immediate 
sanctions policy, intended to deter repeated noncompliance, and by its 
fragmented data systems and incomplete national reporting capabilities. 
First, the complexity of the immediate sanctions policy allows some 
homes with the worst compliance histories--the very homes the policy 
was designed to address--to escape immediate sanctions. For example, 
homes that do not correct deficiencies can avoid immediate sanctions 
because of the requirement for an intervening period of compliance 
between the pair of surveys that identify serious deficiencies--that 
is, a new serious deficiency will not trigger an immediate sanction 
unless the prior serious deficiency has been corrected. Thus, if a 
state survey agency cited a home for a serious deficiency and 2 weeks 
later--before the first deficiency was corrected--cited the home for 
another serious deficiency, the home might not receive an immediate 
sanction. In addition, homes--even those with a history of multiple 
serious deficiencies--may escape immediate sanctions because a routine 
inspection without such a serious deficiency, in effect, clears the 
home's record for determining if immediate sanctions are applicable. 
The immediate sanctions associated with CMS's policy also are often 
inequitable; multiple serious deficiencies during one inspection may 
result in the same sanction as an inspection with a single serious 
deficiency. Second, CMS's fragmented and incomplete data systems 
continue to hamper its ability to monitor enforcement. We previously 
reported that CMS lacked a data system that integrated enforcement data 
nationwide and that the lack of such a system made it difficult for CMS 
to consistently manage and monitor sanctions across states and its 
regional offices. Although CMS has developed a new data system, the 
system's components are not integrated, and the national reporting 
capabilities are not complete. Finally, CMS is taking steps to improve 
its enforcement of nursing home quality requirements. In addition to 
its new data system, the agency piloted new guidance in 2006 designed 
to encourage more consistency across states in the amount of CMPs, 
revised a program that provides enhanced enforcement and monitoring of 
some homes with a history of harming residents in each state, and 
funded studies to examine the effectiveness of nursing home 
enforcement. 

We are recommending that, to increase the deterrent effect of CMPs, the 
Administrator of CMS develop an administrative process to collect CMPs 
prior to exhaustion of appeals, seek legislation for the implementation 
of this process, and address any due process concerns, as appropriate. 
We are also recommending that the CMS Administrator take actions to (1) 
improve the immediate sanctions policy to help ensure that homes that 
repeatedly harm residents or place them in immediate jeopardy do not 
escape immediate sanctions, (2) strengthen the deterrent effect of 
certain sanctions, (3) expand a program of enhanced enforcement for 
homes with a history of noncompliance, and (4) improve the 
effectiveness of the agency's data systems used for enforcement. In 
commenting on a draft of this report, CMS generally concurred with our 
recommendations but did not always specify how it would implement them. 
In addition, CMS noted that implementation of three of our 
recommendations raised resource issues and that others required 
additional research. The four states in which the nursing homes we 
reviewed were located generally concurred with our findings. 

Background: 

Ensuring the quality and safety of nursing home care has been a focus 
of considerable congressional attention since 1998. Titles XVIII and 
XIX of the Social Security Act establish minimum requirements in 
statute that all nursing homes must meet to participate in the Medicare 
and Medicaid programs, respectively. With the Omnibus Budget 
Reconciliation Act of 1987 (OBRA 87), Congress focused the requirements 
on the quality of care actually provided by a home.[Footnote 12] To 
help ensure that homes maintained compliance with the new requirements, 
OBRA 87 also established the range of available sanctions, to include 
CMPs, DPNAs, and termination.[Footnote 13] 

Ensuring Compliance with Federal Quality Requirements: 

CMS contracts with state survey agencies to assess whether homes meet 
federal quality requirements through routine inspections, known as 
standard surveys,[Footnote 14] and complaint investigations. The 
requirements are intended to ensure that residents receive the care 
needed to protect their health and safety, such as preventing avoidable 
pressure sores, weight loss, and accidents. While a standard survey 
involves a comprehensive assessment of federal quality requirements, a 
complaint investigation generally focuses on a specific allegation 
regarding resident care or safety; complaints can be lodged by a 
resident, family member, or nursing home employee. Deficiencies 
identified during either standard surveys or complaint investigations 
are classified in 1 of 12 categories according to their scope (i.e., 
the number of residents potentially or actually affected) and severity. 
An A-level deficiency is the least serious and is isolated in scope, 
while an L-level deficiency is the most serious and is considered to be 
widespread in the nursing home (see table 2).[Footnote 15] When state 
surveyors identify and cite B-level or higher deficiencies, the home is 
required to prepare a plan of correction and, depending on the severity 
of the deficiency, surveyors conduct revisits to ensure that the home 
actually implemented its plan and corrected the deficiencies.[Footnote 
16] 

Table 2: Scope and Severity of Deficiencies Identified during Nursing 
Home Surveys: 

Severity: Immediate jeopardy[A]; 
Scope: Isolated: J; 
Scope: Pattern: K; 
Scope: Widespread: L. 

Severity: Actual harm; 
Scope: Isolated: G; 
Scope: Pattern: H; 
Scope: Widespread: I. 

Severity: Potential for more than minimal harm; 
Scope: Isolated: D; 
Scope: Pattern: E; 
Scope: Widespread: F. 

Severity: Potential for minimal harm[B]; 
Scope: Isolated: A; 
Scope: Pattern: B; 
Scope: Widespread: C. 

Source: CMS. 

[A] Actual or potential for death/serious injury. 

[B] Nursing home is considered to be in substantial compliance. 

[End of table] 

Homes with deficiencies at the A, B, or C levels are considered to be 
in substantial compliance with federal quality requirements, while 
homes with D-level or higher deficiencies are considered noncompliant. 
A noncompliance period begins when a survey finds noncompliance and 
ends when the home either achieves substantial compliance by correcting 
the deficiencies or when the home is terminated from Medicare and 
Medicaid. Since 1998, the deficiencies cited during standard surveys 
have been summarized on CMS's Nursing Home Compare Web site, and CMS 
subsequently added data on the results of complaint 
investigations.[Footnote 17] These data are intended to help consumers 
select a nursing home that takes into account the quality of care 
provided to residents. 

Range of Federal Sanctions: 

CMS and the states can use a variety of federal sanctions to help 
encourage compliance with quality requirements ranging from less severe 
sanctions, such as indicating the specific actions needed to address a 
deficiency and providing an implementation time frame, to those that 
can affect a home's revenues and provide financial incentives to return 
to and maintain compliance (see table 3).[Footnote 18] Overall, two 
sanctions--CMPs and DPNAs--accounted for 80 percent of federal 
sanctions from fiscal years 2000 through 2005. 

Table 3: Sanctions Available to Encourage Nursing Home Compliance with 
Requirements: 

Sanction: CMP; 
Description: The home pays a fine for each day or instance of 
noncompliance. 

Sanction: DPNA; 
Description: Medicare and/or Medicaid payments can be denied for all 
newly admitted eligible residents.[A]. 

Sanction: Directed in-service training; 
Description: The home is required to provide training to staff on a 
specific issue identified as a problem in the survey. 

Sanction: Directed plan of correction; 
Description: The home is required to take action within specified time 
frames according to a plan of correction developed by CMS, the state, 
or a temporary manager. 

Sanction: State monitoring; 
Description: An on-site monitor is placed in the home to help ensure 
that the home achieves and maintains compliance. 

Sanction: Temporary management; 
Description: The nursing home accepts a substitute manager appointed by 
the state with the authority to hire, terminate, and reassign staff; 
obligate funds; and alter the nursing home's procedures, as 
appropriate. 

Sanction: Termination; 
Description: Termination from the Medicare and Medicaid programs. The 
home is no longer eligible to receive Medicare and Medicaid payments 
for beneficiaries residing in the home. 

Source: CMS. 

Notes: Most of the above sanctions are authorized by statute (see 42 
U.S.C. §1395i-3(h) and 42 U.S.C. §1396r(h)), while directed in-service 
training is authorized by regulation (see 42 C.F.R § 488.406(a)). 
Additional or alternative sanctions may also be used (see 42 C.F.R. § 
488.406(c)). 

[A] CMS may also deny payment for all Medicare-and/or Medicaid-covered 
residents but seldom does so because it may severely limit the homes' 
revenues for patient care. 

[End of table] 

The majority of federal sanctions implemented from fiscal years 2000 
through 2005--about 54 percent--were CMPs. CMPs may be either per day 
or per instance. CMS regulations specify a per day CMP range from $50 
to $10,000 for each day a home is noncompliant--from $50 to $3,000 for 
nonimmediate jeopardy and $3,050 to $10,000 for immediate jeopardy. The 
overall amount of the fine increases the longer a home is out of 
compliance.[Footnote 19] For example, a home with a per day CMP of 
$5,000 that is out of compliance for 10 days would accrue a total 
penalty of $50,000. A per day CMP can be assessed retroactively, 
starting from the first day of noncompliance, even if that date is 
prior to the date of the survey that identified the deficiency. 

Per instance CMPs range from $1,000 to $10,000 per episode of 
noncompliance.[Footnote 20] While multiple per instance CMPs can be 
imposed for deficiencies identified during a survey, the total amount 
cannot exceed $10,000. Per day and per instance CMPs cannot be imposed 
as a result of the same survey, but a per day CMP can be added when a 
deficiency is identified on a subsequent survey if a per instance CMP 
was the type of CMP initially imposed. Unlike other sanctions, CMPs 
require no notice period. However, if a home appeals the deficiency, by 
statute, payment of the CMP--whether received directly from the home or 
withheld from the home's Medicare and Medicaid payments--is deferred 
until the appeal is resolved.[Footnote 21] 

DPNAs made up about 26 percent of federal sanctions from fiscal years 
2000 through 2005. A DPNA denies a home payments for new admissions 
until deficiencies are corrected. In contrast to CMPs, CMS regulations 
require that homes be provided a notice period of at least 15 days for 
other sanctions, including DPNAs; the notice period is shortened to 2 
days in the case of immediate jeopardy. As a result, homes can avoid 
DPNAs if they are able to correct deficiencies during the notice 
period, which provides a de facto grace period. Unlike CMPs, DPNAs 
cannot be imposed retroactively, and payment denial is not deferred 
until appeals are resolved. 

Although nursing homes can be terminated involuntarily from 
participation in Medicare and Medicaid, which can result in a home's 
closure, termination is used infrequently.[Footnote 22] Terminations 
were less than 1 percent of total sanctions from fiscal years 2000 
through 2005. Four of the seven types of sanctions described above were 
used less frequently than CMPs and DPNAs--directed plan of correction, 
state monitoring, directed in-service training, and temporary 
management--these sanctions accounted for about 19 percent of sanctions 
nationwide from 2000 through 2005. 

Imposition of Sanctions: 

The statute permits and, in some cases, requires that DPNAs or 
termination be imposed for homes found out of compliance with federal 
quality requirements. Mandatory termination and DPNA are required, as 
follows: 

* Termination--Termination is required by regulations under the statute 
if within 23 days of the end of a survey a home fails to correct 
immediate jeopardy deficiencies,[Footnote 23] or within 6 months of the 
end of a survey the home fails to correct nonimmediate jeopardy 
deficiencies. 

* DPNA--A DPNA is required by statute if within 3 months of the end of 
a survey a home fails to correct deficiencies and return to compliance 
or when a home's last three standard surveys reveal substandard quality 
of care.[Footnote 24] 

The statute also authorizes CMS to impose discretionary DPNAs and 
discretionary terminations in situations other than those specified 
above.[Footnote 25] Federal regulations further stipulate that such 
discretionary sanctions may be implemented as long as a facility is 
given the appropriate notice period. By regulation, the notice period 
for implementing both discretionary and mandatory DPNAs and 
terminations is 15 days; in cases of immediate jeopardy, however, the 
notice period is 2 days. 

