United States General Accounting Office GAO Report to Congressional Requesters July 2003 NURSING HOME QUALITY Prevalence of Serious Problems, While Declining, Reinforces Importance of Enhanced Oversight GAO-03-561 July 2003 NURSING HOME QUALITY Prevalence of Serious Problems, While Highlights of GAO-03-561, a report to Declining, Reinforces Importance of congressional requesters Enhanced Oversight Since July 1998, GAO has reported The proportion of nursing homes with serious quality problems remains numerous times on nursing home unacceptably high, despite a decline in the incidence of such reported quality-of-care issues and identified problems. Actual harm or more serious deficiencies were cited for 20 significant weaknesses in federal percent or about 3,500 nursing homes during an 18-month period ending and state oversight. GAO was January 2002, compared to 29 percent for an earlier period. Fewer asked to assess the extent of the progress made in improving the discrepancies between federal and state surveys of the same homes suggests quality of care provided by nursing that state surveyors are doing a better job of documenting serious homes to vulnerable elderly and deficiencies and that the decline in serious quality problems is potentially disabled individuals, including real. Despite these improvements, the continuing prevalence of and state (1) trends in measured nursing surveyor understatement of actual harm deficiencies is disturbing. For home quality, (2) state responses to example, 39 percent of 76 state surveys from homes with a history of quality- previously identified weaknesses in of-care problems—but whose current survey found no actual harm their survey, complaint, and deficiencies—had documented problems that should have been classified as enforcement activities, and (3) the actual harm or higher, such as serious, avoidable pressure sores. status of oversight and quality improvement efforts by the Centers Weaknesses persist in state survey, complaint, and enforcement activities. for Medicare & Medicaid Services (CMS). According to CMS and states, several factors contribute to the understatement of serious quality problems, including poor investigation and documentation of deficiencies, limited quality assurance systems, and a large number of inexperienced surveyors in some states. In addition, GAO found that about one-third of the most recent state surveys nationwide remained GAO is making several predictable in their timing, allowing homes to conceal problems if they recommendations to the chose to do so. Considerable state variation remains regarding the ease of Administrator of CMS to filing a complaint, the appropriateness of the investigation priorities, and the (1) strengthen the nursing home timeliness of investigations. Some states attributed timeliness problems to survey process, (2) ensure that state survey and complaint inadequate staff and an increase in the number of complaints. Although the activities adequately assess quality- agency strengthened enforcement policy by requiring states to refer for of-care problems, and (3) improve immediate sanction homes that had repeatedly harmed residents, GAO CMS oversight of state survey found that states failed to refer a substantial number of such homes, activities. CMS concurred with the significantly undermining the policy’s intended deterrent effect. report’s recommendations, but its comments on intended actions CMS oversight of state survey activities has improved but requires continued were not fully responsive to all of attention to help ensure compliance with federal requirements. While CMS the recommendations. Eleven strengthened oversight by initiating annual state performance reviews, states provided comments that officials acknowledged that the reviews’ effectiveness could be improved. most often focused on the resource For the initial fiscal year 2001 review, officials said they lacked the capability constraints states face in meeting federal standards for oversight of to systematically distinguish between minor lapses and more serious nursing homes. problems that required intervention. CMS oversight is also hampered by continuing database limitations, the inability of some CMS regions to use available data to monitor state activities, and inadequate oversight in areas such as survey predictability and state referral of homes for enforcement. Three key CMS initiatives have been significantly delayed—strengthening www.gao.gov/cgi-bin/getrpt?GAO-03-561. the survey methodology, improving surveyor guidance for determining the To view the full product, including the scope scope and severity of deficiencies, and producing greater standardization in and methodology, click on the link above. state complaint processes. These initiatives are critical to reducing the For more information, contact Kathryn G. Allen at (202) 512-7118. subjectivity evident in current state survey and complaint activities. Contents Letter 1 Results in Brief 3 Background 6 Magnitude of Problems Remains Cause for Concern Even Though Fewer Serious Nursing Home Quality Problems Reported 11 Weaknesses Persist in State Survey, Complaint, and Enforcement Activities 18 CMS Oversight of State Survey Activities Requires Further Strengthening 29 Conclusions 40 Recommendations for Executive Action 42 Agency and State Comments and Our Evaluation 43 Appendix I Scope and Methodology 51 Appendix II Trends in The Proportion of Nursing Homes Cited for Actual Harm or Immediate Jeopardy Deficiencies, 1997-2002 55 Appendix III Abstracts of Nursing Home Survey Reports That Understated Quality-of-Care Problems 58 Appendix IV Information on State Nursing Home Surveyor Staffing 78 Appendix V Predictability of Standard Nursing Home Surveys 80 Appendix VI Immediate Sanctions Implemented Under CMS’s Expanded Immediate Sanctions Policy 83 Page i GAO-03-561 Nursing Home Quality Appendix VII Cases States Did Not Refer to CMS for Immediate Sanction 86 Appendix VIII HCFA State Performance Standards for Fiscal Year 2001 88 Appendix IX Highlights of State Compliance with CMS Performance Standards 90 Appendix X Comments from the Centers for Medicare & Medcaid Services 91 Appendix XI GAO Contact and Staff Acknowledgements 95 GAO Contact 95 Acknowledgements 95 Related GAO Products 96 Tables Table 1: Scope and Severity of Deficiencies Identified During Nursing Home Surveys 8 Table 2: Change in the Percentage of Nursing Homes Cited for Actual Harm or Immediate Jeopardy during State Standard Surveys between the periods January 1, 1999, through July 10, 2000, and July 11, 2000, through January 31, 2002, by State 13 Table 3: Incidence of Underreported Actual Harm Deficiencies in Surveys GAO Reviewed 17 Table 4: Predictability of Nursing Home Surveys 22 Table 5: Key Findings of Report to CMS on State Complaint Investigation Processes 25 Page ii GAO-03-561 Nursing Home Quality Table 6: Quality of Care Requirements Reviewed in a Sample of State Survey Reports 52 Table 7: Trends in the Percentage of Nursing Homes Cited for Actual Harm or Immediate Jeopardy during State Standard Surveys, by State 56 Table 8: Abstracts of the 39 Nursing Home Deficiencies that Understated Actual Harm from a Sample of 76 Nursing Home Survey Reports 59 Table 9: State Survey Agency Responses to Questions about Surveyor Experience, Vacancies, Hiring Freezes, Competitiveness of Salaries, and Minimum Required Experience 78 Table 10: Predictability of Current Nursing Home Surveys, by State 81 Table 11: Federal Sanctions Implemented against Nursing Homes Referred for Immediate Sanction, January 14, 2000, through March 28, 2002 83 Table 12: Federal CMPs Implemented under CMS’s Immediate Sanctions Policy, January 2000 through March 2002 84 Table 13: Number of Cases States Did Not Refer for Sanction, as Required, and the Number States Appropriately Referred, January 2000 through March 2002 86 Table 14: Overview of HCFA’s Seven State Performance Standards for Nursing Home Survey Activities for Fiscal Year 2001 88 Table 15: State Compliance with Selected CMS Performance Standards, Fiscal Year 2001 90 Figure Figure 1: Four States with the Greatest Number of Cases that Should Have Been Referred for Immediate Sanctions, January 14, 2000, through March 28, 2002 27 Page iii GAO-03-561 Nursing Home Quality Abbreviations ACTS ASPEN Complaint Tracking System CMS Centers for Medicare & Medicaid Services CMP civil money penalties HCFA Health Care Financing Administration MDS minimum data set OSCAR On-Line Survey, Certification, and Reporting system RN registered nurse 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. It may contain copyrighted graphics, images or other materials. Permission from the copyright holder may be necessary should you wish to reproduce copyrighted materials separately from GAO’s product. Page iv GAO-03-561 Nursing Home Quality United States General Accounting Office Washington, DC 20548 July 15, 2003 The Honorable Charles E. Grassley Chairman Committee on Finance United States Senate The Honorable Christopher S. Bond United States Senate A number of congressional hearings since July 1998 have focused considerable attention on the need to improve the quality of care for the nation’s 1.7 million nursing home residents, a highly vulnerable population of elderly and disabled individuals. As we previously reported, poor quality of care at about 15 percent of the nation’s approximately 17,000 nursing homes—an unacceptably high proportion—had repeatedly caused actual harm to residents, such as worsening pressure sores or untreated weight loss, or had placed them at risk of death or serious injury.1 Significant weaknesses in federal and state nursing home oversight that we identified in a series of reports and testimonies since 1998 included (1) periodic state inspections, known as surveys, that understated the extent of serious care problems due to procedural weaknesses, (2) considerable state delays in investigating public complaints alleging harm to residents, (3) federal enforcement policies that did not ensure deficiencies were addressed and remained corrected, and (4) federal oversight of state survey activities that was limited in scope and effectiveness.2 In July 1998, the Health Care Financing Administration (HCFA)—the federal agency with responsibility for managing Medicare and Medicaid and overseeing compliance with federal nursing home quality standards— launched a series of actions intended to address many of the weaknesses we identified.3 Since 1998, the agency has worked to strengthen surveyors’ 1 See U.S. General Accounting Office, Nursing Homes: Proposal to Enhance Oversight of Poorly Performing Homes Has Merit, GAO/HEHS-99-157 (Washington, D.C.: June 30, 1999). 2 A list of related GAO products is at the end of this report. 3 Effective July 1, 2001, HCFA’s name changed to the Centers for Medicare & Medicaid Services (CMS). In this report we continue to refer to HCFA where our findings apply to the organizational structure and operations associated with that name. Page 1 GAO-03-561 Nursing Home Quality ability to detect quality-of-care deficiencies; required states to investigate complaints alleging resident harm within 10 days; mandated immediate sanctions for nursing homes with a pattern of harming residents;4 and begun measuring state compliance with federal survey requirements and reviewing data on the results of state surveys to help pinpoint shortcomings in state survey activities. To evaluate the extent of the progress made in improving the quality of nursing home care since we last addressed this issue in September 2000, you asked us to assess: • trends in measured nursing home quality; • state responses to previously identified weaknesses in their survey, complaint, and enforcement activities; and • the status of key federal efforts to oversee state survey agency performance and improve quality. To assess recent trends in measured nursing home quality, we analyzed survey results for the period July 11, 2000, through January 31, 2002, and compared them to survey results for two earlier 18-month periods: (1) January 1, 1997, through June 30, 1998, and (2) January 1, 1999, through July 10, 2000. Our analysis relied on data from the Centers for Medicare & Medicaid Services’ (CMS) On-Line Survey, Certification, and Reporting (OSCAR) system, which compiles the results of all state nursing home surveys nationwide. To better understand the trends identified through our OSCAR analysis, we analyzed the results of federal comparative surveys, conducted at recently surveyed nursing homes to assess the adequacy of the state surveys, for two time periods—October 1998 through May 2000 and June 2000 through February 2002. We also reviewed 76 survey reports from homes with a history of actual harm deficiencies but whose most recent survey found no such deficiencies in states where the percentage of homes cited for actual harm had declined to below the national average since mid-2000. Our review of deficiencies from these survey reports focused on the types of quality-of-care deficiencies most frequently cited nationwide. 4 The term used in the law and regulations to describe a nursing home penalty for noncompliance is “remedy.” Throughout this report, we use a more common term, “sanction,” to refer to such penalties. Sanctions include actions such as fines, denial of payment for new admissions, and termination from the Medicare and Medicaid programs. Page 2 GAO-03-561 Nursing Home Quality To assess state survey activities as well as federal oversight, we analyzed the conduct and results of fiscal year 2001 state survey agency performance reviews during which CMS regional offices determined state compliance with seven federal standards; we focused on the five standards related to statutory survey intervals, survey documentation, complaint activities, enforcement requirements, and OSCAR data entry. We conducted structured interviews with officials from CMS, CMS’s 10 regional offices, and 16 state survey agencies to discuss trends in survey deficiencies, the underlying causes of problems identified during the performance reviews, and state and federal efforts to address these problems.5 We also discussed these issues with officials from 10 additional states during a governing board meeting of the Association of Health Facility Survey Agencies. We selected the 16 states with the goal of including states that (1) were from diverse geographic areas, (2) had shown either increases or decreases in the percentage of homes cited for actual harm, (3) had been contacted in our prior work, and (4) represented a mixture of strong and weak performance based on the results of federal performance reviews of state survey activities. We also obtained data from most state survey agencies on staffing issues such as nursing home surveyor experience and vacancies. To assess enforcement actions, we analyzed data in CMS’s enforcement database and compared homes identified in OSCAR as requiring immediate sanctions with those actually referred to CMS for sanctions by state survey agencies. See appendix I for a more detailed description of our scope and methodology. Our work was performed from January 2002 through June 2003 in accordance with generally accepted government auditing standards. State survey data indicate that the proportion of nursing homes with Results in Brief serious quality problems remains unacceptably high, despite a decline in such reported problems since mid-2000. Compared to the prior 18-month period, the percentage of nursing homes cited for actual harm or immediate jeopardy from July 2000 through January 2002 declined by about one-third—from 29 percent (about 5,000 homes) to 20 percent (about 3,500 homes). Consistent with this reported improvement in quality, federal comparative surveys completed during a recent 20-month period found actual harm or higher-level deficiencies in 22 percent of 5 We contacted officials in Alabama, California, Colorado, Connecticut, Iowa, Louisiana, Maryland, Michigan, Missouri, Nebraska, New York, Oklahoma, Pennsylvania, Tennessee, Washington, and Virginia. Page 3 GAO-03-561 Nursing Home Quality homes where state surveyors found no such deficiencies, compared to 34 percent in an earlier period. Fewer discrepancies between federal and state surveys suggest that state surveyors’ performance in documenting serious deficiencies has improved and that the decline in serious quality problems nationwide is potentially real. Despite this improvement, however, the magnitude of understatement of actual harm deficiencies remains a cause for concern. Federal surveyors found examples of actual harm deficiencies in about one-fifth of homes that states had judged to be deficiency free. Moreover, 39 percent of 76 surveys we reviewed from homes with a history of quality-of-care problems—but whose current survey indicated no actual harm deficiencies—had documented problems that should have been classified as actual harm: serious, avoidable pressure sores; severe weight loss; and multiple falls resulting in broken bones and other injuries. Weaknesses persist in state survey, complaint investigation, and enforcement activities. Several factors at the state level contribute to the understatement of serious quality-of-care problems. Poor investigation and documentation of deficiencies identified during nursing home surveys preclude a determination of the seriousness of some deficiencies. According to some state officials, the large number of inexperienced surveyors due to high attrition and hiring limitations has also had a negative impact on the quality of surveys. While most of the 16 states we contacted had a quality assurance process in place to review deficiencies cited at the actual harm level and higher, half did not have such a process to help ensure that the scope and severity of less serious deficiencies were not understated. The continued predictability of the occurrence of standard surveys also likely contributes to the understatement of deficiencies. Our analysis of OSCAR data indicated that about one-third of the most recent state surveys nationwide occurred on a predictable schedule, allowing homes to conceal problems if they chose to do so. In addition, many states’ complaint investigation policies and procedures were still inadequate to provide intended protections. For example, 15 states did not provide toll-free hotlines to facilitate the filing of complaints, the majority of states lacked adequate systems for managing complaints, and one or more states in most of CMS’s 10 regions did not correctly determine the investigation priority for complaints. Moreover, most states did not investigate all complaints involving actual harm within 10 days, as required. Some states attributed the timeliness problem to insufficient staff and an increase in the number of complaints. Although HCFA strengthened its enforcement policy by requiring state survey agencies, beginning in January 2000, to refer for immediate sanction homes that had a pattern of harming residents, we found that states failed to refer a Page 4 GAO-03-561 Nursing Home Quality substantial number of such homes, significantly undermining the intended deterrent effect of this policy. While CMS has increased its oversight of state survey and complaint activities, continued attention is required to help ensure compliance with federal requirements. In October 2000, HCFA implemented new annual performance reviews to measure state performance in seven areas, including the timeliness of survey and complaint investigations and the proper documentation of survey findings. The first round of results, however, did not produce information enabling the agency to identify and initiate needed improvements. For example, some regional office summary reports provided too little information to determine if a state did not meet a particular standard by a wide or a narrow margin—information that could help CMS to judge the seriousness of problems identified. We also found inconsistencies in how CMS regions conducted their reviews, raising questions about the validity and fairness of the results. Rather than relying on its regional offices, CMS plans to more centrally manage future state performance reviews to improve consistency and to help ensure that the results of those reviews could be used to more readily identify serious problems. Implementation has been significantly delayed for three other federal initiatives that are critical to reducing the subjectivity evident in the state survey process for identifying deficiencies and investigating complaints. These delayed initiatives were intended to strengthen the methodology for conducting surveys, improve surveyor guidance for determining the scope and severity of deficiencies, and increase standardization in state complaint investigation processes. We are recommending that the Administrator of CMS strengthen survey, complaint, enforcement, and oversight processes by (1) finishing the development of a more rigorous survey methodology, (2) requiring states to implement a quality assurance process to test the validity of cited deficiencies for surveys that include deficiencies below the actual harm level, (3) developing guidance for states that addresses key weaknesses in their complaint investigation processes, and (4) improving the ability of federal oversight of state survey activities to distinguish between systemic and less serious state survey performance problems. Although CMS concurred with our recommendations, its comments did not fully address our concerns about the status of the initiative intended to improve the effectiveness of the survey process or the recommendation regarding state quality assurance systems. Eleven states provided comments that most often focused on the resource constraints states face in meeting federal standards for oversight of nursing homes. Page 5 GAO-03-561 Nursing Home Quality Combined Medicare and Medicaid payments to nursing homes for care Background provided to vulnerable elderly and disabled beneficiaries were expected to total about $63 billion in 2002, with a federal share of approximately $42 billion. Oversight of nursing homes is a shared federal-state responsibility. Based on statutory requirements, CMS defines standards that nursing homes must meet to participate in the Medicare and Medicaid programs and contracts with states to assess whether homes meet these standards through annual surveys and complaint investigations. A range of statutorily defined sanctions is available to help ensure that homes maintain compliance with federal quality requirements. CMS is also responsible for monitoring the adequacy of state survey activities. Standard Surveys Every nursing home receiving Medicare or Medicaid payment must undergo a standard survey not less than once every 15 months, and the statewide average interval for these surveys must not exceed 12 months.6 A standard survey entails a team of state surveyors, including registered nurses (RN), spending several days in the nursing home to assess compliance with federal long-term care facility requirements, particularly whether care and services provided meet the assessed needs of the residents and whether the home is providing adequate quality care, such as preventing avoidable pressure sores, weight loss, or accidents. Based on our earlier work indicating that facilities could mask certain deficiencies, such as routinely having too few staff to care for residents, if they could predict the survey timing, HCFA directed states in 1999 to (1) avoid scheduling a home’s survey for the same month of the year as the home’s previous standard survey and (2) begin at least 10 percent of standard surveys outside the normal workday (either on weekends, early in the morning, or late in the evening). State surveyors’ assessment of the quality of care provided to a sample of residents during the standard survey serves as the basis for evaluating nursing homes’ compliance with federal requirements. CMS establishes specific investigative protocols for state surveyors to use in conducting these comprehensive surveys. These procedural instructions are intended to make the on-site surveys thorough and consistent across states. In response to our earlier recommendations concerning the need to better ensure that surveyors do not miss significant care problems, HCFA 6 CMS generally interprets these requirements to permit a statewide average interval of 12.9 months and a maximum interval of 15.9 months for each home. Page 6 GAO-03-561 Nursing Home Quality planned a two-phase revision of the survey process. In phase one, HCFA instructed states in 1999 to (1) begin using a series of new investigative protocols covering pressure sores, weight loss, dehydration, and other key quality areas, (2) increase the sample of residents reviewed with conditions related to these areas, and (3) review “quality indicator” information on the care provided to a home’s residents, before actually visiting the home, to help guide survey activities. Quality indicators are essentially numeric warning signs of the prevalence of care problems such as greater-than-expected instances of weight loss, dehydration, or pressures sores.7 They are derived from nursing homes’ assessments of residents and rank a facility in 24 areas compared with other nursing homes in the state.8 By using the quality indicators to select a preliminary sample of residents before the on-site review, surveyors are better prepared to identify potential care problems. Surveyors augment this preliminary sample with additional resident cases once they arrive in the home. To address remaining problems with sampling and the investigative protocols, CMS is planning a second set of revisions to its survey methodology. The focus of phase two is (1) improving the on-site augmentation of the preliminary sample selected off-site using the quality indicators and (2) strengthening the protocols used by surveyors to ensure more rigor in their on-site investigations. Complaint Investigations Complaint investigations provide an opportunity for state surveyors to intervene promptly if quality-of-care problems arise between standard surveys. Within certain federal guidelines and time frames, surveyors generally follow state procedures when investigating complaints filed against a home by a resident, the resident’s family, or nursing home employees, and typically target a single area in response to the complaint. 7 Quality indicators were the result of a HCFA-funded project at the University of Wisconsin. The developers based their work on nursing home resident assessment information, known as the minimum data set (MDS)—data on each resident that homes are required to report to CMS. See Center for Health Systems Research and Analysis, Facility Guide for the Nursing Home Quality Indicators (University of Wisconsin-Madison: Sept. 1999). 8 Because resident assessment data are used by CMS and states to calculate quality indicators and to determine the level of nursing homes’ payments for Medicare (and for Medicaid in some states), ensuring accuracy at the facility level is critical. We have made earlier recommendations to CMS on ways to improve the accuracy of these data. See U.S. General Accounting Office, Nursing Homes: Federal Efforts to Monitor Resident Assessment Data Should Complement State Activities, GAO-02-279 (Washington, D.C.: Feb. 15, 2002). Page 7 GAO-03-561 Nursing Home Quality Historically, HCFA had played a minimal role in providing states with guidance and oversight of complaint investigations. Until 1999, federal guidelines were limited to requiring the investigation of complaints alleging immediate jeopardy conditions within 2 workdays. In March 1999, HCFA acted to strengthen state complaint procedures by instructing states to investigate any complaint alleging harm to a nursing home resident within 10 workdays. Additional guidance provided to states in late 1999 specified that, as with immediate jeopardy complaints, investigations should generally be conducted on-site at the nursing home. This guidance also identified techniques to help states identify complaints having a higher level of actual harm. As part of a complaint improvement project, also initiated in late 1999, HCFA plans to issue more detailed guidance to states, such as identifying model programs or practices to increase the effectiveness of complaint investigations. Deficiency Reporting Quality-of-care deficiencies identified during either standard surveys or complaint investigations are classified in 1of 12 categories according to their scope (i.e., the number of residents potentially or actually affected) and their 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 1). States are required to enter information about surveys and complaint investigations, including the scope and severity of deficiencies identified, in CMS’s OSCAR database. Table 1: Scope and Severity of Deficiencies Identified During Nursing Home Surveys Scope Severity Isolated Pattern Widespread Immediate jeopardya J K L Actual harm G H I Potential for more than minimal harm D E F Potential for minimal harmb A B C Source: CMS. a Actual or potential for death/serious injury. b Nursing home is considered to be in “substantial compliance.” Page 8 GAO-03-561 Nursing Home Quality The importance of accurate and timely reporting of nursing home deficiency data has increased with the public reporting of survey deficiencies, which HCFA initiated in 1998 on its Nursing Home Compare Web site.9 The public reporting of deficiency data is intended to assist individuals in differentiating among nursing homes. In November 2002, CMS augmented the deficiency data available on its Web site with 10 clinical indicators of quality, such as the percentage of residents with pressure sores, in nursing homes nationwide. While the intent of this new initiative is worthwhile, CMS had not resolved several important issues that we raised prior to moving from a six-state pilot to nationwide implementation.10 These issues included: (1) the ability of the new information to accurately identify differences in nursing home quality, (2) the accuracy of the underlying data used to calculate the quality indicators, and (3) the potential for public confusion over the available data. Enforcement Policy Ensuring that documented deficiencies are corrected is a shared federal- state responsibility. CMS imposes sanctions on homes with Medicare or dual Medicare and Medicaid certification on the basis of state referrals.11 CMS normally accepts a state’s recommendation for sanctions but can modify it. The scope and severity of a deficiency determine the applicable sanctions that can involve, among other things, requiring training for staff providing care to residents, imposing monetary fines, denying the home Medicare and Medicaid payments for new admissions, and terminating the home from participation in these programs. Before a sanction is imposed, federal policy generally gives nursing homes a grace period of 30 to 60 days to correct the deficiency. We earlier reported, however, that the threat of federal sanctions did not prevent nursing homes from cycling in and out of compliance because they were able to avoid sanctions by returning to compliance within the grace period, even when they had been 9 http://www.medicare.gov/NHCompare/home.asp. 10 U.S. General Accounting Office, Public Reporting of Quality Indicators Has Merit, but National Implementation Is Premature, GAO-03-187 (Washington, D.C.: Oct. 31, 2002). 11 States are responsible for enforcing standards in homes with Medicaid-only certification—about 14 percent of homes. They may use the federal sanctions or rely on their own state licensure authority and nursing home sanctions. States are responsible for ensuring that homes that have a pattern of harming residents are immediately sanctioned. Page 9 GAO-03-561 Nursing Home Quality cited for actual harm on successive surveys.12 In 1998, HCFA began a two- stage phase-in of a new enforcement policy. In the first stage, effective September 1998, HCFA required states to refer for immediate sanction homes found to have a pattern of harming residents or exposing them to actual or potential death or serious injury (H-level deficiencies and above on CMS’s scope and severity grid). Effective January 14, 2000, HCFA expanded this policy to also require referral of homes found to have harmed one or a small number of residents (G-level deficiencies) on successive standard surveys.13 CMS Oversight CMS is responsible for overseeing each state survey agency’s performance in ensuring quality of care in state nursing homes. Its primary oversight tools are statutorily required federal monitoring surveys conducted annually in 5 percent of the nation’s certified Medicare and Medicaid nursing homes, on-site annual state performance reviews instituted during fiscal year 2001, and analysis of periodic oversight reports that have been produced since 2000. Federal monitoring surveys can be either comparative or observational. A comparative survey involves a federal survey team conducting a complete, independent survey of a home within 2 months of the completion of a state’s survey in order to compare and contrast the findings. In an observational survey, one or more federal surveyors accompany a state survey team to a nursing home to observe the team’s performance. Roughly 85 percent of federal surveys are observational. State performance reviews, implemented in October 2000, measure state performance against seven standards, including statutory requirements regarding survey frequency, requirements for documenting deficiencies, timeliness of complaint investigations, and timely and accurate entry of deficiencies into OSCAR. These reviews replaced state self-reporting of their compliance with federal requirements. In October 2000, HCFA also began to produce 19 periodic reports to monitor both state and regional office performance. The reports are based on OSCAR and other CMS databases. Examples of reports that track state activities include pending nursing home terminations (weekly), data entry 12 U. S. General Accounting Office, Nursing Homes: Additional Steps Needed to Strengthen Enforcement of Federal Quality Standards, GAO/HEHS-99-46 (Washington, D.C.: Mar.18, 1999). 