In imposing sanctions, CMS takes into account four factors: (1) the 
scope and severity of the deficiency, (2) a home's prior compliance 
history, (3) desired corrective action and long-term compliance, and 
(4) the number and severity of all the home's deficiencies. In general, 
the severity of the sanction increases with the severity of the 
deficiency. For example, for immediate jeopardy deficiencies (J, K, and 
L on CMS's scope and severity grid) the regulations require that either 
or both temporary management or termination be imposed, and also 
permits use of CMPs of from $3,050 to $10,000 per day or $1,000 to 
$10,000 per instance of noncompliance. Similarly, for deficiencies at 
the actual harm level (G, H, and I on the scope and severity grid) the 
regulations require one or a combination of the following sanctions: 
temporary management, a DPNA, a per day CMP of $50 to $3,000, or a per 
instance CMP of $1,000 to $10,000 per instance of noncompliance. In 
addition to these required sanctions, other sanctions can be included; 
for example, depending on the severity of the deficiency and a home's 
compliance history, it could have a combination of state monitoring, a 
DPNA, and a CMP. Finally, CMS is required to consider the immediacy of 
sanctions. The statute stipulates that sanctions should be designed to 
minimize the time between the identification of violations and the 
final imposition of the sanctions.[Footnote 26] 

State and CMS Roles in Sanctioning Homes: 

Enforcement of nursing home quality-of-care requirements is a shared 
federal-state responsibility. In general, sanctions are (1) initially 
proposed by the state survey agency based on a cited deficiency, (2) 
reviewed and imposed by CMS regional offices, and (3) implemented--that 
is, put into effect--by the same CMS regional office, usually after a 
required notice period (see fig. 1).[Footnote 27] CMS regional offices 
typically accept state-proposed sanctions but can modify them. The 
regional office notifies the home by letter that a sanction is being 
imposed--that is, its intent to implement a sanction--and the date it 
will be implemented. State surveyors may make follow-up visits to the 
home to determine whether the deficiencies have been corrected. The CMS 
regional office implements the sanctions if the deficiencies are not 
corrected. Homes may appeal the cited deficiency and, if the appeal is 
successful, the severity of the sanction could be reduced or the 
sanction could be rescinded. Homes have several avenues of appeal, 
including informal dispute resolution at the state survey agency level 
or a hearing before an administrative law judge, as well as before the 
Department of Health and Human Services Departmental Appeals Board. 
Under CMS policy, homes automatically receive a 35 percent reduction in 
the amount of a CMP if they waive their right to appeal before the 
Departmental Appeals Board.[Footnote 28] 

Figure 1: Federal-State Responsibilities in the Enforcement Process: 

[See PDF for image] 

Source: GAO. 

Notes: States may impose lower-level sanctions, such as state 
monitoring, without federal approval. Some state survey agencies also 
have the ability to impose federal sanctions such as DPNAs. Nursing 
homes are notified of their appeal rights when CMS imposes a sanction. 

[End of figure] 

CMS Enforcement Initiatives: 

In response to our earlier recommendations, CMS undertook a number of 
initiatives intended to strengthen enforcement, many of which we 
reported on in 2005.[Footnote 29] For example, CMS (1) revised its 
revisits policy by requiring surveyors to return to nursing homes to 
verify that serious deficiencies had actually been corrected; (2) hired 
more staff to reduce the backlog of appeals at the Health and Human 
Services Departmental Appeals Boards, the entity that adjudicates 
nursing home appeals of deficiency citations; (3) began annual 
assessments of state survey activities, known as state performance 
reviews, which cover, among other things, the timeliness of sanction 
referrals from state survey agencies to CMS regional offices; and (4) 
revised its past noncompliance policy for citing and reporting serious 
deficiencies that were missed by state surveyors during earlier surveys 
of a home. 

A key CMS enforcement initiative was the two-stage implementation of an 
immediate sanctions policy. In the first stage, effective September 
1998, CMS required states to refer for immediate sanction homes found 
to have a pattern of harming or exposing residents to actual harm or 
potential death or serious injury (H-level or higher deficiencies on 
the agency's scope and severity grid) on successive surveys.[Footnote 
30] Effective January 2000, CMS expanded the policy, requiring referral 
of homes found to have harmed one or a small number of residents (G- 
level deficiencies) on successive routine surveys or intervening 
complaint investigations.[Footnote 31] After expansion of the immediate 
sanctions policy to include G-level deficiencies, it became known as 
the double G immediate sanctions policy. 

CMS also took steps to improve its ability to manage and oversee the 
enforcement process. Our 1999 report described how CMS regions and 
states were using their own systems to track sanctions rather than 
CMS's OSCAR database. Regional office systems ranged from manual, paper-
based records to complex computer programs; none of the four states 
included in our 1999 report had tracking systems compatible with OSCAR 
or the regional office systems in use. Until it implemented a new 
enforcement data collection system, CMS used LTC, an interim 
enforcement tracking system developed and first used by its Chicago 
regional office. LTC was operational in all 10 regions by January 2000. 
CMS's enforcement data collection system--AEM--replaced LTC and was 
implemented 4 years later, on October 4, 2004. 

Recognizing the need to focus more attention on homes that historically 
provided poor care, CMS designed and launched a Special Focus Facility 
program in January 1999, instructing states to select 2 homes each for 
enhanced monitoring. Surveys were to be conducted at 6-month intervals 
rather than annually. In September 2000, CMS reported that semiannual 
surveys had been conducted at a little more than half of the original 
110 facilities. In late 2004, CMS modified the program by (1) expanding 
its scope to include more homes, (2) revising the selection criteria 
for homes, and (3) strengthening sanctions for homes that did not 
improve within 18 months. In a relevant but unrelated initiative, CMS 
established a voluntary program to help nursing homes improve the 
quality of care provided to residents. In 2002, Medicare Quality 
Improvement Organizations (QIO) began working intensively on issues 
such as preventing pressure sores and pain management with 10 percent 
to 15 percent of nursing homes in each state.[Footnote 32] Responding 
to concerns that QIOs were not working with homes that needed the most 
help, CMS established a separate pilot program in 2004; QIOs worked for 
12 months with 1 to 5 nursing homes with significant quality problems 
in 18 states to help them redesign their clinical practices. Unlike the 
Special Focus Facility program, the participation of homes in the pilot 
was voluntary. To distinguish it from the Special Focus Facility 
program, the pilot was known as the Collaborative Focus Facility 
program. 

Number of Sanctions Has Decreased: 

Among the homes we reviewed in four states, the number of implemented 
sanctions and serious deficiencies declined across two time periods-- 
fiscal years 2000 through 2002 and fiscal years 2003 through 2005. 
Federal data show similar declines for homes nationwide, a trend 
consistent with the decline in the proportion of homes cited for 
serious deficiencies that generally result in sanctions.[Footnote 33] 
Despite the decline in the number of serious deficiencies, the homes we 
reviewed generally were cited for more deficiencies that caused harm to 
residents than other homes in the four states. While the numbers of 
implemented CMPs and DPNAs at the homes we reviewed declined across the 
two time periods, the amount of CMPs paid increased. Not all imposed 
sanctions for these homes were implemented, however, which may reduce 
the deterrent effect of sanctions; in fact, we found that the 
implementation rate of certain sanctions, such as DPNAs, decreased. The 
deterrent effect of sanctions for the homes was further eroded because 
CMS generally imposed CMPs on the lower end of the allowable dollar 
range and did not exercise its authority to use discretionary DPNAs and 
terminations, allowing the homes more opportunities to escape sanctions 
prior to implementation. 

Sanctions Have Declined Nationwide: 

Among all nursing homes nationwide, sanctions declined across the two 
time periods--fiscal years 2000 through 2002 and fiscal years 2003 
through 2005.[Footnote 34] Implemented terminations declined the most 
across the two time periods (about 41 percent) and CMPs declined the 
least (about 12 percent), while the number of DPNAs declined by about 
31 percent. In the same time periods, the average number of serious 
deficiencies per home declined by about 33 percent nationwide, from 
about 0.8 to about 0.5. These downward trends are also consistent with 
the nationwide decline in the proportion of homes with serious 
deficiencies--from about 28 percent in fiscal year 2000 to about 17 
percent in fiscal year 2005 (see app. II). While the reported decline 
in serious deficiencies and the proportion of homes cited for such 
deficiencies may be due to improved quality, our earlier reports noted 
similar declines that masked (1) understatement of serious quality 
problems, and (2) inconsistency in how states conduct surveys.[Footnote 
35] For example, our current analysis found that the proportion of 
homes cited for serious deficiencies ranged from a low of about 4 
percent in Florida to a high of about 44 percent in Connecticut during 
fiscal year 2005.[Footnote 36] Across the four states we reviewed, the 
proportion of homes with serious deficiencies in fiscal year 2005 
ranged from 8 percent in California to 23 percent in Michigan. As we 
previously reported, such disparities are more likely to reflect 
inconsistency in how states conduct surveys rather than actual 
differences in the quality of care provided by homes.[Footnote 37] In 
addition, in commenting on a draft of this report, CMS noted concerns 
about whether the immediate sanctions policy has had a negative effect 
on state citations of serious deficiencies. 

Decline in Sanctions and Deficiencies for the Homes Reviewed Is 
Consistent with Nationwide Trends: 

The number of implemented sanctions at the homes we reviewed as well as 
the number of serious deficiencies cited in these homes declined across 
two time periods--fiscal years 2000 through 2002 and fiscal years 2003 
through 2005--consistent with nationwide trends. 

Deficiency trends. The average number of serious deficiencies per home 
we reviewed decreased from about 1.8 in fiscal years 2000 through 2002 
to about 0.7 in fiscal years 2003 through 2005, about a 61 percent 
decline; this decline was consistent with the national trend. During 
both time periods, however, the homes we reviewed generally performed 
more poorly than other homes in their states, having, on average, more 
G-level or higher deficiencies and more double Gs. For example, the 
Texas homes we reviewed had on average 1.3 times as many G-level or 
higher deficiencies as all other homes in the state and the California 
homes we reviewed had on average 3 times as many as all other 
California nursing homes.[Footnote 38] 

CMP trends. Due in part to the closure of some poorly performing homes 
and the citation of fewer serious deficiencies, the homes we reviewed 
had fewer CMPs in fiscal years 2003 through 2005 than in the prior 3 
fiscal years, but the amount paid was higher (see table 4). Among the 
homes, the number of implemented CMPs declined by about 42 percent from 
the first to the second time period. Although the number of CMPs among 
the homes we reviewed decreased, the amount of CMPs paid in Michigan 
more than doubled between the two time periods, accounting for much of 
the increase in the amount of CMPs paid across the two time periods 
(see app. III). States' preferences for either state or federal CMPs 
may in part affect their use. In Michigan, state officials are more 
likely to use federal CMPs and implement them in greater amounts than 
other states we reviewed. In contrast, the homes we reviewed in 
Pennsylvania had only one implemented CMP and paid no federal CMPs from 
fiscal years 2003 through 2005; however, during the same period, the 
Pennsylvania state survey agency implemented seven state CMPs and 
collected $12,050.[Footnote 39] A Pennsylvania state survey agency 
official said that the state prefers to use state sanctions because 
they can be implemented more quickly and are believed to be more 
effective than federal sanctions. The Texas state survey agency does 
not recommend more than one type of money penalty for the same 
deficiency and chooses among one of two state money penalties or a 
federal CMP.[Footnote 40] 

Table 4: Sanctions Implemented for Homes Reviewed, Fiscal Years 2000- 
2002 and 2003-2005: 

Sanction: CMP[A,B]; 
Fiscal years 2000-2002: Number: 93; 
Fiscal years 2000-2002: Duration/ amount paid: $534,527; 
Fiscal years 2003- 2005: Number: 54; 
Fiscal years 2003-2005: Duration/ amount paid: $617,552; 
Percentage change in number between two time periods: -42%. 