13 States are now required to deny a grace period to homes that are assessed one or more deficiencies at the actual harm level or above (G-L on CMS’s scope and severity grid) in each of two successive surveys within a survey cycle. A survey cycle is two successive standard surveys and any intervening survey, such as a complaint investigation. Page 10 GAO-03-561 Nursing Home Quality timeliness (quarterly), tallies of state surveys that find homes deficiency free (semiannually), and analyses of the most frequently cited deficiencies by states (annually). These reports, in a standard format, enable comparisons within and across states and regions and are intended to help identify problems and the need for intervention. Certain reports—such as the timeliness of state survey activities—are used to monitor compliance with state performance standards. The magnitude of the problems uncovered during standard nursing home Magnitude of surveys remains a cause for concern even though OSCAR deficiency data Problems Remains indicate that state surveyors are finding fewer serious quality problems. Compared to an earlier period, the percentage of homes nationwide cited Cause for Concern since mid-2000 for actual harm or immediate jeopardy has decreased in Even Though Fewer over three-quarters of states—with seven states reporting a drop of 20 percentage points or more. State surveys conducted since about mid-2000 Serious Nursing showed less variance from federal comparative surveys, suggesting that Home Quality (1) state surveyors’ performance in documenting serious deficiencies has Problems Reported improved and (2) the decline in serious nursing home quality problems is potentially real. However, federal comparative surveys, as well as our review of a sample of survey reports from homes with a history of quality- of-care problems, continued to find understatement of actual harm deficiencies. Page 11 GAO-03-561 Nursing Home Quality Proportion of Nursing Compared to the preceding 18-month period, the proportion of nursing Homes with Documented homes cited for actual harm or immediate jeopardy has declined Actual Harm or Immediate nationally from 29 percent to 20 percent since mid-2000.14 In contrast, from early 1997 through mid-2000, the percentage of homes cited for such Jeopardy Care Problems serious deficiencies was either relatively stable or increased in 31 states.15 Has Declined since 2000 From July 2000 through January 2002, 40 states cited a smaller percentage of homes with such serious deficiencies, while only 9 states and the District of Columbia cited a larger proportion of homes with such deficiencies.16 Despite these changes, there is still considerable variation in the proportion of homes cited for serious deficiencies, ranging from about 7 percent in Wisconsin to about 50 percent in Connecticut. Appendix II provides trend data on the percentage of nursing homes cited for serious deficiencies for all 50 states and the District of Columbia. Table 2 shows the recent change in actual harm and immediate jeopardy deficiencies for states that surveyed at least 100 nursing homes.17 Specifically: • Twenty-five states had a 5 percentage point or greater decrease in the proportion of homes identified with actual harm or immediate jeopardy. For over two-thirds of these states, the decrease in serious deficiencies was greater than 10 percentage points. Seven states—Arizona, Alabama, 14 We analyzed OSCAR data for surveys performed from January 1, 1999, through July 10, 2000, and from July 11, 2000, through January 31, 2002, and entered into OSCAR as of June 24, 2002. See app. I for our complete scope and methodology. Our analysis considered only standard surveys. In commenting on a draft of this report, Missouri stated that our findings would have shown that quality had remained “fairly stable” had we included actual harm and immediate jeopardy deficiencies identified during complaint investigations in our analysis in table 2. However, we found that both nationally and in Missouri, the proportion of homes cited for actual harm or immediate jeopardy showed a similar decline even when complaint surveys were considered. 15 The two earlier time periods we analyzed are for surveys conducted from January 1, 1997, through June 30, 1998, and from January 1, 1999, through July 10, 2000. See U.S. General Accounting Office, Nursing Homes: Sustained Efforts Are Essential to Realize Potential of the Quality Initiatives, GAO/HEHS-00-197 (Washington, D.C.: Sept. 28, 2000). 16 The proportion of nursing homes in Utah cited with serious deficiencies remained the same between the two time periods. 17 We excluded Alaska, Delaware, the District of Columbia, Hawaii, Idaho, Nevada, New Hampshire, New Mexico, North Dakota, Rhode Island, Utah, Vermont, and Wyoming from this analysis because fewer than 100 homes were surveyed and even a small increase or decrease in the number of homes with serious deficiencies in such states produces a relatively large percentage point change. Page 12 GAO-03-561 Nursing Home Quality California, Michigan, Indiana, Pennsylvania, and Washington— experienced declines of 15 percentage points or more. • Two states, South Dakota and Colorado, experienced an increase of 5 percentage points or greater in the proportion of homes with actual harm or immediate jeopardy deficiencies (6.6 and 10.8, respectively). • The remaining 11 states were relatively stable—experiencing approximately a 4 percentage point change or less. Table 2: Change in the Percentage of Nursing Homes Cited for Actual Harm or Immediate Jeopardy during State Standard Surveys between the periods January 1, 1999, through July 10, 2000, and July 11, 2000, through January 31, 2002, by State Percentage of homes with actual harm or immediate jeopardy deficiencies Number of homes surveyed Percentage point a State (7/00-1/02) 1/99-7/00 7/00-1/02 differenceb Decrease of 5 percentage points or greater Arizona 147 33.8 8.8 -25.0 Alabama 228 42.2 18.4 -23.8 Pennsylvania 764 32.2 11.6 -20.6 California 1,348 29.1 9.3 -19.9 Indiana 573 45.3 26.2 -19.1 Michigan 441 42.1 24.7 -17.4 Washington 275 54.1 38.5 -15.6 Oregon 152 47.5 33.6 -13.9 Illinois 881 29.3 15.4 -13.9 Mississippi 219 33.2 19.6 -13.5 Minnesota 431 31.7 18.8 -12.9 Montana 103 37.5 25.2 -12.3 Missouri 569 22.3 10.2 -12.1 South Carolina 180 28.7 17.8 -10.9 North Carolina 419 40.8 30.1 -10.7 Arkansas 267 37.7 27.3 -10.4 Massachusetts 512 33.0 22.9 -10.2 Iowa 494 19.3 9.9 -9.4 Tennessee 377 26.0 16.7 -9.3 Nation 17,149 29.3 20.5 -8.8 Virginia 285 19.9 11.6 -8.3 Kansas 400 37.1 29.0 -8.1 Nebraska 243 26.0 18.9 -7.1 Wisconsin 421 14.0 7.1 -6.9 Maryland 248 25.6 20.2 -5.5 Ohio 1,029 29.0 23.7 -5.3 Change of less than 5 percentage points Kentucky 306 28.8 25.2 -3.7 Page 13 GAO-03-561 Nursing Home Quality Percentage of homes with actual harm or immediate jeopardy deficiencies Number of homes surveyed Percentage point a b State (7/00-1/02) 1/99-7/00 7/00-1/02 difference New Jersey 366 24.5 22.4 -2.1 Georgia 370 22.6 20.5 -2.0 West Virginia 143 15.6 14.0 -1.7 Texas 1,275 26.9 25.5 -1.5 Florida 742 20.8 20.1 -0.8 Maine 124 10.3 9.7 -0.6 New York 671 32.2 32.3 0.2 Connecticut 259 48.5 49.4 0.9 Louisiana 367 19.9 23.4 3.5 Oklahoma 394 16.7 20.6 3.9 Increase of 5 percentage points or greater South Dakota 114 24.1 30.7 6.6 Colorado 225 15.4 26.2 10.8 Source: GAO analysis of OSCAR data as of June 24, 2002. a Includes only those states in which 100 or more homes were surveyed since July 2000. b Differences are based on numbers before rounding. States offered several explanations for the declines in actual harm and immediate jeopardy deficiencies, including (1) changing guidance from CMS regional offices as to what constitutes actual harm, (2) hiring additional staff, and (3) surveyors failing to properly identify actual harm deficiencies. Federal Comparative Our analysis of federal comparative surveys conducted nationwide prior to Surveys Show Decreased and since June 2000 showed a decreased variance between federal and Variance with State Survey state survey findings (see app. I for a description of our scope and methodology). For comparative surveys completed from October 1998 Findings, but through May 2000, federal surveyors found actual harm or higher-level Understatement of Actual deficiencies in 34 percent of homes where state surveyors had found no Harm Deficiencies such deficiencies, compared to 22 percent for comparative surveys Continued completed from June 2000 through February 2002. In addition, while federal surveyors found more serious care problems than state surveyors on 70 percent of the earlier comparative surveys, this percentage declined to 60 percent for the more recent surveys. Despite the decline in understatement of actual harm deficiencies from 34 percent to 22 percent, the magnitude of the state surveyors’ Page 14 GAO-03-561 Nursing Home Quality understatement of quality problems remains an issue. For example, from June 2000 through February 2002, federal surveyors found at least one actual harm or immediate jeopardy quality-of-care deficiency in 16 of the 85 homes (19 percent) that the states had found to be free of deficiencies. For example, federal surveyors found that 1 of the 16 homes failed to prevent pressure sores, failed to consistently monitor pressure sores when they did develop, and failed to notify the physician promptly so that proper treatment could be started. The federal surveyors who conducted the comparative survey of this nursing home noted in the file that a lack of consistent monitoring of pressure sores existed at the home during the time of the state’s survey and that the state surveyors should have found the deficiency. Several states that reviewed a draft of this report questioned the value of federal comparative surveys because of their timing. Arizona noted that comparative surveys do not have to begin until up to 2 months after the state’s survey, and Iowa and Virginia officials said they might occur so long after the state’s survey that conditions in the home may have significantly changed. Although legislation requires comparative surveys to begin within 2 months of the state’s survey, CMS is continuing to make progress in reducing the timeframe between the state and the comparative survey. Based on our earlier recommendation that comparative surveys begin as soon after the state’s survey as possible, CMS instructed the regions to begin these surveys no later than one month following the state’s survey, and the average time between surveys nationally has decreased from 33 calendar days in 1999 to about 26 calendar days for surveys conducted from June 2000 through February 2002.18 Quality-of-Care Problems Even with the reported decline in serious deficiencies, an unacceptably Were Understated in high number of nursing homes—one in five nationwide—still had actual Homes with a History of harm or immediate jeopardy deficiencies. Moreover, we found widespread understatement of actual harm deficiencies in a sample of surveys we Problems reviewed that were conducted since July 2000 at homes with a history of harming residents (see app. I for a description of our methodology in selecting this sample). In 39 percent of the 76 survey reports we reviewed, we found sufficient evidence to conclude that deficiencies cited at a lower level (generally, potential for more than minimal harm, D or E) should 18 U.S. General Accounting Office, Nursing Homes: Enhanced HCFA Oversight of State Programs Would Better Ensure Quality, GAO/HEHS-00-6 (Washington, D.C.: Nov. 4, 1999). Page 15 GAO-03-561 Nursing Home Quality have been cited at the level of actual harm or higher (G level or higher on CMS’s scope and severity grid). We were unable to assess whether the scope and severity of other deficiencies in our sample of surveys were also understated because of weaknesses in the investigations conducted by surveyors and in the adequacy with which they documented those deficiencies. Of the surveys we reviewed, 30 (39 percent) contained sufficient evidence for us to conclude that deficiencies cited at the D and E level should have been cited as at least actual harm because a deficient practice was identified and linked to documented actual harm involving at least one resident (see table 3). These 30 survey reports depicted examples of actual harm, including serious, avoidable pressure sores; severe weight loss; and multiple falls resulting in broken bones and other injuries (see app. III for abstracts of these 30 survey reports). The following example illustrates understated actual harm involving the failure to provide necessary care and services. A nurse at one facility noted a large area of bruising and swelling on an 89-year-old resident’s chest. Nothing further was done to explore this injury until 11 days later when the resident began to experience shortness of breath and diminished breath sounds. Then a chest x ray was taken, revealing that the resident had sustained two fractured ribs and fluid had accumulated in the resident’s left lung. A facility investigation determined that the resident had been injured by a lift used to transfer the resident to and from the bed. It was clear from the surveyor’s information that the facility failed to take appropriate action to assess and provide the necessary care until the resident developed serious symptoms of chest trauma. Nevertheless, the surveyor concluded that there was no actual harm and cited a D-level deficiency—potential for more than minimal harm. Page 16 GAO-03-561 Nursing Home Quality Table 3: Incidence of Underreported Actual Harm Deficiencies in Surveys GAO Reviewed Number of surveys in which GAO Number of G-level Number of surveys identified G-level deficiencies GAO State from state deficiencies identified Alabama 6 2 2 Arizona 3 1 2 California 22 13 17 Iowa 7 5 7 Maryland 3 1 1 Minnesota 5 0 0 Mississippi 1 0 0 Missouri 4 1 1 Nebraska 4 2 2 Pennsylvania 11 2 3 South Carolina 1 0 0 Virginia 7 3 4 West Virginia 1 0 0 Wisconsin 1 0 0 Total 76 30 39 Source: GAO analysis of state surveys. Note: We reviewed surveys where state surveyors had cited deficiencies at the D or E level (potential for more than minimal harm) in one or more of four quality-of-care areas (see app. I, table 6). We reviewed all such deficiencies to determine if, in our judgment, the deficiencies should have been cited at the G level or higher (actual harm). State survey agency officials in Alabama, California, Iowa, and Nebraska told us that surveyors had originally cited G-level deficiencies in 10 of the surveys we reviewed, but that the deficiencies had been reduced to the D level during the states’ reviews because of inadequate surveyor documentation. We concluded that 5 of the 10 surveys did contain adequate documentation to support actual harm because there was a clear link between the deficient facility practice and the documented harm to a resident. For example, the survey managers in one state changed a G- to a D-level deficiency because the surveyor only cited one source of evidence to support the deficiency—nurses’ notes in the residents’ medical records.19 According to the surveyor, a resident with dementia, experiencing long- and short-term memory problems, fell 11 times and 19 Instructions from the state’s CMS regional office suggest, but do not require, the use of more than one source of information to support a deficiency. Page 17 GAO-03-561 Nursing Home Quality sustained a fractured wrist, three fractured ribs, and numerous bruises, abrasions, and skin tears. According to the notes of facility nurses, a personal alarm unit was in place as a safety device to prevent falls. The surveyor found that the facility had (1) failed to provide adequate interventions to prevent accidents and (2) continued to use the alarm unit even though it did not prevent any of the falls. The medical record documentation of these events was extensive and, in our judgment, was sufficient evidence of a deficiency that resulted in actual harm to the resident. In many of the 76 surveys we reviewed, including surveys in which we found no D- or E-level deficiencies that would appear to meet the criteria for actual harm deficiencies, we identified serious investigation or documentation weaknesses that could further contribute to the understatement of serious deficiencies in nursing homes. In some cases, the survey did not clearly describe the elements of the deficient practice, such as whether the resident developed a pressure sore in the facility or what the facility did to prevent the development of a facility-acquired pressure sore. In other cases, the survey omitted critical facts, such as whether a pressure sore had worsened or the size of the pressure sore. Widespread weaknesses persist in state survey, complaint investigation, Weaknesses Persist in and enforcement activities despite increased attention to these issues in State Survey, recent years. Several factors at the state level contribute to the understatement of serious quality-of-care problems, including poor Complaint, and investigation and documentation of deficiencies, the absence of adequate Enforcement quality assurance processes, and a large number of inexperienced surveyors in some states due to high attrition or hiring limitations. In Activities addition, our analysis of OSCAR data indicated that the timing of a significant proportion of state surveys remained predictable, allowing homes to conceal problems if they choose to do so. Many states’ complaint investigation policies and procedures were still inadequate to provide intended protections. For example, many states do not investigate all complaints identified as alleging actual harm in a timely manner, a problem some states attributed to insufficient staff and an increase in the number of complaints. Although HCFA strengthened its enforcement policy by requiring state survey agencies, beginning in January 2000, to refer for immediate sanction homes that had a pattern of harming residents, we found that many states did not fully comply with this new requirement. States failed to refer a substantial number of homes for sanction, significantly undermining the policy’s intended deterrent effect. Page 18 GAO-03-561 Nursing Home Quality Investigation Weaknesses CMS and state officials identified several factors that they believe and Other Factors contribute to state surveys continuing to miss significant care problems. Contribute to These weaknesses persist, in part, because many states lack adequate quality assurance processes to ensure that deficiencies identified by Underreporting of Care surveyors are appropriately classified. According to officials we Problems interviewed, the large number of inexperienced surveyors in some states due to high attrition has also had a negative impact on the quality of state surveys and investigations. Our analysis of OSCAR data also indicated that nursing homes could conceal problems if they choose to do so because a significant proportion of current state surveys remain predictable. Investigation and Consistent with the investigation and documentation weaknesses we Documentation Weaknesses found in our review of a sample of survey reports from homes with a history of actual harm deficiencies, CMS officials told us that their own activities had identified similar problems that could contribute to an understatement of serious deficiencies at nursing homes. • CMS reviews of state survey reports during fiscal year 2001 demonstrated weaknesses in a majority of states, including: (1) inadequate investigation and documentation of a poor outcome, such as reviewing available records to help identify when a pressure sore was first observed and how it changed over time, (2) failure to specifically identify the deficient practice that contributed to a poor outcome, or (3) understatement of the seriousness of a deficiency, such as citing a deficiency at the D level (potential for actual harm) when there was sufficient evidence in the survey report to cite the deficiency at the G level (actual harm). • State survey agency officials expressed confusion about the definition of “actual harm” and “immediate jeopardy,” suggesting that such confusion contributes to the variability in state deficiency trends. For example, officials in one state told us that, in their view, residents must experience functional impairment for state surveyors to cite an actual harm deficiency, an interpretation that CMS officials told us was incorrect. Under such a definition, repeated falls that resulted in bruises, cuts, and painful skin tears would not be cited as actual harm, even if the facility failed to assess the resident for measures to prevent falls. • CMS officials also told us that, contrary to federal guidance, state surveyors in at least one state did not cite all identified deficiencies but rather brought them to the homes’ attention with the expectation that the deficiencies would be corrected. CMS officials told us that they identified the problem by asking state officials about the unusually high number of homes with no deficiencies on their standard surveys. Page 19 GAO-03-561 Nursing Home Quality Inadequate Quality Assurance Some state officials told us that considerable staff resources are devoted Processes to scrutinizing the support for actual harm and higher-level deficiencies that could lead to the imposition of a sanction. While most of the 16 states we contacted had quality assurance processes to review deficiencies cited at the actual harm level and higher, half did not have such processes to help ensure that the scope and severity of less serious deficiencies were not understated.20 State officials generally told us that they lacked the staff and time to review deficiencies that did not involve actual harm or immediate jeopardy, but some states have established such programs. For example, Maryland established a technical assistance unit in early 2001 to review a sample of survey reports; the review looks at all deficiencies— not just those involving actual harm or immediate jeopardy. A Maryland official told us that she had the resources to do so because the state legislature authorized a substantial increase in the number of surveyors in 1999. However, staff cutbacks in late 2002 due to the state’s budget crisis have resulted in the reviews being less systematic than originally planned. In Colorado, two long-term-care supervisors reviewed all 1,351 deficiencies cited in fiscal year 2001. Maryland and Colorado officials told us that the reviews have identified shortcomings in the investigation and documentation of deficiencies, such as the failure to interview residents or the classification of deficiencies as process issues when they actually involved quality of care. The reviews, we were told, provide an opportunity for surveyor feedback or training that improves the quality and consistency of future surveys. Inexperienced State Surveyors State officials cited the limited experience level of state surveyors as a factor contributing to the variability in citing actual harm or higher-level deficiencies and the understatement of such deficiencies. Data we obtained from 42 state survey agencies in July 2002 revealed the magnitude of the problem: in 11 states, 50 percent or more of surveyors had 2-years’ experience or less; in another 13 states, from 30 percent to 48 percent of surveyors had similarly limited experience (see app. IV). For example, Alabama’s and Louisiana’s recent annual attrition rates were 29 percent and 18 percent, respectively, and, as a result, almost half of the surveyors in both states had been on the job for 2 years or less. In California and Maryland—states that hired a significant number of new surveyors since 2000—52 percent and 70 percent of surveyors, 20 Officials explained the focus on actual harm or higher-level deficiencies by noting that the potential for sanctions increased the likelihood that the deficiencies would be challenged by the nursing home and perhaps appealed in an administrative hearing. Page 20 GAO-03-561 Nursing Home Quality respectively, had less than 2 years of on-the-job experience.21 According to CMS regional office and state officials, the first year for a new surveyor is essentially a period of training and low productivity, and it takes as long as 3 years for a surveyor to gain sufficient knowledge, experience, and confidence to perform the job well. High staff turnover was attributed, in part, to low salaries for RN surveyors—salaries that may not be competitive with other employment opportunities for nurses. Overall, 29 of the 42 states that responded to our inquiry indicated that they believed nurse surveyor salaries were not competitive (see app. IV). Officials in several states also told us that the combination of low starting salaries and a highly competitive market forced them to hire less qualified candidates with less breadth of experience. Predictable Surveys Even though HCFA directed states, beginning January 1, 1999, to avoid scheduling a nursing home’s survey for the same month of the year as its previous survey, over one-third of state surveys remain predictable. Our analysis demonstrated little change in the proportion of predictable nursing home surveys. Predictable surveys can allow quality-of-care problems to go undetected because homes, if they choose to do so, may conceal problems.22 We recommended in 1998 that HCFA segment the standard survey into more than one review throughout the year, simultaneously increasing state surveyor presence in nursing homes and decreasing survey predictability. Although HCFA disagreed with segmenting the survey, it did recognize the need to reduce predictability. Our analysis of OSCAR data demonstrated that, on average, the timing of 34 percent of current surveys nationwide could have been predicted by nursing homes, a slight reduction from the prior surveys when about 38 percent of all surveys were predictable. The predictability of current surveys ranged from 83 percent in Alabama to 10 percent in Michigan (see app. V for data on all 50 states and the District of Columbia). In 34 states, 25 percent to 50 percent of current surveys were predictable, as shown in 21 As of July 2002, both states had vacant surveyor positions and a surveyor hiring freeze. 22 In commenting on a draft of this report, Arizona disagreed with the significance we attribute to survey predictability, questioning whether poor homes would, or even could, hide problems if they knew a survey was imminent. However, advocates and family members have told us that a home that operates with too few staff could temporarily augment its staff during the expected period of a survey in order to mask an otherwise serious deficiency—a common practice based on advocates’ own observations. Page 21 GAO-03-561 Nursing Home Quality table 4. In 9 states, more than 50 percent of current surveys were predictable.23 Table 4: Predictability of Nursing Home Surveys a b Percentage of predictable surveys Number of states More than 50 percent 9 25 percent to 50 percent 34 Less than 25 percent 8 Source: GAO analysis of OSCAR data as of April 9, 2002. a We considered surveys to be predictable if (1) homes were surveyed within 15 days of the 1-year anniversary of their prior surveys, or (2) homes were surveyed within 1 month of the maximum 15- month interval between standard surveys. b Includes the District of Columbia. Many State Complaint Most state agencies did not investigate serious complaints filed against Investigation Systems Still nursing homes within required time frames, and practices for investigating Have Timeliness Problems complaints in many states may not be as effective as they could be. A CMS review of states’ timeliness in investigating complaints alleging harm to and Other Weaknesses residents revealed that most states did not investigate all such complaints within 10 days, as CMS requires. Additionally, a CMS-sponsored study of complaint practices in 47 states raised concerns about state approaches to accepting and investigating complaints. Until March 1999, states could set their own complaint investigation time frames, except that they were required to investigate within 2 workdays all complaints alleging immediate jeopardy conditions. In March 1999, we reported that inadequate complaint intake and investigation practices in states we reviewed had too often resulted in extensive delays in investigating serious complaints.24 As a result of our findings, HCFA began requiring states to investigate complaints that allege actual harm, but do 23 We considered surveys to be predictable if (1) homes were surveyed within 15 days of the 1-year anniversary of their prior surveys (13 percent of homes, nationally) or (2) homes were surveyed within 1 month of the maximum 15-month interval between standard surveys (21 percent of homes, nationally). Because homes know the maximum allowable interval between surveys, those whose prior surveys were conducted 14 or 15 months earlier are aware that they are likely to be surveyed soon. 24 U.S. General Accounting Office, Nursing Homes: Complaint Investigation Processes Often Inadequate to Protect Residents, GAO/HEHS-99-80 (Washington, D.C.: Mar. 22, 1999). Page 22 GAO-03-561 Nursing Home Quality not rise to the level of immediate jeopardy, within 10 workdays.25 CMS’s 2001 review of a sample of complaints in all states demonstrated that many states were not complying with these requirements. Specifically, 12 states were not investigating all immediate jeopardy complaints within the required 2 workdays, and 42 states were not complying with the requirement to investigate actual harm complaints within 10 days.26 The agency also found that the triaging of complaints to determine how quickly each complaint should be investigated was inadequate in many states. The extent to which states did not meet the 2-day and 10-day investigation requirements varied considerably. Officials from 12 of the 16 states we contacted indicated that they were unable to investigate complaints on time because of staff shortages. Oklahoma investigated only 3 of the 21 immediate jeopardy complaints that CMS sampled within the required 2- day period and none of 14 sampled actual harm complaints in 10 days. Oklahoma officials attributed this timeliness problem to staff shortages and a surge in the number of complaints received in 2000, from about 5 per day to about 35. The rising volume of complaints is a particular problem for California, which receives about 10,000 complaints annually, and had a 20 percent increase in complaints from January 2001 through July 2002. State officials told us that California law requires all complaints alleging immediate jeopardy to a resident to be investigated within 24 hours and all others to be investigated within 10 days, and that the increase in the number of complaints requires an additional 32 surveyor positions.27 CMS regional officials told us that the vast majority of California complaints were investigated within 10 days. However, the 2001 review also showed that about 9 percent of the state’s standard surveys were conducted late.28 Both CMS and California officials indicated that the priority the state attaches to investigating complaints affected survey timeliness. Officials 25 In some states, the 10-day requirement significantly compressed the time frame in which complaints alleging potential actual harm must be investigated. For instance, prior to HCFA’s change, such complaints were supposed to be investigated within 30 days in Michigan and 60 days in Tennessee. 26 Staff from each of CMS’s regional offices reviewed a 10 percent random sample of complaint files (maximum of 40 files) in each state. 27 According to a state official, a hiring freeze precluded increasing the number of surveyors. 