Sanction: DPNA[C]; 
Fiscal years 2000-2002: Number: 52; 
Fiscal years 2000-2002: Duration/ amount paid: 2,451days; 
Fiscal years 2003-2005: Number: 30; 
Fiscal years 2003-2005: Duration/ amount paid: 1,245 days; 
Percentage change in number between two time periods: -42%. 

Sanction: Involuntary termination; 
Fiscal years 2000-2002: Number: 1; 
Fiscal years 2000-2002: Duration/ amount paid: NA; 
Fiscal years 2003-2005: Number: 1; 
Fiscal years 2003-2005: Duration/ amount paid: NA; 
Percentage change in number between two time periods: 0%. 

Source: GAO analysis of LTC data, AEM, CMS regional office and state 
enforcement case files, and CMPTS. 

Note: Includes homes that were open for at least part of the 6-year 
period. 

NA = Not applicable. 

[A] Includes per day and per instance CMPs. 

[B] Amount paid for CMPs implemented in these fiscal years. 

[C] Includes mandatory and discretionary DPNAs. 

[End of table] 

DPNA trends. The number of DPNAs declined by 42 percent from fiscal 
years 2000 through 2002 to fiscal years 2003 through 2005 for the homes 
we reviewed. Overall, the duration of the DPNAs decreased by 12 percent 
from the first to the second time period. The duration of DPNAs among 
the Texas homes we reviewed decreased the most--from an average of 46 
days in the first time period to an average of 26 days in the second 
time period. The duration of DPNAs among the Michigan and Pennsylvania 
homes also decreased (see app. III). In California, however, the DPNAs 
were in effect longer in the second time period--from an average of 39 
days in fiscal years 2000 through 2002 to an average of 63 days in 
fiscal years 2003 through 2005. As a result, homes in California were 
out of compliance for longer periods of time. 

Termination trends. Only two of the homes we reviewed closed 
involuntarily--that is, they were terminated for cause by CMS because 
of health and safety issues. One of the two homes has since been 
certified to participate in Medicare again.[Footnote 41] An additional 
nine other homes closed voluntarily, although four reopened at some 
point during fiscal years 2000 through 2005.[Footnote 42] However, a 
home's voluntary closure may not accurately reflect the degree to which 
the home had quality problems, such as a history of harming residents, 
that put the home at risk of involuntary termination. The reasons for 
closure, as recorded by CMS, are general and do not always reflect that 
homes may have histories of harming residents and may have been at risk 
of involuntary termination. 

Implementation Rate of Some Sanctions Has Declined for the Homes 
Reviewed: 

The implementation rate of DPNAs and terminations declined for the 
homes we reviewed, while the implementation rate of CMPs increased 
across three time periods (see fig. 2). Some sanctions are never 
implemented because CMS rescinds them if homes correct deficiencies 
before the implementation date, a situation we noted in our 1999 
report.[Footnote 43] Thus, sanctions may be considered more of a threat 
than a real consequence of noncompliance. 

We compared the implementation rates of CMPs, DPNAs, and terminations 
across three time periods: (1) July 1995 to October 1998, the time 
period covered in our March 1999 report;[Footnote 44] (2) fiscal years 
2000 through 2002; and (3) fiscal years 2003 through 2005. From the 
first time period to the third, the implementation rate for DPNAs 
declined by about 20 percent and the implementation rate for 
terminations declined by about 97 percent. In contrast, across the same 
time periods, the overall implementation rate for CMPs increased from 
32 percent in the first time period to 86 percent in the third time 
period, an almost threefold increase. The timing of this increase 
coincides with the January 2000 implementation of the immediate 
sanctions policy, suggesting that the increase may in part be related 
to the policy's implementation. 

Figure 2: Percentage of Implemented Sanctions for Homes Reviewed Over 
Three Time Periods (July 1995-October 1998, Fiscal Years 2000-2002, and 
Fiscal Years 2003-2005): 

[See PDF for image] 

Source: GAO analysis of LTC, OSCAR, and CMS regional office and state 
enforcement case files. 

[End of figure] 

CMS Did Not Take Advantage of the Full Range of Sanctions for the Homes 
Reviewed: 

Among the homes we reviewed, CMS did not use the full range of its 
sanctions authority, generally imposing CMPs on the lower end of the 
allowable range.[Footnote 45] In addition, CMS imposes DPNAs and 
involuntary terminations when they are mandatory, but generally not 
when they are discretionary. Homes subject to such mandatory sanctions 
have more opportunities to escape sanctions prior to implementation. 
The median per instance CMP implemented was $2,000 in fiscal years 2000 
through 2002 and $1,750 in fiscal years 2003 through 2005, although the 
maximum per instance CMP can be as high as $10,000. The median per day 
CMP implemented for nonimmediate jeopardy deficiencies was $500 in 
fiscal years 2000 through 2002 and $350 in fiscal years 2003 through 
2005, significantly below the maximum of $3,000 per day. In cases in 
which homes were cited for immediate jeopardy and the maximum potential 
per day CMP is $10,000, the median per day CMP implemented was $3,050 
in fiscal years 2000 through 2002 and $5,050 in fiscal years 2003 
through 2005. According to one CMS official, the agency generally 
hesitates to impose CMPs that are higher than $200 per day, in part 
because of concerns that higher per day CMPs could bankrupt some 
homes.[Footnote 46] But the same official noted that the CMPs being 
imposed are not enough to "make nursing homes take notice" or to deter 
them from deficient practices. Another CMS official stated that some 
homes consider CMPs a part of the "cost of doing business" or as having 
no more effect than a "slap on the wrist." Table 5 provides examples of 
homes we reviewed with implemented CMPs that were at the low end of the 
allowable CMP range. 

Table 5: Examples of Homes with Low Implemented CMPs: 

Home's location: Michigan; 
Surveyors' comments: "A significant medication error occurred when 
resident #8 was administered [the wrong medication] over a three day 
period. The resident experienced hypoglycemia and required 
hospitalization. Upon return from the hospital there was evidence of 
actual harm: a decline in ability to perform activities of daily 
living."; 
Summary of deficiencies: 1 G; 
CMP implemented: $1,500 per instance; 
Allowable CMP range: $1,000 - $10,000 per instance of noncompliance. 

Home's location: Texas; 
Surveyors' comments: "Facility nurse aides failed to promptly report an 
allegation of possible sexual abuse. Resident reported the incident to 
two nurse aides, however, it was not reported. Also, reference checks 
were not documented for 4 employees and 4 employees had not attended an 
inservice [training session] on abuse."; 
Summary of deficiencies: 3 F, 1 E; 
CMP implemented: $250 per day for 150 days; 
Allowable CMP range: $50 - $3,000 per day for noncompliance other than 
immediate jeopardy. 

Home's location: Texas; 
Surveyors' comments: "There was an [immediate jeopardy deficiency for 
staff mistreatment of residents]. There was a failure to monitor 
residents in distress."; 
Summary of deficiencies: 1 L (immediate jeopardy), 9 G; 
CMP implemented: Immediate jeopardy: $3,050 per day for 14 days; After 
immediate jeopardy removed: $400 per day for 86 days; $300 per day for 
46 days; $50 per day for 6 days; 
Allowable CMP range: $3,050 - $10,000 per day for immediate jeopardy; 
$50 - $3,000 per day for nonimmediate jeopardy. 

Source: GAO analysis of CMS regional office and state case files and 
LTC. 

Note: In addition to CMPs, CMS also imposed DPNAs and terminations-- 
either mandatory or discretionary. All of the DPNAs but none of the 
terminations were implemented. 

[End of table] 

CMS is likely to impose DPNAs and terminations only when required to do 
so. However, CMS also has broad authority to impose DPNAs and 
terminations at its discretion, which can facilitate quicker 
implementation. Discretionary DPNAs and terminations can be implemented 
any time after a survey if the sanction is appropriate for the cited 
deficiencies and the required notice period is met. In contrast, the 
soonest that mandatory DPNAs and terminations for nonimmediate jeopardy 
can be implemented is 3 and 6 months, respectively, after the survey on 
which the deficiencies were cited.[Footnote 47] Despite the greater 
expediency of discretionary DPNAs, 64 percent of the DPNAs CMS imposed 
were mandatory for fiscal years 2000 through 2005 for the homes we 
reviewed. For example, CMS imposed a total of six DPNAs during fiscal 
years 2000 through 2003 on a Pennsylvania home with demonstrated 
compliance problems. Of those six DPNAs, the first five were mandatory 
DPNAs. Only the last DPNA--imposed after multiple years of repeated 
noncompliance at the G-level or higher--was a discretionary DPNA. 

Moreover, CMS imposed significantly more mandatory terminations than 
discretionary terminations; in fiscal years 2000 through 2005, 118 
mandatory and 5 discretionary terminations were imposed on the homes we 
reviewed.[Footnote 48] None of the mandatory terminations were 
implemented, but 2 discretionary terminations were implemented--one 
each in Michigan and Texas.[Footnote 49] An official from the Texas 
state survey agency said that the CMS regional office in Dallas prefers 
to impose mandatory terminations, unless there is cause to believe 
there will be no improvements in the care provided by the nursing home. 
Mandatory terminations give homes 6 months to correct deficiencies 
before being implemented, as opposed to discretionary terminations, 
which can be implemented more quickly. 

Even when CMS imposes terminations, their deterrent effect is weakened 
because the agency sometimes extends the termination dates. For 
example, CMS extended the discretionary termination dates for up to 6 
months for some of the Texas homes we reviewed if the nursing homes had 
lower-level deficiencies on subsequent surveys. The termination date 
imposed on one Texas nursing home we reviewed was extended three times 
in fiscal year 2001 from the original date of April 18 to June 26, then 
to July 26, and finally to September 26. The first extension occurred 
because the home corrected the deficiencies that caused immediate 
jeopardy cited during the first survey. Therefore, despite the fact 
that this home continued to be found out of compliance for deficiencies 
such as mistreatment or neglect of residents during subsequent surveys, 
CMS extended the termination date twice to give the home an additional 
opportunity to correct those deficiencies and achieve substantial 
compliance. The termination ultimately was rescinded because the home 
corrected the deficiencies, but the home was subsequently cited for 
eight G-level deficiencies such as inadequate treatment or prevention 
of pressure sores, employing convicted abusers, and poor accident 
supervision or prevention. In 2004, the home closed voluntarily. 

Despite Changes in Federal Enforcement Policy, Many Homes Continued to 
Cycle In and Out of Compliance: 

Despite changes in federal enforcement policy, almost half of the homes 
we reviewed--homes with prior serious quality problems--continued to 
cycle in and out of compliance, continuing to harm residents. These 
homes corrected deficiencies only temporarily and, despite having 
sanctions implemented, were again found to be out of compliance during 
subsequent surveys. Our analysis also showed that in some cases the 
double Gs did not result in immediate sanctions as required, even 
though about 40 percent of the homes were cited for double Gs during 
fiscal years 2000 through 2005. In addition, the term "immediate 
sanctions policy" is misleading because the policy requires only that 
sanctions be imposed, that is, that homes be notified immediately of 
CMS's intent to implement sanctions, not that sanctions must be 
implemented immediately. Furthermore, when a sanction is implemented 
for a double G citation, there is a lag time between when the double G 
occurs and the sanction's effective date. CMS cited double Gs multiple 
times at several of the homes we reviewed, suggesting that immediate 
sanctions did not deter future noncompliance as intended. Terminations 
of homes is infrequent, in part because of concerns such as local 
access to other nursing facilities and the effect on residents if they 
are moved, and in part because CMS allows some problem homes to 
continue operating until the homes eventually close voluntarily. 

Many Homes Cycled In and Out of Compliance, Continuing to Harm 
Residents: 

Consistent with our earlier work, our current analysis showed that 
sanctions appear to have induced homes to correct deficiencies only 
temporarily because surveyors found that many of the homes we reviewed 
with implemented sanctions were again out of compliance on subsequent 
surveys.[Footnote 50] Commenting on this phenomenon, state survey 
agency officials said that improvements resulting from sanctions might 
last about 6 months. From fiscal years 2000 through 2005, 31 of the 63 
homes we reviewed (about 49 percent) cycled in and out of compliance 
more than once, harming residents, even after sanctions had been 
implemented, including 8 homes that did so seven times or more (see 
fig. 3). 

Figure 3: Frequency that Reviewed Homes Cycled In and Out of 
Compliance, Fiscal Years 2000-2005: 

[See PDF for image] 

Source: GAO analysis of LTC, OSCAR, and CMS regional office and state 
enforcement case files. 

Note: This figure illustrates the concept of a yo-yo pattern of 
compliance. While the time periods that a home is in or out of 
compliance appear to be of uniform duration, the duration can vary. 

[End of figure] 

Each of the 31 homes that cycled in and out of compliance more than 
once during the period we reviewed had at least one G-level or higher 
deficiency in at least one period of noncompliance; 19 had at least one 
G-level or higher deficiency in every noncompliance period. Table 6 
shows the number and length of noncompliance periods for a Michigan 
home we reviewed that cycled in and out of compliance nine times from 
fiscal years 2000 through 2005; the home remained open as of November 
2006. Appendix IV provides similar examples for homes in California, 
Pennsylvania, and Texas. Homes' correction of deficiencies often was 
temporary, despite receiving sanctions. Thus, once the homes we 
reviewed corrected deficiencies, they maintained compliance for a 
median of 133 days and then cycled out of compliance again. Some homes 
cycled out of compliance more quickly--homes were again out of 
compliance in 30 days or less about 8 percent of the time and within 60 
days about 28 percent of the time. 

Table 6: Example of a Michigan Nursing Home That Frequently Cycled In 
and Out of Compliance and Was Still Open as of November 2006: 

Noncompliance period in fiscal years 2000-2005 (no. of days): 1[ST] (41 
days); 
Examples of the nature of deficiencies[A]: 
* Inadequate treatment or prevention of pressure sores; 
* Poor quality of care; 
Summary of G-level or higher deficiencies: 1 G; 
Enforcement action implemented[B]: * Per instance CMP ($1,000). 

Noncompliance period in fiscal years 2000-2005 (no. of days): 2nd (185 
days); 
Examples of the nature of deficiencies[A]: 
* Poor nutrition; 
* Poor quality of care; 
Summary of G-level or higher deficiencies: 1 G; 
Enforcement action implemented[B]: 
* 1[ST] per day CMP ($10,000/day); 
* 2nd per day CMP ($100/day); 
* Per instance CMP ($1,500); 
* Mandatory DPNA (109 days). 

Noncompliance period in fiscal years 2000-2005 (no. of days): 3rd (176 
days); 
Examples of the nature of deficiencies[A]: 
* Inadequate treatment or prevention of pressure sores; 
* Poor accident supervision or prevention; 
Summary of G-level or higher deficiencies: 5 G; 
Enforcement action implemented[B]: 
* Per instance CMP ($10,000); 
* Mandatory DPNA (85 days). 

Noncompliance period in fiscal years 2000-2005 (no. of days): 4th (158 
days); 
Examples of the nature of deficiencies[A]: 
* Resident abuse; 
* Employing convicted abusers; 
Summary of G-level or higher deficiencies: 1 J (immediate jeopardy), 3 
G; 
Enforcement action implemented[B]: 
* 1[ST] per day CMP ($850/day); 
* 2nd per day CMP ($3,500/day); 
* 3rd per day CMP ($1,000/day); 
* Discretionary DPNA (127 days). 

Noncompliance period in fiscal years 2000-2005 (no. of days): 5[TH] 
(107 days); 
Examples of the nature of deficiencies[A]: 
* Resident abuse; 
* Failure to provide necessary services for daily living; 
Summary of G-level or higher deficiencies: 3 H, 3 G; 
Enforcement action implemented[B]: 
* Per day CMP ($200/day); 
* Discretionary DPNA (74 days). 

Noncompliance period in fiscal years 2000-2005 (no. of days): 6[TH] (94 
days); 
Examples of the nature of deficiencies[A]: 
* Poor accident supervision or prevention; 
Summary of G-level or higher deficiencies: 1 G; 
Enforcement action implemented[B]: 
* Per day CMP ($350/day); 
* Discretionary DPNA (62 days). 

Noncompliance period in fiscal years 2000-2005 (no. of days): 7[TH] 
(127 days); 
Examples of the nature of deficiencies[A]: 
* Failure to provide necessary services for daily living; 
* Poor accident supervision or prevention; 
Summary of G-level or higher deficiencies: 1 J (immediate jeopardy), 1 
G; 
Enforcement action implemented[B]: 
* 1[ST] per day CMP ($3,550/day); 
* 2nd per day CMP ($450/day); 
* Mandatory DPNA (35 days). 

Noncompliance period in fiscal years 2000-2005 (no. of days): 8[TH] (89 
days); 
Examples of the nature of deficiencies[A]: 
* Inadequate treatment or prevention of pressure sores; 
* Employing convicted abusers; 
* Medication errors; 
Summary of G-level or higher deficiencies: 2 G; 
Enforcement action implemented[B]: 
* Per day CMP ($500/day); 
* Discretionary DPNA (59 days). 

Noncompliance period in fiscal years 2000-2005 (no. of days): 9[TH] (83 
days); 
Examples of the nature of deficiencies[A]: 
* Inadequate treatment or prevention of pressure sores; 
* Medication errors; 
Summary of G-level or higher deficiencies: 1 H; 
Enforcement action implemented[B]: 
* Per day CMP ($750/day); 
* Discretionary DPNA (51 days). 

Source: GAO analysis of OSCAR, ETS, and AEM data. 

Note: The table only includes federal sanctions imposed and 
implemented; sanctions imposed but not implemented and state sanctions 
are not included. 

[A] Examples of the nature of deficiencies include D-level or higher 
deficiencies. 

[B] In a number of cases, more than one per day CMP is listed because 
CMS can raise or lower per day CMP amounts based on changes in 
deficiencies. 

[End of table] 

Relatively Few Homes Reviewed Were Cited for Double Gs: 

Despite the large number of G-level or higher deficiencies cited for 
the homes we reviewed, relatively few of these homes were cited for 
double Gs, and some double G citations did not result in sanctions. 
Over the 6-year period, 27 of the homes we reviewed had 69 double Gs. 
However, 47 of the homes had 444 G-level or higher deficiencies. We 
found no record that CMS imposed a sanction for 15 of the 69 double Gs, 
but the data did show that CMS implemented sanctions for the remaining 
double G cases.[Footnote 51] 

Across the four states we reviewed, there was variation in the citation 
of G-level or higher deficiencies and the implementation of immediate 
sanctions. For example, from fiscal years 2000 through 2005, 35 percent 
of G-level or higher deficiencies and 52 percent of double Gs among the 
homes we reviewed were cited in Michigan, while 9 percent of the G- 
level or higher deficiencies and 4 percent of the double Gs were cited 
in homes in California. In California, complaints typically are 
investigated under state licensure authority and the findings generally 
are not recorded in the same manner as deficiencies cited under the 
federal process,[Footnote 52] which may contribute to lower double G 
citation rates in the state.[Footnote 53] Thus, California homes are 
not cited for a double G when the subsequent deficiency equivalent to a 
G-level or higher deficiency was found during a complaint 
investigation.[Footnote 54] Complaint surveys with G-level or higher 
deficiencies often lead to double Gs. One CMS official stated that if 
complaints against California nursing homes were investigated under the 
federal complaint investigation procedure, more double Gs would be 
cited in California.[Footnote 55] The California Department of Health 
Services conducted a pilot to test the use of the federal complaint 
procedure in select district offices, in part because of the low double 
G citation rate. As of November 2006, the department decided not to 
expand or complete a formal evaluation of the pilot; instead, the 
department is focusing on eliminating its backlog of complaints and 
initiating complaint investigations within required time 
frames.[Footnote 56] 

Immediate Sanctions Often Not Immediate and Do Not Appear to Deter 
Noncompliance: 

Although referred to as the "immediate sanctions" policy, the term is 
misleading because (1) there is a lag between when the double G is 
cited and when the sanction is implemented, negating the sanction's 
immediacy; (2) the policy only requires that sanctions be imposed 
immediately, which does not guarantee that the sanction will be 
implemented; and (3) homes may not actually pay a CMP, the most 
frequently implemented sanction, until years after citation of the 
double G because payment is suspended until after appeals have been 
adjudicated. Delays in implementing DPNAs and in collecting CMPs--which 
diminish their immediacy--coupled with their nominal amounts may 
undermine their deterrent effect. 

Immediate sanctions often are not immediate because there is a lag time 
between the identification of deficiencies during the survey and when a 
sanction (i.e., a CMP or DPNA) is actually implemented.[Footnote 57] 
CMS implemented about 68 percent of the DPNAs for double Gs among the 
homes we reviewed during fiscal years 2000 through 2005 more than 30 
days after the survey (see app. V). In contrast, CMPs can go into 
effect as early as the first day the home was out of compliance, even 
if that date is prior to the survey date, because, unlike DPNAs, CMPs 
do not require a notice period.[Footnote 58] About 98 percent of CMPs 
imposed for double Gs took effect on or before the survey date. Figure 
4 illustrates the lag time that can occur between the survey date and 
the implementation date of the sanction, especially with regard to 
DPNAs. For example, in fiscal years 2000 through 2005, 60 percent of 
the DPNAs in the homes we reviewed were implemented 31 to 60 days from 
the date of the survey citing deficiencies. In contrast, nearly all 
CMPs were implemented on or before the survey date. 

Figure 4: Number of Days between Survey and Implementation of CMPs and 
DPNAs among Homes Reviewed, Fiscal Years 2000-2005: 

[See PDF for image] 

Source: GAO analysis of LTC, OSCAR, and CMS regional office and state 
enforcement case files. 

Note: CMPs can take effect prior to the date of the survey, if the date 
of noncompliance can be established. In cases where an appeal has 
changed the determination of the date of noncompliance, the 
implementation date of CMPs would be modified accordingly. Some CMPs 
and DPNAs were not included in this analysis because implementation 
dates were not available. 

[End of figure] 

While the immediate sanctions policy requires that sanctions be imposed 
immediately, it is silent on how quickly sanctions should be 
implemented. A sanction is considered imposed when a home is notified 
of CMS's intent to implement a sanction--15 days from the date of the 
notice. If during the 15-day notice period the nursing home corrects 
the deficiencies, no sanction is implemented. Thus, even under the 
immediate sanctions policy, which is intended to eliminate grace 
periods for nursing homes repeatedly cited for deficiencies at the 
actual harm level or higher, nursing homes have a de facto grace 
period. 