28 Because CMS based its analysis of timeliness only on nursing homes that actually were surveyed during fiscal year 2001—and not on all homes in the state—the 9 percent figure is understated. Our analysis of all homes indicated that about 12 percent of the state’s homes were not surveyed within the required time frame. Page 23 GAO-03-561 Nursing Home Quality from Washington told us that their practice of investigating facility self- reported incidents led to their not meeting the 10-day requirement on all complaints that CMS reviewed. Washington investigated 18 of 20 sampled actual harm complaints on time—missing the 10-day requirement for the other two by 2 days and 4 days, respectively. Washington officials pointed out that the two complaints not investigated within 10 days were facility self-reported incidents and commented that many other states do not even require investigation of such incidents. Thus, in these other states, such incidents would not even have been included in CMS’s review. In its review of state complaint files, CMS also evaluated whether states had appropriately triaged complaints—that is, determined how quickly each complaint should be investigated. Most of the regions told us that one or more of their states had difficulty determining the investigation priority for complaints. In an extreme case, a regional office discovered that one of its states was prioritizing its complaints on the basis of staff availability rather than on the seriousness of the complaints. Several regions indicated that some states improperly assigned complaints to categories that permitted longer investigation time frames, and one region indicated that triaging difficulties involved state personnel not collecting enough information from the complainant to make a proper decision. Officials from some of the 16 state survey agencies we contacted indicated that HCFA’s 1999 guidance to states on what constitutes an actual harm complaint was unclear and confusing. In an effort to improve state responsiveness to complaints, HCFA hired a contractor in 1999 to assess and recommend improvements to state complaint practices. The study identified significant problems with states’ complaint processes, including complaint intake activities, investigation procedures, and complaint substantiation practices.29 For example, the report noted that 15 states did not have toll-free hotlines for the public to file complaints. In our earlier reports, we noted that the process of filing a complaint should not place an unnecessary burden on a complainant and that an easy-to-use complaint process should include a toll-free number that permits the complainant to leave a recorded message when state staff 29 Center for Health Systems Research and Analysis at the University of Wisconsin, Madison, Final Report: Complaint Improvement Project, prepared for CMS, June 3, 2002. The report is based on a questionnaire sent to the 50 states, the District of Columbia, Puerto Rico, and CMS’s 10 regional offices. Three states did not respond to the questionnaire. The report treated the District of Columbia and Puerto Rico as states. Page 24 GAO-03-561 Nursing Home Quality are unavailable.30 Table 5 summarizes major findings from the contractor’s report to CMS. Table 5: Key Findings of Report to CMS on State Complaint Investigation Processes Finding Description States vary in the ease with Thirty-four states indicated that they provide toll-free which the public can file a hotlines for the public to file complaints. Twenty-nine of complaint. the 34 states indicated that they operate their hotlines 24 hours a day, 7 days a week, and 5 said their hotlines were answered during business hours. Nineteen states had no provisions or plans to handle non-English speaking complainants. States need to improve their States need to better triage their complaints and decide complaint intake and triaging which complaints should be referred to other agencies for systems. investigation. They should also improve procedures for merging complaints with ongoing survey activities at a nursing home. More consistency is needed in handling facility self-reported incidents. State survey staffs that States should use staff dedicated to investigating conduct complaint intake complaints to improve the quality of investigations. This and investigation often have might include assigning responsibility for a state’s total additional duties. complaint system to a single complaint supervisor or coordinator and also may require more careful hiring standards with specific job qualifications. Investigation procedures States do not use all available data when preparing for a vary across states. complaint investigation. There is little agreement among states regarding how many resident records should be sampled during a complaint investigation.a Complaint investigation Specialized complaint training and periodic refresher training is needed. training on complaint intake, triaging, and investigation techniques are needed to improve the quality of complaint investigations. Resolution of complaints is States have developed varying criteria for determining inconsistent across states. what constitutes a substantiated complaint and varying practices for communicating the results of investigations to complainants. Twenty-two states could not indicate how long it takes them to provide the results of an investigation to the complainant, and at least four states do not inform the complainant of the results. Not all states have Twenty states indicated that they could track the status of comprehensive complaint complaints and produce summary reports. tracking systems, and CMS tracking systems are not up- to-date or user friendly.b Source: CMS. 30 See GAO/HEHS-99-80 and U.S. General Accounting Office, Medicare Home Health Agencies: Weaknesses in Federal and State Oversight Mask Potential Quality Issues, GAO-02-382 (Washington, D.C.: July 19, 2002). Page 25 GAO-03-561 Nursing Home Quality Note: GAO analysis of information from Center for Health Systems Research and Analysis at the University of Wisconsin, Madison, Final Report: Complaint Improvement Project, prepared for CMS, June 3, 2002. a In 1999, we reported that HCFA had not provided states with guidance on when to expand a complaint review beyond the residents who were the subject of the original complaint. See GAO/HEHS-99-80. b CMS is planning to implement a new complaint tracking system nationwide that should address this shortcoming. States Did Not Refer a State survey agencies did not refer 711 cases in which nursing homes were Substantial Number of found to have a pattern of harming residents to CMS for immediate Nursing Homes to CMS for sanction as required by CMS policy.31 Our earlier work found that nursing homes tended to “yo-yo” in and out of compliance, in part because HCFA Immediate Sanctions rarely imposed sanctions on homes with a pattern of deficiencies that harmed residents.32 In response, the agency required that homes found to have harmed residents on successive standard surveys be referred to it for immediate sanction.33 Most states did not refer at least some cases that should have been referred under this policy.34 Figure 1 shows the results of our analysis for the four states—Massachusetts, New York, Pennsylvania, and Texas—with the greatest numbers of cases that should have been 31 Using CMS data, we identified 1,334 cases that appeared to meet the criteria for immediate sanctions but that did not appear to have been referred to CMS by states. (See app. I for a description of our methodology.) We use the term “cases” rather than “nursing homes” because some nursing homes had multiple referrals for immediate sanctions. At our request, CMS reviewed most of these cases and determined that 711 (62 percent of those CMS reviewed) should have been—but were not—referred for immediate sanction. CMS did not analyze 155 of the cases we asked it to examine and was unable to determine the status of an additional 30 cases. 32 See GAO/HEHS-99-46. 33 This policy was implemented in two stages, and our analysis focused on implementation of the second stage in January 2000. Beginning in September 1998, HCFA required states to refer homes that had a pattern of harming a significant number of residents or placed residents at high risk of death or serious injury (H-level deficiencies and above on CMS’s scope and severity grid). Effective January 14, 2000, HCFA expanded this policy by requiring state survey agencies to refer for immediate sanction homes that had harmed residents—G-level deficiencies on the agency’s scope and severity grid—on successive surveys. States are now required to deny a grace period to homes that are assessed one or more deficiencies at the actual harm level or above (G-L on CMS’s scope and severity grid) in each of two surveys within a survey cycle. A survey cycle is two successive standard surveys and any intervening survey, such as a complaint investigation. 34 We found that states did refer 4,310 cases over a 27-month period. See app. VI for a summary of all sanctions that were implemented, including the amount of civil money penalties (CMPs) by state. Page 26 GAO-03-561 Nursing Home Quality referred and for the nation (see app. VII for information on all states). These four states accounted for 55 percent of the 711 cases. Figure 1: Four States with the Greatest Number of Cases that Should Have Been Referred for Immediate Sanctions, January 14, 2000, through March 28, 2002 State 169 a Texas 423 66 140 New York 22 0 46 Massachusetts 81 0 38 Pennsylvania 164 50 711 4310 Nation 155 0 100 200 300 400 500 600 700 800 5000 Cases Cases that should have been referred and were not Cases that were referred CMS did not determine if cases should have been referred Source: GAO and CMS analysis of OSCAR and enforcement data. Note: Analysis includes cases entered in CMS’s enforcement database by March 28, 2002. a According to a Dallas regional office official, Texas referred most of the 423 cases because the nursing homes had a “poor enforcement history,” not because of repeat harm level deficiencies. However, based on other information, the region coded these cases as requiring immediate sanction. State and CMS officials identified several reasons why state agencies failed to forward cases to CMS for immediate sanction, including (1) an initial misunderstanding of the policy on the part of some states and regions, (2) poor state systems for monitoring the survey history of homes to identify those meeting the criteria for referral for immediate sanction, and (3) actions, by two states, that were at variance with CMS policy. First, officials from some states—and some CMS regional officials as well—told Page 27 GAO-03-561 Nursing Home Quality us that they did not initially fully understand the criteria for referring homes for immediate sanction.35 For example, several states and CMS regional offices reported that they did not understand that CMS required states to look back before the January 2000 policy implementation date to determine if there was an earlier survey with an actual-harm-level deficiency. The look-back requirement was specifically addressed in a February 10, 2000, CMS policy clarification specifying that state agencies were to consider the home’s survey history before the January 14, 2000, implementation date in determining if a home met the criteria for immediate referral for sanction. However, officials in one region told us that they had instructed three of four states not to look back before the January 2000 implementation date of the policy. Two other regional offices told us that CMS policy did not require the state to look back before January 2000 for earlier surveys. Officials at another regional office did not recall the look-back policy at the time we talked to them in mid-2002, and were not sure what advice they had given their states when the policy was first implemented. Second, some state survey agencies told us that their managers responsible for enforcement did not have an adequate methodology for checking the survey history of homes to identify those meeting the criteria. Some states said that their managers relied on manual systems, which are less accurate and sometimes failed to identify cases that should have been referred. Officials in one state told us that its district offices had no consistent procedure for checking the survey history of homes. An official in another state told us that some cases were not referred because time lags in reporting some surveys meant that an earlier survey—such as a complaint survey—with an actual harm deficiency might not have been entered in the state’s tracking system until after a later survey that also found harm-level deficiencies. Third, two states did not implement CMS’s expanded policy on immediate sanctions. New York was in direct conflict with CMS policy. Although CMS policy calls for state referrals to CMS regardless of the type of deficiency, 35 Arizona’s comments on a draft of this report indicated that eight of the nine cases not referred for immediate sanction were during the period January through October 2000 when the state was struggling with various interpretations of CMS’s new requirement. Similarly, Missouri officials indicated in their comments that the majority of cases they did not refer occurred during the initial stages of the new policy, which Missouri believes was “complicated, at best.” Missouri officials added that the number of missed cases significantly declined as the state gained a better understanding of the policy. Page 28 GAO-03-561 Nursing Home Quality a state agency official told us that the state only referred a home to CMS for immediate sanction if both actual harm citations were for the exact same deficiency.36 A CMS official indicated that New York began complying with the policy in September 2002.37 Texas, the second state, did not implement the CMS policy statewide until July 2002, when it received our inquiry about the cases not referred for immediate sanction. In the interim from January 2000 through July 2002, three of Texas’s 11 district offices specifically requested from state survey agency officials, and were granted, permission to implement the policy. While CMS has increased its oversight of state survey and complaint CMS Oversight of activities and instituted a more systematic oversight process by initiating State Survey annual state performance reviews, CMS officials acknowledged that the effectiveness of the reviews could be improved. In particular, CMS Activities Requires officials told us that for the initial state performance review in fiscal year Further Strengthening 2001, they lacked the capability to systematically distinguish between minor lapses identified during the reviews and more serious problems that require intervention. CMS oversight is also hampered by continuing limitations in OSCAR data, the inability or reluctance of some CMS regions to use such data to monitor state activities, and inadequate oversight of certain areas, such as survey predictability and state referral of homes for immediate enforcement actions. CMS has restructured regional office responsibilities to improve the consistency of federal oversight and plans to further strengthen oversight by increasing the number of federal comparative surveys. However, three federal initiatives critical to reducing the subjectivity evident in the current survey process and the investigation of complaints have been delayed. 36 This New York state official told us that the state believed it was in compliance with CMS’s policy because it imposed one of two minor federal sanctions and a state civil money penalty on all consecutive G-level cases. This state official also indicated that state fines were imposed in place of federal civil money penalties in all cases. (The maximum state fine is $2,000 per violation, lower than the federal maximum of $10,000 per instance or per day of noncompliance.) However, when we discussed this explanation with officials in the CMS central office, they disagreed that the state was in compliance. 37 In commenting on a draft of our report, New York officials indicated that their initial failure to refer nursing homes for immediate sanctions was based on their misinterpretation of the new policy and not on a deliberate refusal to implement it. They also indicated that their procedures are now consistent with the federal policy. Page 29 GAO-03-561 Nursing Home Quality CMS Reviews of State In the first of what is planned as an annual process, CMS’s 10 regional Performance Have offices reviewed states’ fiscal year 2001 performance for seven standards Identified Areas for to determine how well states met their nursing home survey responsibilities (see app. VIII for a description of the seven standards).38 Improvement This enhanced oversight of state survey agency performance responds to our prior recommendations. In 1999, we reported that HCFA’s oversight of state efforts had limitations preventing it from developing accurate and reliable assessments of state performance.39 HCFA regional office policies, practices, and oversight had been inconsistent, a reflection of coordination problems between HCFA’s central office and its regional staffs. In important areas, such as the adequacy of surveyors’ findings and complaint investigations, HCFA relied on states to evaluate their own performance and report their findings to HCFA. Although OSCAR data were available to HCFA for monitoring state performance, they were infrequently used, and neither the states nor HCFA’s regional offices were held accountable for failing to meet or enforce established performance standards. To promote consistent application of the standards across the 10 regions, the agency developed detailed guidance for measuring each standard, including the method of evaluation, the data sources to be used, and the criteria for determining whether a state met a standard. Only two states met each of the five standards we reviewed and many did not meet several standards. Appendix IX identifies the standards we analyzed and the results of CMS’s review of these standards. During the 2001 review, CMS elected not to impose the most serious sanctions available for inadequate state performance, including reducing federal payments to the state or initiating action to terminate the state’s agreement, but advised the states that annual reviews in subsequent years will serve as the basis for such actions. While imposing no sanctions during the 2001 review, CMS did require several states to prepare corrective action plans. Each year, CMS plans to update and improve the standards based on experience gained in prior years. 