While CMPs can be implemented closer to the date of survey than DPNAs, 
the immediacy and the effect of CMPs may be diminished by (1) the 
significant time that can pass between the citation of deficiencies on 
a survey and the home's payment of the CMP and (2) the low amounts 
imposed, as described earlier in this report.[Footnote 59] By statute, 
payment of CMPs is delayed until appeals are exhausted.[Footnote 60] 
For example, a Michigan home did not pay its CMP of $21,600 until more 
than 2 years after a February 2003 survey had cited a G-level 
deficiency.[Footnote 61] (See fig. 5.) The February G-level citation 
was a repeat deficiency: less than a month earlier, the home had 
received another G-level deficiency in the same quality of care area. 
The delay in collecting the fine in this case is consistent with a 2005 
report from the Office of Inspector General of the Department of Health 
and Human Services that found that the collection of CMPs in appealed 
cases takes an average of 420 days--a 110 percent increase in time over 
nonappealed cases--and "consequently, nursing homes are insulated from 
the repercussions of enforcement by well over a year."[Footnote 62] 

Unlike the Social Security Act, the federal Surface Mining Control and 
Reclamation Act of 1977 provides for the collection of CMPs prior to 
exhaustion of administrative appeals.[Footnote 63] Under this statute, 
mining operators charged with civil money penalties have 30 days to 
either pay the penalty in full or forward the proposed amount for 
placement in an escrow account pending resolution of appeals. This 
provision, requiring escrow deposit of a proposed penalty assessment, 
has been upheld by three federal circuit courts of appeal, all citing 
the various procedural safeguards as helping to ensure sufficient due 
process to affected operators.[Footnote 64] For example, these courts 
cited the availability of an informal conference at which mining 
operators may present information relevant to an assessment of a 
penalty. It is unclear whether the informal dispute resolution process 
available to nursing homes would provide due process similar to that 
provided under the Federal Mining statute. Nonetheless, the Social 
Security Act would preclude a more expeditious collection of nursing 
home CMPs. 

Figure 5: Lag Time between Survey and CMP Payment for a Michigan 
Nursing Home: 

[See PDF for image] 

Source: GAO analysis of LTC, OSCAR, and CMS regional office and state 
enforcement case files. 

[End of figure] 

Despite the potentially negative consequences, CMS's implementation of 
the immediate sanctions policy does not appear to deter homes from 
harming residents in the future. Two-thirds (18) of the 27 nursing 
homes cited for double Gs that subsequently had sanctions implemented 
went on to be cited again for one or more additional double Gs. (See 
fig. 6.) 

Figure 6: Number of Homes with One or More Double Gs, Fiscal Years 2000-
2005: 

[See PDF for image] 

Source: GAO analysis of CMS PDQ. 

[End of figure] 

Termination Used Infrequently: 

Nursing homes, even those that repeatedly harm residents, are 
infrequently terminated because of CMS's concerns about access to other 
sources of nursing care and the impact of moving residents. Of the 
homes we reviewed, two were terminated involuntarily for cause. Another 
nine homes closed voluntarily,[Footnote 65] which is not a sanction 
because the homes chose to close. However, the actual reason for 
closure is not always clear; a home may close to avoid involuntary 
termination because of quality problems cited by state 
surveyors.[Footnote 66] Allowing a problem home to close voluntarily 
rather than terminating it may result in continuing harm to residents 
until the home decides to close. For example, two homes we reviewed in 
Pennsylvania and Texas closed voluntarily, but the histories of both 
homes show that they were repeatedly cited for harming residents from 
fiscal year 2000 through the time of their closures, over 4 years later 
in January 2004. The Pennsylvania home cycled in and out of compliance 
4 times during the period we reviewed and had noncompliance periods 
lasting an average of 170 days. The Texas home cycled in and out of 
compliance 10 times during the period reviewed and had average 
noncompliance periods of 46 days. On average, both homes had about 6 G- 
level or higher deficiencies per year in areas such as inadequate 
treatment or prevention of pressure sores and resident abuse.[Footnote 
67] The home in Pennsylvania had an average of 31 other deficiencies 
per year and the Texas home had an average of 27.[Footnote 68] 

Four homes we reviewed had similar deficiency histories. Two closed 
voluntarily and two remained open as of November 2006 (see table 7). 
Although the homes that remained open met the deadline to correct 
deficiencies before the termination would have been implemented, a 
home's ability to correct deficiencies in a specified period of time 
may not be the strongest criteria upon which to determine whether a 
home should remain open, because correcting deficiencies does not 
ensure that the home will improve residents' quality of care and does 
not prevent the home from again falling out of compliance. For example, 
the California and Michigan homes in table 7 were still operating as of 
November 2006 but cycled in and out of compliance four and seven times, 
respectively. 

Table 7: Examples of Homes' Deficiency Histories and Termination 
Actions, Fiscal Years 2000-2005: 

California home[C]. 

Examples of deficiencies causing harm to residents[A]: 
* A resident choked to death when the suction machines that should have 
been maintained in working order did not have the requisite parts. 
Indeed, during an unannounced inspection 2 days following the death of 
this resident, it was noted that there were no functional suction 
machines in the facility; 
Deficiency history: 
* 173 D-level or higher deficiencies; 
* Cycled in and out of compliance 4 times; 
Enforcement history[B]: 
* DPNA (142 days); 
* CMP ($193,780); 
* Mandatory termination imposed (4 times); 
* Discretionary termination imposed (0 times); 
Current status: In operation as of November 2006. 

Michigan home. 

Examples of deficiencies causing harm to residents[A]: 
* The facility failed to provide proper respiratory treatment and care 
for a resident, resulting in the resident's hospitalization for acute 
respiratory failure; 
* During an inspection, several residents' pressure sores were observed 
to be untreated. For example, one resident had two areas of dead tissue 
on his feet. The facility acknowledged that the resident should have 
been wearing protective heel pads when in bed, and yet his bare feet 
were uncovered, both heels rested directly on the mattress, and he was 
not wearing heel protectors, which were lying nearby; 
Deficiency history: 
* 95 D-level or higher deficiencies; 
* Cycled in and out of compliance 7 times; 
Enforcement history[B]: 
* DPNA (58 days); 
* CMP ($40,970); 
* Mandatory termination imposed (7 times); 
* Discretionary termination imposed (0 times); 
Current status: In operation as of November 2006. 

Pennsylvania home. 

Examples of deficiencies causing harm to residents[A]: 
* "Resident eloped and was found on the courtyard froze (sic) to 
death."; 
* "A resident was found to have bruises on the inner thighs and arms 
and appeared to be a victim of abuse. The staff did not report this to 
the local police and bathed resident prior to assessment for sexual 
abuse."; 
Deficiency history: 
* 159 D-level or higher deficiencies, fiscal years 2000-2004; 
* Cycled in and out of compliance 4 times; 
Enforcement history[B]: 
* DPNA (229 days); 
* CMPs ($47,700); 
* Mandatory termination imposed (6 times); 
* Discretionary termination imposed (0 times); 
Current status: Closed January 2004; Reason for closure: voluntary-
merger/closure. 

Texas home. 

Examples of deficiencies causing harm to residents[A]: 
* "Conditions remain poor, residents are not clean or groomed, drug 
errors continue, restorative care is poor. Will give facility the full 
6 months to try to come into compliance, continue all remedies."; 
Deficiency history: 
* 141 D-level or higher deficiencies, fiscal years 2000-2004; 
* Cycled in and out of compliance 10 times; Enforcement history[B]: 
* DPNA (228 days); 
* CMPs ($146,244); 
* Mandatory or discretionary termination imposed (10 times)[D]; 
Current status: Closed January 2004; Reason for closure: voluntary-
merger/closure. 

Source: GAO analysis of LTC, OSCAR, CMPTS, and CMS regional office and 
state enforcement files. 

[A] Statements are from surveyors' notes and are either paraphrased or 
direct quotes. 

[B] The CMP amount reflects the amount payable by the home, but is not 
necessarily the amount the home actually paid. 

[C] These data likely understate the quality problems at this home 
because California primarily conducts complaint investigations under 
its state licensure authority and did not record serious deficiencies 
identified during such investigations in OSCAR. In commenting on a 
draft of this report, California noted that this home did receive the 
highest state deficiency citation and was assessed a state CMP of 
$60,000. 

[D] Because the Texas data did not always allow us to distinguish 
between mandatory and discretionary terminations, we report the total 
number of imposed terminations. 

[End of table] 

According to CMS and state officials, factors that may prevent or delay 
termination of problem nursing homes include (1) concerns regarding 
lack of access to alternate local nursing facilities, (2) the potential 
for resident trauma as a result of transfer to another home, (3) the 
preference of residents' families for homes located close by, and (4) 
pressure to keep homes open from families and other 
stakeholders.[Footnote 69] Our analysis of alternatives to the 4 poorly 
performing homes in table 7--those that closed voluntarily or are still 
open--showed that there were from 2 to 37 homes within 10 miles of 
these homes, and from 5 to 120 homes within 25 miles.[Footnote 70] 

Complex Immediate Sanctions Policy and Data Limitations Hamper CMS 
Management of Enforcement: 

While the goal of enforcement is to help ensure nursing home compliance 
with federal quality requirements, CMS management of the process is 
hampered by the complexity of its immediate sanctions policy and by its 
fragmented and incomplete data systems. The agency's immediate 
sanctions policy, intended to deter repeat noncompliance, fails to hold 
some homes accountable for repeatedly harming residents. In addition, 
although CMS has developed a new data system, the system's components 
are not integrated and the national reporting capabilities are not 
complete, hampering the agency's ability to track and monitor 
enforcement. Finally, CMS has taken some steps intended to improve 
enforcement of nursing home quality requirements, such as developing 
guidance to help ensure greater consistency across states in CMP 
amounts, revising its Special Focus Facility program, and commissioning 
two studies to examine the effectiveness of nursing home enforcement. 
It is not clear, however, the extent to which--or when--these 
initiatives will address the enforcement weaknesses we found. 

Immediate Sanctions Policy Is Complex and Fails to Hold Some Homes 
Accountable: 

The double G immediate sanctions policy is complex and fails to hold 
some homes accountable. In 2003, we reported that the early 
implementation of the policy was flawed.[Footnote 71] We found that 
between January 2000 and March 2002 over 700 cases that should have 
been referred for immediate sanctions were not because (1) the policy 
was misunderstood by some states and regional offices, (2) states 
lacked adequate systems for identifying deficiencies that triggered an 
immediate sanction, and (3) actions of two of the four states were at 
variance with CMS policy. CMS developed an on-line reporting tool for 
use by survey agency and regional office staff to automate the 
identification of double Gs.[Footnote 72] CMS also offered training 
sessions and issued additional guidance to state survey agencies and 
regional offices. While the on-line reporting tool and training were 
useful, they did not address the underlying complexity of the policy. 
For example, CMS staff told us that in developing the tool they had 
initially misinterpreted the double G immediate sanctions policy. As a 
result, the tool produced many false positives: that is, it identified 
deficiencies as triggering an immediate sanction that in fact did not 
occur. Moreover, a December 2005 report by the Office of the Inspector 
General of the Department of Health and Human Services also reported 
that state survey agency staff continued to have difficulty identifying 
double G cases.[Footnote 73] 

Furthermore, our analysis of CMS's application of the policy to the 
homes we reviewed demonstrated that the policy's complex rules allowed 
homes to escape immediate sanctions even if they repeatedly harmed 
residents; these rules include (1) the requirement for an intervening 
period of compliance, (2) the clearing effect of standard surveys, and 
(3) the lack of differentiation between single and multiple instances 
of harm. Such rules may in part explain why the homes we reviewed only 
had 69 instances of immediate sanctions over a 6-year period, despite 
being cited 444 times for deficiencies that harmed residents. 

Intervening period of compliance. G-level or higher deficiencies only 
count toward a double G immediate sanction if the home has an 
intervening period of compliance between the two G-level or higher 
deficiencies. In order to receive an immediate sanction, a home has to 
achieve substantial compliance between the pair of surveys on which the 
G-level or higher deficiencies are cited. As a result of this rule, 
homes that do not correct deficiencies do not receive immediate 
sanctions, while homes that do correct deficiencies do receive 
immediate sanctions. CMS officials stated that the intent of the policy 
as written was to give nursing homes a chance to correct deficiencies 
and achieve a period of compliance. Without this provision, CMS 
officials believe that homes could get caught in endless double G 
cycles. 

The following example illustrates how the policy allows nursing homes 
to escape immediate sanctions if they do not correct deficiencies and 
have ongoing noncompliance periods.[Footnote 74] 

* In a 9-month time period, a Pennsylvania home had seven surveys, each 
with at least one G-level deficiency (a total of 19 G-level 
deficiencies).[Footnote 75] However, double G immediate sanctions were 
triggered by only two pairs of surveys because the home had failed to 
correct some deficiencies before the next survey that again found 
actual harm.[Footnote 76] Figure 7 illustrates how some pairs of 
surveys with G-level deficiencies do not count as a double G because of 
the intervening period of compliance rule. For example, both the March 
and April surveys cited G-level deficiencies. However, the pair of 
surveys did not result in a double G, which would have triggered 
immediate sanctions because the home did not correct the G-level 
deficiency cited on the March survey before the next G-level deficiency 
was cited in April. Following the April survey, the home corrected the 
deficiencies, resulting in a period of compliance. In July, another 
survey found a new G-level deficiency. Because of the intervening 
period of compliance, the March and July surveys resulted in a double 
G, for which immediate sanctions would have been warranted. 

Figure 7: Impact of Intervening Periods of Compliance Rule on Immediate 
Sanctions for One Pennsylvania Nursing Home, 2000: 

[See PDF for image] 

Source: GAO analysis of OSCAR and PDQ. 

[End of figure] 

Clearing effect of standard surveys. Under the double G immediate 
sanctions policy, a standard survey without a G-level or higher 
deficiency "clears the home's record" for the purposes of determining 
whether a double G occurred.[Footnote 77] As a result of this rule, 
surveys with G-level or higher deficiencies that occurred before the 
standard survey without a G-level or higher deficiency are not 
considered in determining whether a double G should be cited and an 
immediate sanction should be imposed. CMS officials believe that it is 
appropriate for standard surveys without G-level or higher deficiencies 
to clear the home's record for double G purposes because standard 
surveys are comprehensive and occur regularly. Yet, we have previously 
reported that weaknesses in the survey process result in surveyors' 
missing serious deficiencies on standard surveys.[Footnote 78] 
Moreover, variability among states in the citation of serious 
deficiencies suggests that some states may not be citing deficiencies 
at the appropriate scope and severity (see app. II). For example, 
according to California officials, the guidance the state received from 
the CMS regional office created confusion as to what constituted actual 
harm, and this confusion contributed to the decline in citations of 
serious deficiencies in California. The regional office clarified its 
guidance in late 2004. 

The following example illustrates how a standard survey without G-level 
or higher deficiencies affects double G determinations and how having 
uncorrected deficiencies can prevent a home from receiving an immediate 
sanction.[Footnote 79] 

* In approximately a 12-month period, a Michigan home had five surveys, 
four of which had one G-level deficiency. However, the G-level 
deficiencies triggered double G immediate sanctions only once instead 
of three times because in one instance a standard survey cited no G- 
level deficiencies and in the other there was no intervening period of 
compliance.[Footnote 80] Figure 8 illustrates how some pairs of surveys 
with G-level deficiencies do not count as double Gs because of the 
clearing effect of standard surveys. For example, state surveyors found 
a G-level deficiency during a January 2000 complaint survey. However, 
on the home's standard survey a month later (February 2000), no G-level 
or higher deficiencies were found by surveyors. As a result, when 
surveyors found another G-level deficiency on a complaint survey 
several months later (November 2000), the G-level deficiency on the 
home's January survey was not considered, and no immediate sanctions 
were triggered. The pair of surveys in January 2000 and November 2000 
did not trigger immediate sanctions because, in effect, the February 
2000 standard survey cleared the home's record. 

Figure 8: Impact of Clearing Effect Rule on Immediate Sanctions for One 
Michigan Nursing Home, 2000-2002: 

[See PDF for image] 

Source: GAO analysis of OSCAR and PDQ. 

[End of figure] 

Multiple instances of harm. Multiple G-level or higher deficiencies 
identified on a survey that results in an immediate sanction are 
sometimes treated the same, in terms of enforcement, as a single 
instance of harm or immediate jeopardy cited on a survey. We examined 
the sanctions imposed for a single versus multiple instances of harm 
and found that the sanctions can be quite similar, despite the 
significant differences in the number of deficiencies.[Footnote 81] The 
following example involves two surveys of a Michigan home with a 
history of repeated noncompliance. On a survey with only 1 G-level 
deficiency, CMS implemented a $350 per day CMP and a discretionary 
DPNA. On a different survey with 33 D-level or higher deficiencies and 
6 G-level or higher deficiencies, CMS implemented a $200 per day CMP 
and a discretionary DPNA. We found similar examples among other homes 
we reviewed. 

We discussed our concerns with CMS about how the double G immediate 
sanctions policy allows some homes to avoid immediate sanctions. CMS 
officials stated that regardless of the policy, state and regional 
office officials retain the discretion to impose immediate sanctions 
even when not required by the policy. However, based on a discussion 
with CMS officials, we believe that, instead of imposing sanctions of 
appropriate severity, state and regional office officials may impose 
weaker sanctions for problem homes that have escaped immediate 
sanctions because of the complexities of the policy. CMS agreed that 
this could happen. 

CMS Oversight Continues to Be Hampered by Data Limitations: 

Fragmented data systems and incomplete national reporting capabilities 
continue to hamper CMS's ability to track and monitor enforcement. In 
March 1999, we reported that CMS lacked a system for effectively 
integrating enforcement data nationwide and that the lack of such a 
system weakened oversight.[Footnote 82] Since 1999, CMS has made 
progress in developing an enforcement data collection system called the 
ASPEN Enforcement Manager (AEM). However, while AEM collects valuable 
data from the states and regions, it is not fully integrated with other 
CMS systems used to track nursing home survey and enforcement 
activities. For example, when regional and state survey officials want 
to evaluate complaint and enforcement data, they must access one system 
for complaint data and then access another system, AEM, for enforcement 
data. Because there is no direct interface between the two systems, CMS 
and states must rely on fragmented data systems for tracking and 
monitoring enforcement. Furthermore, CMS officials told us that the 
agency does not have a concrete plan to use the enforcement data to 
improve monitoring and oversight but that some national enforcement 
reports are under development. 

From 2000 to 2004, CMS tracked sanctions with LTC, a data system 
developed in the Chicago region that became operational in all 10 CMS 
regions in 2000. LTC was a relatively simple system designed to collect 
sanctions data, automatically generate sanction imposition letters, and 
automatically calculate the 35 percent reduction in CMPs for homes that 
waive the right to appeal deficiencies. LTC was not always useful for 
enforcement oversight because it was sometimes incomplete. Data entry 
into the LTC system was optional, and many regional and state surveyors 
continued to rely on their own, state-specific tracking systems. 
Moreover, during the time LTC was in use, states and regions were 
expected to continue updating the enforcement component of OSCAR, which 
duplicated some of the information in LTC. This required separate 
manual data entry into both LTC and OSCAR. We were told by regional 
office officials that sometimes only one of the files would be updated. 
Furthermore, LTC had no internal quality control checks for ensuring 
all fields were completed or that the data were accurate; in its design 
of LTC, CMS chose flexibility in modifying the data to accommodate 
special circumstances over a more rigid field edits system that would 
have controlled the data more tightly. 

Since October 1, 2004, CMS has used AEM to collect state and regional 
data on sanctions and improve communications between state survey 
agencies and CMS regional offices. Specifically, AEM was designed to 
provide real-time entry and tracking of sanctions, issue monitoring 
alerts, generate enforcement letters, and facilitate analysis of 
enforcement patterns. CMS expects that the data collected in AEM will 
enable states, CMS regional offices, and the CMS central office to more 
easily track and evaluate sanctions against nursing homes as well as 
respond to emerging issues. Developed by CMS's central office primarily 
for use by states and regions, AEM is one module of a broader data 
collection system called ASPEN. There are a number of other modules 
under the ASPEN umbrella, including the ASPEN Complaints/Incidents 
Tracking System (ACTS) module. The ASPEN modules--and other data 
systems related to enforcement such as the financial management system 
for tracking CMP collections--are fragmented and lack automated 
interfaces with each other. As a result, enforcement officials must 
pull discrete bits of data from the various systems and manually 
combine the data to develop a full enforcement picture. For example, if 
regional office officials want to review a home's complaint history, 
they must access ACTS to print a report on complaints, access AEM to 
print a report on corresponding sanctions, manually compare the two 
reports, and then access the CMP tracking system to determine whether a 
corresponding CMP was paid. Each step adds to staff workload. 

AEM collects potentially useful enforcement data from the states and 
regions, but, as described, CMS has not integrated AEM with the other 
data collection systems (e.g., ACTS); furthermore, the agency has not 
defined a plan for using the AEM data to inform the tracking and 
monitoring of enforcement through national enforcement reports. In a 
December 2004 CMS report, the agency stated that AEM "will permit 
meaningful comparisons of like measures and will serve as a primary 
tool on which to base policy decisions, new initiatives and strategies 
for improving care to our Nation's nursing home population."[Footnote 
83] While CMS is developing a few draft national enforcement reports, 
it has not developed a concrete plan and timeline for producing a full 
set of reports that use the AEM data to help in assessing the 
effectiveness of sanctions and its enforcement policies. In addition, 
while the full complement of enforcement data recorded by the states 
and regional offices in AEM is now being uploaded to CMS's national 
system, CMS does not intend to upload any historical data. Efforts to 
track and monitor enforcement would be greatly enhanced by reports that 
contain the historical data; for example, with historical data the 
agency could generate reports that provide a longitudinal perspective 
of a home's compliance history, compare trends across states and 
regions, and, overall, help evaluate the effectiveness of sanctions and 
policies. Finally, like LTC, AEM has quality control weaknesses. While 
AEM has some automatic quality control mechanisms to ensure that the 
data entered are complete and in a valid format, there are no 
systematic quality control mechanisms to ensure that the data entered 
are accurate. For example, while the system automatically requires the 
entry of valid survey dates, CMS does not conduct periodic data audits 
to check that the survey dates are correct. 

CMS officials told us they will continue to develop and implement 
enhancements to AEM to expand its capabilities over the next several 
years. However, until CMS develops a plan for integrating the 
fragmented systems and for using AEM data--along with other data the 
agency collects--efficient and effective tracking and monitoring of 
enforcement will continue to be hampered and, as a result, CMS will 
have difficulty assessing the effectiveness of sanctions and its 
enforcement policies. 