38 The CMS regions assessed each state’s performance by (1) reviewing a set of standardized reports drawn from information contained in CMS’s databases and (2) visiting states to review procedures and to examine a sample of records, such as complaint investigation files. Some reviews, such as assessing state complaint investigation timeliness, were performed semiannually, enabling regional office staff to provide midpoint feedback intended to correct any deficiencies identified. 39 GAO/HEHS-00-6. Page 30 GAO-03-561 Nursing Home Quality CMS’s State Performance Characterizing its fiscal year 2001 state performance review as a “shakeout Standards and Review Had cruise,” CMS is working to address several weaknesses identified during Shortcomings the reviews, including difficulty in determining if identified problems were isolated incidents or systemic problems, flawed criteria for evaluating a critical standard, and inconsistencies in how regional offices conducted the reviews. In our discussions of the results of the performance reviews with officials of CMS’s regional offices, it was evident that some regions had a much better appreciation of the strengths and weaknesses of survey activities in their respective states than was reflected in the state performance reports. However, this information was not readily available to CMS’s central office. In addition, CMS has not released a summary of the review to permit easy comparison of the results. For subsequent reviews, CMS plans to more centrally manage the process to improve consistency and help ensure that future reviews distinguish serious from minor problems. Distinctions in State CMS officials acknowledged that the first performance review did not Performance Were Hard to provide adequate information regarding the seriousness of identified Identify problems. The agency indicated that it had since revised the performance standards to enable it to determine the seriousness of the problems identified. Some regional office summary reports provided insufficient information to determine whether a state did not meet a particular standard by a wide or a narrow margin. For example, although California did not meet the standard to investigate all complaints alleging actual harm within 10 days, the regional office summary provided no details about the results. Regional officials told us that they found very few California complaints that were not investigated within the 10-day deadline and those that were not were generally investigated by the 13th day.40 Conversely, although the report for Oregon shows that the state met the 10-day requirement, our discussions with regional officials revealed that serious shortcomings nevertheless existed in the state’s complaint investigation practices. 41 Officials in the Seattle region told us that for many years Oregon had contracted out investigations of complaints to local government entities not under the control of the state agency and, as 40 According to CMS regional officials, California state law requires that all complaints other than those alleging immediate jeopardy be investigated within 10 days, irrespective of the seriousness of the allegation. 41 CMS’s database showed that Oregon conducted only 14 on-site complaint investigations during fiscal year 2001. Because of this low number, the region reviewed the entire universe of complaints (instead of a sample), but did not identify the number reviewed in its report. Page 31 GAO-03-561 Nursing Home Quality a result, exercised little control over the roughly 2,000 complaints the state receives against nursing homes each year. For instance, under this arrangement, information about complaint investigations, including deficiencies identified, was not entered into CMS’s database. Regional officials told us that the Oregon state agency recently assumed responsibility for investigating complaints filed by the public, but that the local government entities continue to investigate facility self-reported incidents. CMS’s Standard for Measuring CMS’s standard for measuring how well states document deficiencies States’ Documentation of identified during standard surveys was flawed because it mixed major and Deficiencies Was Flawed minor issues, blurring the significance of findings. CMS’s protocol required assessment of 33 items, ranging from the important issue of whether state surveyors cited deficiencies at the correct scope and severity level to the less significant issue of whether they used active voice when writing deficiencies. Because of the complexity of the criteria and concerns about the consistency of regional office reviews of states’ documentation practices, CMS decided not to report the results for this standard for 2001. For the 2002 review, CMS reduced the number of criteria to be assessed from 33 to 7.42 Based on the available evidence of the understatement of actual harm deficiencies, we believe that successful implementation of the documentation standard in 2002 and future years is critical to help ensure that deficiencies are cited at the appropriate scope and severity level. CMS Regions’ Reviews Were CMS’s regional offices were sometimes inconsistent in how they Inconsistent conducted their reviews, raising questions about the validity and fairness of the results. For example: • Although the guidelines for the review indicated that the regional offices were to assess the timeliness of complaint investigations based on the state’s prioritization of the complaint, officials from one region told us that they judged timeliness based on their opinion of how the complaint should have been prioritized. 42 CMS’s criteria for evaluating the documentation standard in 2002 are (1) the proper regulation is cited for each deficiency, (2) evidence supports the cited area of noncompliance, (3) several components required by the relevant regulation for each deficiency, such as identifying the citation number, are included, (4) the deficient practice is identified, (5) the cited severity of each deficiency is accurate, (6) the cited scope of each deficiency is accurate, and (7) the sources and identifiers in the deficient practice statement match the sources and identifiers in the findings. Page 32 GAO-03-561 Nursing Home Quality • Two regional offices acknowledged that they did not use clinicians to review complaint triaging. Officials from two states questioned the credibility of reviews not conducted by clinicians. • Although one objective of the reviews was to review some immediate jeopardy complaints in every state, the random samples selected in some states did not yield such complaints. In such cases, one region indicated that it specifically selected a few immediate jeopardy complaints outside the sample while another region did not. To eliminate this inconsistency in future years, CMS has instructed the regions to expand their sample to ensure that at least two immediate jeopardy complaints are reviewed in each state. • While some regions examined more than the required number of complaints to assess overall timeliness, one region felt that additional reviews were unnecessary. For instance, surveyors reviewing California, which receives thousands of complaints per year, expanded the number of complaints reviewed beyond the minimum number required because they felt that the required random sample of 40 complaints did not provide sufficient information about overall timeliness in the state. To assess overall timeliness, they visited all but 1 of the state’s 17 district offices to review complaints. However, surveyors from another CMS region reviewed only 3 or 4 of the roughly 18 complaints a state received and told us that additional reviews were unnecessary because the state had already failed the timeliness criterion based on the few complaints reviewed. Although the review of 3 or 4 complaints technically met CMS’s sampling requirement, we believe examination of most or all of the relatively few remaining complaints would have provided a more complete picture of the state’s overall timeliness. Performance Standards While CMS has addressed some of the weaknesses in its 2001 state Excluded Some Important performance review by revising the standards and guidance for the 2002 Areas review, including simplifying the criteria for assessing documentation and requiring regions to assess states’ complaint prioritization efforts separately from the timeliness issue, the performance standards do not yet address certain issues that are important for assessing state performance and that would further strengthen CMS oversight of state survey activities. These issues include: • Assessing the predictability of state surveys. Although CMS monitored compliance with its requirement for state survey agencies to initiate at least 10 percent of their standard surveys outside normal working hours to reduce predictability, it did not examine compliance with its 1999 instructions for states to avoid scheduling a home’s survey during the same month each year. As shown in app. V, our analysis of CMS data found that from 10 percent to 31 percent of surveys in 31 states were Page 33 GAO-03-561 Nursing Home Quality predictable because they were initiated within 15 days of the 1-year anniversary of the prior survey. • Evaluating states’ compliance with the requirement to refer nursing homes that have a pattern of harming residents for immediate sanctions. CMS officials confirmed that there was no consistent oversight of state agencies’ implementation of this policy. Several CMS regional offices generally did not know, for example, how their states were monitoring homes’ survey history to detect cases that should be referred for immediate sanction. CMS could have used the enforcement database to determine that New York was not adhering to the agency’s immediate sanctions policy. During calendar years 2000 and 2001, New York cited actual harm at a relatively high proportion of its nursing homes but only referred 19 cases for immediate sanction. Over a comparable period, New Jersey, a state with far fewer homes and citations, referred almost three times as many cases.43 • Developing better measures of the quality of state performance, in addition to process measures. Several CMS regional officials believed that the scope of the state performance standards should address additional areas of performance, including assessing the adequacy of nursing homes’ plans of correction submitted in response to deficiencies and the appropriateness of states’ recommended enforcement remedies. In particular, several regions noted that rather than focusing only on the timeliness of complaint investigations, regions should also assess the adequacy of the investigation itself, including whether the complaint should have been substantiated. The introduction of a new CMS complaint tracking database, discussed below, should enable regions to automate the review of complaint timeliness, thereby allowing them to focus more attention on such issues. Data Limitations and CMS’s oversight of state survey activities is further hampered by Inconsistent Use of limitations in the data used to develop the 19 periodic reports intended to Periodic Reports Hamper assist the regions in monitoring state performance and by the regions’ inconsistent use of the reports.44 For instance, CMS’s current complaint Oversight database does not provide key information about the number of 43 While cases referred by states were typically recorded in CMS’s enforcement database, a New York regional official indicated that because of the departure of key staff members, the region had not entered all cases into the database. 44 CMS’s central office and the regions have jointly produced the reports since they were created in 2000. As CMS’s systems become more user-friendly, the regions will be able to produce them independently. Page 34 GAO-03-561 Nursing Home Quality complaints each state receives (including facility self-reported incidents) or the time frame in which each complaint is investigated.45 In addition, officials from one region emphasized to us that information about complaints provided in the reports did not correspond with CMS’s required complaint investigation time frames. The reports identify the number of state on-site complaint investigations that took place in three different time periods—3 days, from 4 to 14 days, and 15 days or more; however, required time frames for complaint investigations are 2 days for complaints alleging immediate jeopardy and 10 days for those alleging harm. Additionally, a regional official pointed out that investigations shown in one of the reports as taking place within 3 days do not necessarily represent complaints that the state prioritized as immediate jeopardy. Despite the problems with these data, however, several regional offices indicated that the reports could at least serve as a starting point for discussions with states about their complaint programs and often lead to a better understanding of state complaint activities. CMS indicated that the deficiencies in complaint data should be addressed by the new automated complaint tracking system that it is developing for use by all states as part of the redesign of OSCAR.46 Officials from several regions also told us that the value of some of the 19 periodic reports was unclear, and officials in three regions said they either lacked the staff expertise or the time to use the reports routinely to oversee state activities. For example, officials in one region told us that 45 As we reported previously, although HCFA standards require states to report information about complaints, the process for collecting it results in inaccurate and incomplete information. For example, the form CMS requires states to use to record the results of complaint investigations was created to record information about a single complaint, but many states investigate multiple complaints at a nursing home during one on-site visit. As a result, the timeliness, prioritization, and other important tracking information related to multiple complaints is reported as though it applies to one complaint. See GAO/HEHS-99-80. 46 CMS planned to implement the new system, known as the ASPEN Complaint Tracking System, or ACTS, nationwide in October 2002. However, implementation was delayed because of several issues that surfaced during pilot testing: (1) states have different policies regarding the treatment of self-reported facility incidents, (2) complaints filed with some states may be investigated by entities other than the state survey agency (for instance, the Board of Nursing), and (3) 8 to 10 states have indicated that their current state complaint tracking systems have superior capability to ACTS and they do not wish to discontinue using their own system or maintain separate systems. CMS plans to evaluate this last issue during the extended pilot test. As of July 2003, nationwide implementation had been further delayed by the need to obtain approval from the Office of Management and Budget for publication of a notice in the Federal Register, a procedure that applies to establishing a system of federal records. Page 35 GAO-03-561 Nursing Home Quality they used one of the reports about complaints to ask states questions about their prioritization practices. But a different region appeared unaware that the reports showed that two of its states might be outliers in terms of the percentage of complaints they prioritized as actual harm or immediate jeopardy. Additionally, because the periodic reports do not include trend data, many regional offices were unaware of the trends in the percentage of homes cited in their states for actual harm or immediate jeopardy. We believe that such data could be useful to CMS’s regions in identifying significant trends in their states. CMS indicated that it is continuing to make progress in redesigning the OSCAR reporting system. In 1999, we recommended that the agency develop an improved management information system that would help it track the status and history of deficiencies, integrate the results of complaint investigations, and monitor enforcement actions.47 Another objective of the OSCAR redesign is to make it easier to analyze the data it contains, addressing the problem that generating analytical reports from OSCAR was difficult and most regions lacked the expertise to do so. The redesigned system, called the Quality Improvement and Evaluation System, would also eliminate the need for duplicate data entry, which should reduce the potential for data entry errors to which OSCAR is susceptible.48 CMS has faced some problems in the implementation of the new system, such as inadvertent modifications of survey data results when data are transferred from the old OSCAR database into the new system, but the agency indicated that its target date for completing the redesign is 2005. CMS Is Making Progress CMS has taken, or is undertaking, several other efforts to improve federal but Also Encountering oversight and survey procedures, including making structural changes to Delays in Several Key the regional offices to improve coordination, expanding the number of comparative surveys conducted each year, improving the survey Efforts methodology, developing clearer guidance for surveyors, and developing additional guidance to states for investigating complaints. As of April 2003, only the effort to restructure the regional offices had been completed. The 47 GAO/HEHS-99-46. 