Other CMS Initiatives to Improve Enforcement: 

In addition to its efforts to implement a new data system for managing 
enforcement, CMS has taken other steps to improve its enforcement of 
nursing home quality requirements. For example, the agency has 
developed guidance to help ensure greater consistency across states in 
CMP amounts imposed, revised its Special Focus Facility program, and 
commissioned two studies to examine the effectiveness of nursing home 
enforcement.[Footnote 84] 

To ensure greater consistency in CMP amounts proposed by states and 
imposed by regions, CMS, in conjunction with state survey agencies, 
developed a grid that provides guidance for states and regions. The CMP 
grid lists ranges for minimum CMP amounts while allowing for 
flexibility to adjust the penalties on the basis of factors such as the 
deficiency's scope and severity, the care areas where the deficiency 
was cited, and a home's past history of noncompliance. In August 2006, 
CMS completed the regional office pilot of its CMP grid. The results of 
the pilot, which are currently being analyzed, will be used to 
determine how the grid should be used by states; its use would be 
optional to provide states flexibility to tailor sanctions to specific 
circumstances. 

CMS revised its Special Focus Facility program, an initiative intended 
to increase the oversight of homes with a history of providing poor 
care. We had previously reported that the program was worthwhile but 
that its narrow scope excluded many homes that provide poor 
care.[Footnote 85] Moreover, according to CMS, the goal of two surveys 
per home per year was never achieved because of the relatively low 
priority assigned to the program and the lack of state survey agency 
resources. In December 2004, CMS announced three changes in the 
operation of the program. First, CMS expanded the scope of the program 
from about 100 homes nationwide to about 135 homes by making the number 
of Special Focus Facilities in each state proportional to the number of 
nursing homes. Second, CMS revised the method for selecting nursing 
homes by reviewing 3 years' rather than 1 year's worth of deficiency 
data. This change was intended to ensure that the homes in the program 
had a history of noncompliance rather than a single episode of 
noncompliance. Third, CMS strengthened its enforcement for Special 
Focus Facilities by requiring immediate sanctions for homes that failed 
to significantly improve their performance from one survey to the next 
and by requiring termination for homes with no significant improvement 
after three surveys over an 18-month period.[Footnote 86] Despite these 
changes, however, many homes that could benefit from enhanced oversight 
and enforcement are still excluded from the program. As noted earlier, 
few of the homes we reviewed were or are part of CMS's Special Focus 
Facilities program. In 2005, only 2 were designated Special Focus 
Facilities and in 2006, the number increased to 4. Of the 8 homes that 
cycled in and out of compliance seven or more times (see fig. 3), 6 are 
still open but only 1 is now a Special Focus Facility. Although CMS now 
requires QIOs to work with poorly performing nursing homes, this 
initiative also only targets a small number of homes--as few as 1 to 3 
facilities in each state. 

To enhance its understanding of and ability to improve the enforcement 
process, CMS has funded two studies that will examine the steps that 
lead to sanctions as well as the impact of enforcement on homes' 
quality-of-care processes. 

* Qualitative Enforcement Case Studies. This study, which began in the 
spring of 2003 and is scheduled to be completed in early 2007, required 
research nurses to visit 25 nursing homes in four states to evaluate 
how the survey and enforcement processes are carried out and assess the 
extent to which the enforcement process results in changes in nursing 
staff behavior and improved compliance with federal requirements. 

* Impact of Sanctions on Quality. The objective of this study is to 
test the effects of sanctions on facility behavior and resident 
outcomes. Researchers will identify and compare a group of nursing 
homes that had both deficiencies and sanctions to a group of nursing 
homes that had similar levels of deficiencies but no sanctions. A year 
later, researchers will review the nursing home's subsequent survey to 
determine whether the sanctions resulted in any significant changes in 
the quality of care delivered. The study began in the fall of 2004 and 
the first report is scheduled to be completed by mid-2007. 

Although CMS has taken several steps to improve its enforcement of 
nursing home requirements, its Nursing Home Compare Web site does not 
include information on sanctions. Thus, CMS does not indicate what 
sanctions have been implemented against nursing homes, nor does it 
identify homes that have received immediate sanctions for repeatedly 
harming residents. 

As noted throughout this report, we found variation among the states we 
reviewed in areas such as the number and amount of CMPs implemented and 
the proportion of homes with double Gs. In general, these differences 
reflect the state survey agencies' views on the effectiveness of 
certain sanctions and differences in state enforcement policies. For 
example, Pennsylvania state officials prefer state rather than federal 
sanctions because they believe the former are more effective, have a 
greater deterrent effect on providers, and are easier and quicker to 
impose. Pennsylvania requires homes to pay a state CMP prior to appeal, 
even if the home appeals the deficiency. In contrast, homes need not 
pay a federal CMP until after an appeal is resolved. Pennsylvania 
rarely implemented federal CMPs on the 14 state homes whose compliance 
history we reviewed, preferring to use state sanctions instead. In 
Michigan, state officials are more likely to use federal CMPs and 
implement them in greater amounts than other states we reviewed. Texas 
state officials often use state rather than federal sanctions for G- 
level or higher deficiencies, in part because they cannot propose a 
federal CMP if they impose a state sanction and because the total state 
money penalty that may be imposed may be higher than federal CMPs. 
California had fewer sanctions than Michigan. California typically 
investigates complaints under its state licensure authority, which may 
partly explain why California has fewer reported deficiencies and 
federal sanctions. We believe it is important for CMS to explore the 
differences in state enforcement approaches and policies so that it can 
both identify problem areas and identify best practices that could be 
disseminated nationwide. 

Conclusions: 

Although CMS has taken steps to strengthen the nursing home enforcement 
process, our review of 63 homes in four states with a history of 
quality problems identified design weaknesses as well as flaws in the 
way sanctions are implemented that diminish their full deterrent 
effect. Some of these homes repeatedly harmed residents over a 6-year 
period and yet remain in the Medicare and Medicaid programs. Until 
these systemic weaknesses are addressed, the effectiveness of sanctions 
in encouraging homes to return to and maintain compliance will remain 
questionable and the safety and security of vulnerable residents will 
remain at risk. 

CMS's immediate sanctions policy fails to hold homes with a long 
history of harming residents accountable for the poor care provided. 
The policy's complexity, such as the requirement for an intervening 
period of compliance, prevents its use for the very homes it was 
designed to address--those with systemic quality problems. Furthermore, 
the immediate sanctions label is misleading because sanctions are not, 
in fact, immediate. The notice period required by CMS regulations for 
sanctions such as DPNAs and terminations provides homes with a de facto 
grace period during which they can correct deficiencies to avoid an 
immediate sanction. Moreover, in one state we reviewed, the immediate 
sanctions policy does not fully identify all homes with repeat serious 
deficiencies because most complaint deficiencies, which can often 
trigger a double G, were being cited under state licensure authority, 
not federal. Consequently, some problem homes in the state were not 
identified by the policy and thus were able to avoid double G immediate 
sanctions. 

Although CMPs and DPNAs were the most frequently used sanctions 
nationwide and for the homes we reviewed, their effectiveness was 
undermined by a number of weaknesses. The CMPs levied against the homes 
we reviewed were often nominal, significantly less than the maximum 
amounts Congress provided for in statute. To strengthen CMPs, CMS has 
been developing a CMP grid since 2004 to guide states and regional 
offices in determining appropriate CMP amounts, and CMS regional 
offices piloted the grid in 2006. However, its implementation is 
expected to be optional for states, once again contributing to 
interstate variation. Despite the nominal amounts, CMPs, unlike DPNAs, 
do not require a notice period and may be imposed retroactively before 
the date of the survey. However, these advantages are countered by the 
fact that, under the Social Security Act, payment by homes of federally 
imposed CMPs is deferred if they appeal their deficiencies, a process 
that can take years, diminishing the immediacy of the sanction and 
further undermining the sanction's deterrent effect. While there is 
precedent under the federal surface mining statute, which permits the 
collection of CMPs before exhaustion of appeals, it is unclear if the 
informal dispute resolution process available to nursing homes provides 
the same type of procedural safeguards that courts have pointed to in 
upholding the mining statute provision. Some states choose to use their 
own authority to impose state fines, which can sometimes be implemented 
faster than is possible under federal law. Although CMS has the 
authority to implement discretionary DPNAs after a 15-day notice period 
for the homes we reviewed, it did not generally do so. It imposes 
mandatory DPNAs when criteria are met, which provide homes a 3-month de 
facto grace period to correct deficiencies. Because many homes we 
reviewed returned to compliance within 3 months--though often only 
temporarily--the DPNAs frequently were rescinded. 

Termination--the most powerful enforcement tool--was used infrequently 
nationwide and for the homes we reviewed because of states' and CMS's 
concerns about potential access to care and resident transfer trauma. 
However, we found that some poorly performing homes are located in 
areas with several other nearby nursing homes. Even though some homes 
we reviewed cycled in and out of compliance numerous times while 
continuing to harm residents, CMS allowed them to determine for 
themselves whether and when to leave the Medicare and Medicaid 
programs. Even when terminations were imposed, their deterrent effect 
was undermined by extending some termination dates to give the homes 
more time to correct deficiencies. CMS's earlier termination of such 
troubled homes could have cut short the cycle of poor care. CMS's 
revamped Special Focus Facility program would provide for termination 
of poorly performing homes within 18 months if they fail to show 
significant improvement in the quality of care provided to residents. 
Despite the expansion of the program from about 100 to about 135 homes, 
the number of Special Focus Facilities is inadequate because, as our 
work has demonstrated, the program still fails to include many homes 
with a history of repeatedly harming residents. 

Although CMS has made progress in establishing a database to help it 
track and monitor the nursing home enforcement process, the development 
of AEM is not yet complete. AEM is not integrated with other important 
databases to help ensure that CMS has a comprehensive picture of a 
home's deficiency history, and CMS has not developed a concrete plan 
for using national enforcement reports--built off of AEM data--to help 
evaluate the effectiveness of sanctions and its enforcement polices. 
Having longitudinal enforcement data available for homes would enable 
CMS to pursue increasing the severity of sanctions for homes that 
repeatedly harm residents. Furthermore, CMS has not developed a system 
of quality checks to ensure the accuracy and integrity of AEM data. 

CMS's Nursing Home Compare Web site has been modified a number of times 
to add important quality information about nursing homes. While CMS now 
summarizes the results from both standard surveys and complaint 
investigations, the Web site contains no information about sanctions 
implemented against nursing homes, nor does it identify homes that have 
received immediate sanctions for repeatedly harming residents. Such 
information could be valuable to consumers who use the Web site to help 
choose a home for family members or friends. 

Recommendations for Executive Action: 

To address weaknesses that undermine the effectiveness of the immediate 
sanctions policy, we recommend that the Administrator of CMS reassess 
and revise the policy to ensure that it accomplishes the following 
three objectives: (1) reduce the lag time between citation of a double 
G and the implementation of a sanction, (2) prevent nursing homes that 
repeatedly harm residents or place them in immediate jeopardy from 
escaping sanctions, and (3) hold states accountable for reporting in 
federal data systems serious deficiencies identified during complaint 
investigations so that all complaint findings are considered in 
determining when immediate sanctions are warranted. 

To strengthen the deterrent effect of available sanctions and to ensure 
that sanctions are used to their fullest potential, we recommend that 
the Administrator of CMS take the following three actions: 

* Ensure the consistency of CMPs by issuing guidance such as the 
standardized CMP grid piloted during 2006. 

* Increase use of discretionary DPNAs to help ensure the speedier 
implementation of appropriate sanctions. 

* Strengthen the criteria for terminating homes with a history of 
serious, repeated noncompliance by limiting the extension of 
termination dates, increasing the use of discretionary terminations, 
and exploring alternative thresholds for termination, such as the 
cumulative duration of noncompliance. 

To collect CMPs more expeditiously, which could increase their 
deterrent effect, we recommend that the Administrator of CMS develop an 
administrative process under which CMPs would be paid--or Medicare and 
Medicaid payments in equivalent amounts would be withheld--prior to 
exhaustion of appeals and seek legislation for the implementation of 
this process, as appropriate. Payments could be refunded with interest 
if the deficiencies are modified or overturned at appeal. 