48 Until recently, states had to manually enter data into a computerized system that generated survey reports and then manually reenter much of the same data into OSCAR. This duplicative data entry process increased the chances for errors in OSCAR. Page 36 GAO-03-561 Nursing Home Quality other efforts critical to reducing the subjectivity evident in the current survey process and the investigation of complaints have been delayed. CMS Is Taking Additional Steps In December 2002, CMS reduced the number of regional managers in to Address Inconsistencies in charge of survey activities from 10 (1 per region) to 5, a change intended Regional Office Performance to provide more management attention to survey matters and to improve and Improve Federal Oversight accountability, direction, and leadership. Our prior and current work found that regional offices’ policies, practices, and oversight were often inconsistent. For example, in 1999 we reported that regional offices used different criteria for selecting and conducting comparative surveys. The 5 regional managers will be responsible only for survey and certification activities, while in the past many of the 10 were also responsible for managing their regions’ Medicaid programs. In response to our prior recommendations, CMS plans to more than double the number of federal comparative surveys in which federal surveyors resurvey a nursing home within 2 months of the state survey to assess state performance. We noted in 1999 that, although insufficient in number, comparative surveys were the most effective technique for assessing state agencies’ abilities to identify serious deficiencies in nursing homes because they constitute an independent evaluation of the state survey. CMS plans to hire a contractor to perform approximately 170 additional comparative surveys per year, bringing the annual total of comparative surveys performed by both CMS surveyors and the contractor to about 330. Although CMS had intended to award a contract and begin surveys by spring 2003, as of July 2003, it was still in the process of identifying qualified contractors. CMS officials stated that using a contractor would provide CMS flexibility because if it suspects that a state or region is having problems with surveys, it can quickly have the contractor conduct several comparative surveys there. Being able to direct the contractor to quickly focus on states or regions where state surveys may be problematic could represent a significant improvement in CMS’s oversight of state survey agencies. Key Initiatives to Improve CMS’s implementation schedules have slipped for three critical initiatives Survey Consistency and intended to enhance the consistency and accuracy of state surveys and Complaint Investigations Have complaint investigations, delaying the introduction of improved Been Delayed methodologies or guidance until 2003 or 2004. Because surveyors often missed significant care problems due to weaknesses in the survey process, HCFA took some initial steps to strengthen the survey methodology, with the goal of introducing an improved survey process in 2000. In July 1999, the agency introduced quality indicators to help surveyors do a better job of selecting a resident sample, instructed states to increase the sample size Page 37 GAO-03-561 Nursing Home Quality in areas of particular concern, and required the use of investigative protocols in certain areas, such as pressures sores and nutrition, to help make the survey process more systematic.49 However, HCFA recognized that additional steps were required to ensure that surveyors thoroughly and systematically identify and assess care problems. To address remaining problems with sampling and the investigative protocols, CMS contracted for the development of a revised survey methodology. The contractor has proposed a two-phase survey process.50 In the first phase, surveyors would initially identify potential care problems using quality indicators generated off-site prior to the start of the survey and additional, standardized information collected on-site, from a sample of as many as 70 residents. During the second phase, surveyors would conduct an investigation to confirm and document the care deficiencies initially identified.51 According to CMS officials, this process differs from the current methodology because it would more systematically target potential problems at a home and give surveyors new tools to more adequately document care outcomes and conduct on-site investigations. Use of the new methodology could result in survey findings that more accurately identify the quality of care provided by a nursing home to all of its residents.52 Initial testing to evaluate the proposed methodology focused primarily on the first phase and was completed in 49 Quality indicators are derived from nursing homes’ assessments of residents and rank a facility in 24 areas compared with other nursing homes in a state. By using the quality indicators to select a preliminary sample of residents before the on-site review, surveyors are better prepared to identify potential care problems. 50 The agency is committed to implementing only those portions of the new methodology that are proven to be significantly more effective than the current survey methodology. CMS officials said the new process must be manageable and easy to use, add no additional time to surveys, and require limited additional training resources. Given the high turnover among surveyors and state budget constraints, the agency is particularly concerned about imposing new training requirements that would interfere with the conduct of mandatory surveys. 51 A minimum of three residents would be included in the sample for each of the care problems identified in phase one, which covers as many as 33-35 resident-care areas. 52 The goals of the new survey methodology are to (1) ensure that all areas of care are addressed, (2) make the survey process more data-driven and less reliant on surveyor judgment, thus reducing variability in the citation of serious deficiencies, (3) focus surveyors’ attention more on nursing homes with poor quality and less on better performing homes, (4) more reliably determine the scope of deficiencies at nursing homes, that is, the number of residents potentially or actually affected, and (5) produce better documented and defensible survey deficiencies. Page 38 GAO-03-561 Nursing Home Quality three states during 2002. As of April 2003, a CMS official told us that the agency lacked adequate funding to conduct further testing that more fully incorporates phase two. As a result, it is not clear when changes to survey methodology will be implemented. We continue to believe that redesign of the survey methodology, under way since 1998, is necessary for CMS to fully respond to our past recommendation to improve the ability of surveys to effectively identify the existence and extent of deficiencies. While CMS’s goal of not adding additional time to surveys is an important consideration, it should not take priority over the goal of ensuring that surveys are as effective as possible in identifying the quality of care provided to residents. Recognizing inconsistencies in how the scope and severity of deficiencies are cited across states, in October 2000, HCFA began developing more structured guidance for surveyors, including survey investigative protocols for assessing specific deficiencies. The intent of this initiative is to enable surveyors to better (1) identify specific deficiencies, (2) investigate whether a deficiency is the result of poor care, and (3) document the level of harm resulting from a home’s identified deficient care practices. The areas originally targeted for this initiative included deficiencies related to pressure sores, urinary catheters and incontinence, activities programming, safe food handling, and nutrition. Delays have occurred because CMS is committed to incorporating the work of multiple expert panels and two rounds of public comments for each deficiency. The project has been further delayed because the approach used to identify resident harm shifted during the course of work. The process should proceed more quickly, however, now that CMS has developed its approach. CMS expected to release the first new guidance, addressing pressure sores, in early 2003, but officials were unable to tell us how many of the 190 federal nursing home requirements will ultimately receive new guidance or a specific time line for when this initiative will be completed.53 As discussed earlier, CMS’s state performance reviews include an assessment of state surveyors’ documentation of the scope and severity of a sample of deficiencies cited, which should provide CMS with an opportunity to assess the effectiveness of the new guidance. Finally, despite initiation of a complaint improvement project in 1999, CMS has not yet developed detailed guidance for states to help improve their complaint systems. Effective complaint procedures are critical 53 As of July 2003, the guidance had not yet been released. Page 39 GAO-03-561 Nursing Home Quality because complaints offer an opportunity to assess nursing home care between standard surveys, which can be as long as 15 months apart. In 1999, HCFA commissioned a contractor to assess and recommend improvements to state complaint practices. CMS received the contractor’s final report in June 2002, and indicated agreement with the contractor that reforming the complaint system is urgently needed to achieve a more standardized, consistent, and effective process. The study identified serious weaknesses in state complaint processes (see table 5) and made numerous recommendations to CMS for strengthening them. Key recommendations were that CMS increase direction and oversight of states’ complaint processes and establish mechanisms to monitor states’ performance. CMS indicated that it has already taken steps to address these recommendations by initiating annual performance reviews that include evaluating the timeliness of state complaint investigations and the accuracy of states’ complaint triaging decisions, and by developing the new ASPEN complaint tracking system, which should provide more complete data about complaint activities than the current system. The contractor also recommended that CMS (1) expand outreach for the initiation of complaints, such as use of billboards or media advertising, (2) enhance complaint intake processes by using professional intake staff, (3) improve investigation and resolution processes by using available data about the home being investigated and establishing uniform definitions and criteria for substantiating complaints, (4) make the process more responsive by conducting timely investigations and allowing the complainant to track the progress of the investigation, and (5) establish a higher priority for complaint investigations in the state survey agency. CMS noted that some of these recommendations are beyond the agency’s purview and will require the support of all stakeholders to accomplish. CMS told us that it plans to issue new guidance to the states in late fiscal year 2003—about 4 years after the complaint improvement project initiative was launched. As we reported in September 2000, continued federal and state attention is Conclusions required to ensure necessary improvements in the quality of care provided to the nation’s vulnerable nursing home residents. The reported decline in the percentage of homes cited for serious deficiencies that harm residents is consistent with the concerted congressional, federal, and state attention focused on addressing quality-of-care problems. More active and data- driven oversight is increasing CMS’s understanding of the nature and extent of weaknesses in state survey activities. Despite these efforts, however, the proportion of homes reported to have harmed residents is still unacceptably high. It is therefore essential that CMS fully implement Page 40 GAO-03-561 Nursing Home Quality key initiatives to improve the rigor and consistency of state survey, complaint investigation, and enforcement processes. The seriousness of the challenge confronting CMS in ensuring consistency in state survey activities is also becoming more apparent. Our work, as well as that of CMS, demonstrates the persistence of several long-standing problems and also provides insights on factors that may be contributing to these shortcomings: • state surveyors continue to understate serious deficiencies that caused actual harm or placed residents in immediate jeopardy; • deficiencies are often poorly investigated and documented, making it difficult to determine the appropriate severity category; • states focus considerable effort on reviewing proposed actual harm deficiencies, but many have no quality assurance processes in place to determine if less serious deficiencies are understated or have investigation and documentation problems; • the timing of too many surveys remains predictable, allowing problems to go undetected if a home chooses to conceal deficiencies; • numerous weaknesses persist in many states’ complaint processes, including the lack of consumer toll-free hotlines in many states, confusion over prioritization of complaints, inconsistent complaint investigation procedures, and the failure of most states to investigate all complaints alleging actual harm within 10 days, as required; and • states did not refer a substantial number of homes that had a pattern of harming residents to CMS for immediate sanctions. Over the past several years, CMS has taken numerous steps to improve its oversight of state survey agencies, but needs to continue its efforts to help better ensure consistent compliance with federal requirements. Several areas that require CMS’s ongoing attention include (1) the newly established standard performance reviews to ensure that critical elements of the review, such as assessing states’ ability to properly document deficiencies, are successfully implemented, (2) the successful modernization of CMS’s data system by 2005 to support the survey process and provide key information for monitoring state survey activities, (3) the planned expansion of comparative surveys to improve federal oversight of the state survey process, (4) the survey methodology redesign intended to make the survey process more systematic, (5) the development of more structured guidance for surveyors to address inconsistencies in how the scope and severity of deficiencies are cited across states, and (6) the provision of detailed guidance to states to ensure thorough and consistent complaint investigations. Some of these efforts have been under way for Page 41 GAO-03-561 Nursing Home Quality several years, and CMS has consistently extended their estimated completion and implementation dates. We believe that effective implementation of planned improvements in each of these six areas is critical to ensuring better quality care for the nation’s 1.7 million nursing home residents. To strengthen the ability of the nursing home survey process to identify Recommendations for and address problems that affect the quality of care, we recommend that Executive Action the Administrator of CMS • finalize the development, testing, and implementation of a more rigorous survey methodology, including guidance for surveyors in documenting deficiencies at the appropriate level of scope and severity. To better ensure that state survey and complaint activities adequately address quality-of-care problems, we recommend that the Administrator • require states to have a quality assurance process that includes, at a minimum, a review of a sample of survey reports below the level of actual harm (less