To strengthen sanctions for homes with a history of noncompliance, such 
as a large number of deficiencies or a large number of actual harm and 
immediate jeopardy deficiencies, we recommend that the Administrator of 
CMS consider further expanding the Special Focus Facility program with 
its enhanced enforcement requirements to include all homes that meet a 
threshold, established by CMS, to qualify as poorly performing homes. 

To improve the effectiveness of its new enforcement data system, we 
recommend that the Administrator of CMS take the following three 
actions: 

* Develop the enforcement-related data systems' abilities to interface 
with each other in order to improve the tracking and monitoring of 
enforcement, such as by developing an automatic interface between 
systems such as AEM and ACTS. 

* Expedite the development of national enforcement reports, including 
longitudinal and trend reports designed to evaluate the effectiveness 
of sanctions and enforcement policies, and a concrete plan for using 
the reports. 

* Develop and institute a system of quality checks to ensure the 
accuracy and integrity of AEM data, such as periodic data audits 
conducted as part of CMS's annual state performance reviews. 

To improve public information available to consumers that helps them 
assess the quality of nursing home care, we recommend that the 
Administrator of CMS expand CMS's Nursing Home Compare Web site to 
include implemented sanctions, such as the amount of CMPs and the 
duration of DPNAs, and homes subjected to immediate sanctions. 

Agency and State Comments and Our Evaluation: 

We obtained written comments on our draft report from CMS and three of 
the four states in which the homes we studied were located--California, 
Michigan, and Texas. We also received e-mail comments from the Director 
of the Division of Nursing Care Facilities in Pennsylvania. CMS's 
comments are reproduced in appendix VI. California's, Michigan's and 
Texas's comments are reproduced in appendixes VII, VIII, and IX, 
respectively. CMS generally concurred with our 12 recommendations in 
six areas intended to strengthen the enforcement process but did not 
always specify how it would implement the recommendations. In addition, 
CMS noted that implementation of 3 of our recommendations raised 
resource issues and that others required additional research. 
California concurred with our conclusions and recommendations, while 
Michigan and Pennsylvania indicated appreciation or general agreement. 
However, most state comments, including Texas's, were technical in 
nature. Our evaluation responds to CMS and state comments in the six 
areas covered by our recommendations. 

Addressing weaknesses in the double G immediate sanctions policy. CMS 
agreed that homes that repeatedly harm residents should not escape 
immediate sanctions and stated that it would remove the limitation on 
applying an additional sanction when a home failed to correct a 
deficiency that gave rise to a prior sanction. CMS also agreed to 
reduce the lag time between citation and implementation of a double G 
immediate sanction by limiting the prospective effective date for DPNAs 
to no more than 30 to 60 days. Reducing the lag time as much as 
possible is critical because it provides homes with a de facto grace 
period in which to correct deficiencies and avoid sanctions. Michigan 
commented about the need to increase the immediacy of DPNAs, noting 
that even the 15-day notice period associated with discretionary DPNAs 
was outdated now that homes are notified electronically and delivery 
can be verified. Currently, CMS has an incomplete picture of serious 
deficiencies cited against homes that could result in immediate 
sanctions because California investigates many nursing home complaints 
under state licensure authority. CMS agreed to collect additional 
information on complaints for which data are not reported in federal 
data systems. We believe that CMS's commitment to do this will help 
better identify and deal with consistently poorly performing homes. CMS 
commented that the Social Security Act does not provide authority for 
CMS to require states to report enforcement actions taken under state- 
only authority if federal resources are not used for the complaint 
investigation; however, to the extent that federal funds are used for 
complaint investigations, our findings and recommendations remain 
valid. Michigan concurred that CMS needs the complete compliance 
history of a facility to assess its overall performance. 

CMS acknowledged that the complexity of its immediate sanctions policy 
may be an inherent limitation and indicated that it intends to either 
strengthen the policy or replace it with a policy that achieves similar 
goals through alternative methods. CMS noted that it is concerned about 
whether the immediate sanctions policy has negatively affected the 
rates of state deficiency citations and may ultimately be ineffective 
with the most problematic facilities. We believe the policy has merit 
but that its complex requirements have prevented many homes from 
receiving immediate sanctions. 

Strengthening the deterrent effect of sanctions. CMS agreed to issue a 
CMP analytic tool, or grid, and to provide states with further guidance 
on discretionary DPNAs and terminations. The CMP grid is a tool to help 
ensure national consistency in CMPs and to assist CMS regional offices 
in monitoring enforcement actions. Texas commented that it had been 
using the grid since June 2006 and found it to be very helpful. 
Michigan noted that it had independently developed and implemented a 
CMP grid in 2000 but expressed disappointment that CMS had not mandated 
state use of the agency's grid. In addition, Michigan supported the 
need for additional CMS guidance on the use of discretionary 
termination. Such guidance, it commented, was necessary to ensure a 
consistent national approach. In response to our recommendation to 
increase the use of discretionary terminations, CMS stated that it will 
continue its research to design proposals that yield a more effective 
combination of robust enforcement actions but that do not penalize 
vulnerable residents. While we encourage CMS's commitment to further 
research to improve the effectiveness of enforcement actions, we 
believe that CMS must also be committed to protecting residents from 
actual harm in poorly performing facilities--including terminating 
homes from the Medicare or Medicaid programs--when other steps fail to 
ensure the quality of resident care. 

Collecting CMPs more expeditiously. CMS agreed to seek legislative 
authority to collect CMPs prior to the exhaustion of appeals, which 
could increase their deterrent effect. California commented that it 
supported this recommendation. 

Expanding the Special Focus Facility program. CMS agreed with the 
concept of expanding the program to include all homes that meet a 
threshold to qualify as poorly performing homes, but said it lacks the 
resources needed for this expansion because of decreases in its budget 
and increases in both the number of providers and quality assurance 
responsibilities for state and federal surveyors. CMS stated that it 
envisioned expansion of the program if Congress fully funds the 
President's proposed fiscal year 2008 budget for survey and 
certification activities. CMS specified other initiatives it will 
implement to improve the Special Focus Facility program. 

Improving the effectiveness of enforcement data. CMS agreed to develop 
and implement a system of quality checks to ensure the accuracy of its 
data systems, including AEM. While the agency agreed to study the 
feasibility of linking the separate data systems used for enforcement 
and to develop other national standard enforcement reports, CMS 
indicated that available resources may limit its ability to take 
further action on these issues. CMS has already invested significant 
resources in developing potentially powerful data systems intended to 
improve the tracking and monitoring of enforcement, and we believe the 
agency should place a priority on ensuring that these systems operate 
effectively. 

Improving information available to consumers. Rather than agreeing to 
report all implemented sanctions on its Nursing Home Compare Web site, 
CMS proposed reporting implemented sanctions only for poorly performing 
homes that meet an undefined threshold. CMS's response was therefore 
not fully responsive to our recommendation. By only reporting sanctions 
for homes that meet a certain threshold--eight or more sanctions in a 3-
year period, in an example provided by CMS--consumers might incorrectly 
assume that other homes have received no sanctions. Furthermore, CMS's 
plan to post such limited sanctions data in an accessible location on 
its Web site is vague. We believe that consumers must be able to easily 
link deficiency and sanctions data. 

CMS and three of the four states also provided technical comments, 
which we incorporated as appropriate. 

As arranged with your office, unless you publicly announce its contents 
earlier, we plan no further distribution of this report until 30 days 
after its issue date. At that time, we will send copies to the 
Administrator of the Centers for Medicare & Medicaid Services and 
appropriate congressional committees. We will also make copies 
available to others upon request. In addition, the report will be 
available at no charge on the GAO Web site at http://www.gao.gov. 

If you or your staff 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 X. 

Sincerely yours, 

Signed by: 

Kathryn G. Allen: 
Director, Health Care: 

[End of section] 

Appendix I: Scope and Methodology: 

This appendix provides a more detailed description of our scope and 
methodology and generally follows the order that findings appear in the 
report. We analyzed the fiscal years 2000 through 2005 enforcement and 
deficiency history for a total of 63 of the 74 nursing homes in four 
states--California, Michigan, Pennsylvania, and Texas--whose compliance 
history informed the conclusions of our March 1999 report.[Footnote 87] 
These homes had a history of providing poor quality care to residents 
prior to 1999. We excluded 11 of the original 74 homes from our 
analysis because they either closed before fiscal year 2000 or closed 
within 6 months of the beginning of fiscal year 2000 and had few or no 
deficiencies or sanctions.[Footnote 88] Some of the remaining 63 homes 
participated in the Medicare and Medicaid programs for only a portion 
of fiscal years 2000 through 2005 because they either closed 
permanently or closed temporarily and were subsequently reinstated. For 
these homes, we set a criterion that required that the home participate 
for at least 6 months of the fiscal year in order for its enforcement 
data in that fiscal year to be included in our analysis. Table 8 shows 
the distribution of homes across the four states in our 1999 report, 
the distribution of those homes for this report, and the number of 
providers participating for at least 6 months by fiscal year. Although 
the table shows some year-to- year fluctuation in the number of 
providers, the changes do not significantly influence our findings. 
While the focus of our analysis was the compliance history of these 63 
homes, we also analyzed general trends in (1) implemented sanctions 
nationwide for the same 6-year period and (2) the proportion of homes 
in each state cited for serious deficiencies--that is, those at the 
actual harm or immediate jeopardy level. 

Table 8: Number of Nursing Homes Reviewed in 1999 That Were Included in 
Our Analysis for This Report, by State: 

State: California; 
1999 report: 12; 
Current report: Any participation in fiscal years 2000-2005: 10; 
Current report: 2000: 10; 
Current report: 2001: 10; 
Current report: 2002: 10; 
Current report: 2003: 10; 
Current report: 2004: 10; 
Current report: 2005: 10. 

State: Michigan; 
1999 report: 18; 
Current report: Any participation in fiscal years 2000-2005: 16; 
Current report: 2000: 14; 
Current report: 2001: 14; 
Current report: 2002: 13; 
Current report: 2003: 13; 
Current report: 2004: 12; 
Current report: 2005: 13. 

State: Pennsylvania; 
1999 report: 17; 
Current report: Any participation in fiscal years 2000-2005: 14; 
Current report: 2000: 14; 
Current report: 2001: 13; 
Current report: 2002: 13; 
Current report: 2003: 12; 
Current report: 2004: 10; 
Current report: 2005: 10. 

State: Texas; 
1999 report: 27; 
Current report: Any participation in fiscal years 2000-2005: 23; 
Current report: 2000: 23; 
Current report: 2001: 22; 
Current report: 2002: 22; 
Current report: 2003: 23; 
Current report: 2004: 22; 
Current report: 2005: 22. 

State: Total; 
1999 report: 74; 
Current report: Any participation in fiscal years 2000-2005: 63; 
Current report: 2000: 61; 
Current report: 2001: 59; 
Current report: 2002: 58; 
Current report: 2003: 58; 
Current report: 2004: 54; 
Current report: 2005: 55. 

Source: GAO. 

[End of table] 

CMS deficiency data. To determine the number, scope, and severity of 
deficiencies cited for the 63 homes, we analyzed OSCAR (On-Line Survey, 
Certification, and Reporting system) deficiency data resulting from 
standard surveys and complaint investigations. We also used OSCAR data 
on deficiencies identified during standard surveys to analyze state 
trends in the proportion of nursing homes cited for actual harm or 
immediate jeopardy during fiscal years 2000 through 2005.