than G level) to assess the appropriateness of the scope and severity cited and to help reduce instances of understated quality-of-care problems. • finalize the development of guidance to states for their complaint investigation processes and ensure that it addresses key weaknesses, including the prioritization of complaints for investigation, particularly those alleging harm to residents; the handling of facility self-reported incidents; and the use of appropriate complaint investigation practices. To better ensure that states comply with statutory, regulatory, and other CMS nursing home requirements designed to protect resident health and safety, we recommend that the Administrator • further refine annual state performance reviews so that they (1) consistently distinguish between systemic problems and less serious issues regarding state performance, (2) analyze trends in the proportion of homes that harm residents, (3) assess state compliance with the immediate sanctions policy for homes with a pattern of harming residents, and (4) analyze the predictability of state surveys. Page 42 GAO-03-561 Nursing Home Quality We provided a draft of this report to CMS and the 22 states we contacted Agency and State during the course of our review. (CMS’s comments are reproduced in app. Comments and Our X.) CMS concurred with our findings and recommendations, stating that it already had initiatives under way to improve the effectiveness of the Evaluation survey process, address the understatement of serious deficiencies, provide better data on state complaint activities, and improve the annual federal performance reviews of state survey activities. Although CMS concurred with our recommendations, its comments on intended actions did not fully address our concerns about the status of the initiative to improve the effectiveness of the survey process or the recommendation regarding state quality assurance systems. Eleven of the 22 states also commented on our draft report.54 CMS and state comments generally covered five areas: survey methodology, state quality assurance systems, definition of actual harm, survey predictability, and resource constraints. Survey Methodology In response to our recommendation that the agency finalize the Redesign development, testing, and implementation of a more rigorous nursing home survey methodology, under way since 1998, CMS commented that it had already taken steps to improve the effectiveness of the survey process, such as the development of surveyor guidance on a series of clinical issues.55 However, the agency did not specifically comment on any actions it would take to finalize and implement its new survey methodology, which is broader than the actions CMS described. Our draft report noted that, earlier this year, CMS said it lacked adequate funding for the additional field testing needed to implement the new survey methodology. Through September 2003, CMS will have committed $4.7 million to this effort. While CMS did not address the lack of adequate funding in its comments on our draft report, a CMS official subsequently told us that about $508,000 has now been slated for additional field testing. This amount, however, has not yet been approved. Not funding additional field testing could jeopardize the entire initiative, in which a substantial investment has already been made. We continue to believe that CMS should implement a revised survey methodology to address our 1998 54 States that commented included Alabama, Arizona, California, Connecticut, Iowa, Missouri, Nebraska, New York, Pennsylvania, Tennessee, and Virginia. 55 Our draft report discussed the problems CMS encountered in developing this guidance and pointed out that the guidance on the first clinical issue to be addressed, pressure sores, was expected in early 2003. As of July 2003, the guidance had not yet been released. Page 43 GAO-03-561 Nursing Home Quality finding that state surveyors often missed significant care problems due to weaknesses in the survey process. State Quality Assurance We recommended that CMS require states to have a quality assurance Systems process that includes, at a minimum, a review of a sample of survey reports below the level of actual harm to help reduce instances of understated quality-of-care problems. CMS commented on the importance of this concept and noted it had already incorporated such reviews into CMS regional offices’ reviews of the state performance standards. However, the agency did not indicate whether it would require states to initiate an ongoing process that would evaluate the appropriateness of the scope and severity of documented deficiencies, as we recommended. While federal oversight is critical, the annual performance reviews conducted by federal surveyors examine only a small, random sample of state survey reports and should not be considered a substitute for appropriate and ongoing state quality assurance mechanisms. In its comments, New York stated that, in April 2003, it had implemented a process consistent with our recommendation and it had already realized positive results. New York is using the results of these reviews to provide surveyor feedback and expects that instances where deficiencies may be understated will decrease. California also commented that it fully supports this recommendation but indicated that a new requirement could not be implemented without additional resources. State Resource Constraints Officials from five states indicated that resource shortages are a challenge in meeting federal standards for oversight of nursing homes. Alabama commented that there is a relationship among (1) the scheduling of nursing home standard surveys, (2) the number and timing of complaint surveys, (3) the tasks that must be accomplished during each survey, and (4) the resources that are available to state agencies. According to Alabama, the funding provided by CMS is insufficient to meet all of the CMS workload demands, and many of the serious problems identified in our draft report were attributable to insufficient funding for state agencies to hire and retain the staff necessary to do the required surveys. For example, Alabama indicated that the inability of some states to meet survey time frames—maintaining a 12-month average between standard surveys and investigating complaints alleging actual harm within 10 days— is almost always the result of states not having enough surveyors to accomplish the required workload. Page 44 GAO-03-561 Nursing Home Quality Comments from other states echoed Alabama’s concerns about the adequacy of funding provided by CMS. Arizona said that, in order to hire and retain qualified surveyors, it increased surveyor salaries in 2001. Because CMS did not increase the state’s survey and certification budget to accommodate these increases, the state left surveyor positions unfilled and curtailed training to make up for the funding shortfall. Arizona also observed that CMS’s priorities sometimes conflict, further complicating effective resource use. CMS’s performance standards require states to investigate all complaints alleging immediate jeopardy or actual harm in 2 and 10 days, respectively. For budgeting purposes, however, CMS ranks complaint investigations as a lower priority than annual surveys and instructs states to ensure that annual surveys will be completed before beginning work on complaints. California and Connecticut officials said that the growing volume of complaints in their states, combined with limited resources, is a concern. California officials observed that the growth in the number of complaints, coupled with the lack of significant funding increase from CMS, has made it impossible to meet all federal and state standards. They added that they received a 3-percent increase in survey funding from fiscal years 2000 through 2003, but documented the need for a 24-percent increase over this period. As noted in our draft report, the higher priority California attaches to investigating complaints affected survey timeliness—about 12 percent of the state’s homes were not surveyed within the required 15 months. Connecticut indicated that 90 percent of the complaints it receives allege actual harm and require investigation within 10 days, but that with fairly stagnant budget allocations from CMS, its ability to initiate investigations of so many complaints within 10 days was limited. CMS’s fiscal year 2001 state performance review found that Connecticut did not investigate about 30 percent of the sampled actual harm complaints in a timely manner. Although not specifically mentioning resources, New York noted that the increasing volume of complaints was a concern and indicated that any assistance CMS could provide would be welcome. Definition of Actual Harm Comments from four states on our analysis of a sample of survey deficiencies from homes with a history of harming residents revealed state confusion about CMS’s definition of actual harm and immediate jeopardy, a situation that contributes to the variability in state deficiency trends shown in table 2. CMS’s written comments did not address our review of these deficiencies; however, during an interview to follow up on state comments, CMS officials told us that they agreed with our determinations of actual harm as detailed in appendix III. Page 45 GAO-03-561 Nursing Home Quality Arizona and California agreed that some of the deficiencies we reviewed for nursing homes in their states should have been cited at the level of actual harm. However, their disagreement regarding others stemmed from differing interpretations of CMS guidance, particularly the language on the extent of the consequences to a resident resulting from a deficiency.56 For example, Arizona stated that one of the two deficiencies we reviewed could not be supported at the actual harm level because the injuries from multiple falls—including skin tears and lacerations of the extremities and head requiring suturing—did not compromise the residents’ ability to function at their highest optimal level (table 8, Arizona 3). In these cases, it was documented that nursing home staff had failed to implement plans of care intended to prevent such falls. In contrast, California agreed with us that state surveyors should have cited actual harm for similar injuries resulting from falls—head lacerations and a minimal impaction fracture of the hip—due to the inappropriate use of bed side rails (table 8, California 9). CMS officials noted that the definition of actual harm uses the term “well-being” rather than function because harm can be psychological as well as physical. Moreover, they indicated that whether the consequence was small or large was irrelevant to determining harm. CMS central office officials acknowledged that the language linking actual harm to practices that have “limited consequences” for a resident has created confusion for state surveyors and that this reference will be eliminated in an upcoming revision of the guidance. Regarding preventable stage II pressure sores, California stated that guidance received from CMS’s San Francisco regional office in November 2000 precluded citing actual harm unless the pressure sores had an impact on residents’ ability to function.57 According to a California official, this and similar guidance on weight loss was the CMS regional office’s reaction to the growing volume of appeals by nursing homes of actual harm 56 CMS guidance to states in the Medicare State Operations Manual defines actual harm as “noncompliance that results in a negative outcome that has compromised the resident’s ability to maintain and/or reach his/her highest practicable physical, mental and psychosocial well-being as defined by an accurate and comprehensive resident assessment, plan of care, and provision of services. This does not include a deficient practice that only could or has caused limited consequence to the resident.” 57 Stages of pressure sore formation are I—skin of involved area is reddened; II—upper layer of skin is involved and blistered or abraded; III—skin has an open sore and involves all layers of skin down to underlying connective tissue; and IV—tissue surrounding the sore has died and may extend to muscle and bone and involve infection. Page 46 GAO-03-561 Nursing Home Quality citations as well as a reaction to administrative law hearing decisions.58 Prior to this written guidance, which California received in late 2000, it routinely cited preventable stage II pressure sores as actual harm. The guidance noted that small stage II pressure sores seldom cause actual harm because they have the potential to heal relatively quickly and are usually of limited consequence to the resident’s ability to function. We discussed the San Francisco regional office guidance with another regional office as well as with CMS central office officials, who agreed that the San Francisco region’s pressure sore guidance was inconsistent with CMS’s definition of harm, which judges the impact of a deficiency on a resident’s “well-being” rather than functioning. Moreover, central office officials indicated that the regional office’s guidance should have been submitted to CMS’s Policy Clearinghouse for approval. This entity was created in June 2000 to ensure that regional directives to states are consistent with national policy. San Francisco regional office officials indicated that the individual responsible for the guidance provided to California had since left the agency. California also disagreed with our assessment that state surveyors should have cited immediate jeopardy for a resident who repeatedly wandered (eloped) outside the facility near a busy intersection. According to state officials, California’s policy on immediate jeopardy requires the surveyor to witness the incident. A San Francisco regional office official told us that surveyors did not have to witness an elopement to cite immediate jeopardy. An official from a different regional office agreed and noted that repeated elopements suggested the existence of a systemic problem that warranted citation of immediate jeopardy. Although Iowa and Nebraska did not comment specifically on the deficiencies in their surveys that we determined to be actual harm, they did address the definition of harm and the role of surveyor judgment in classifying deficiencies. Iowa officials indicated that a more precise definition of harm is needed because of varying emphasis over the last several years on the degree of harm—harm that has a small consequence for the resident or serious harm. Nebraska commented that we may have based our conclusion that two deficiencies in its surveys should have been cited at the actual harm level on insufficient information because citing 58 Nursing homes can appeal civil money penalties imposed by CMS when they are found to have serious deficiencies. The appeals are decided by the Department of Health and Human Service’s Departmental Appeals Board. Page 47 GAO-03-561 Nursing Home Quality actual harm is a judgment call that varies among state and federal surveyors based on experience and expertise. As noted in our draft report, we found sufficient evidence in the surveys we reviewed to conclude that some deficiencies should have been cited as actual harm because a deficient practice was identified and linked to documented actual harm. Survey Predictability CMS, Arizona, and Iowa commented that nursing home surveys, as currently structured, are inherently predictable because of the statutory requirement to survey nursing homes on average every 12 months with a maximum interval of 15 months between each home’s survey. We agree but believe that survey predictability could be further mitigated by segmenting the surveys into more than one visit, a recommendation we made in 1998 but that CMS has not implemented.59 Currently, surveys are comprehensive reviews that can last several days and entail examining not only a home’s compliance with resident care standards but also with administrative and housekeeping standards. Dividing the survey into segments performed over several visits, particularly for those homes with a history of serious deficiencies, would increase the presence of surveyors in these homes and provide an opportunity for surveyors to initiate broader reviews when warranted. With a segmented set of inspections, homes would be less able to predict their next scheduled visit and adjust the care they provide in anticipation of such visits. CMS also commented that our report captures only the number of days since the prior survey and does not take into account other predictors, for example the time of day or day of the week. Rather than segmenting standard surveys as we earlier recommended, the agency instructed states to reduce survey predictability by starting at least 10 percent of surveys outside the normal workday—either on weekends, in the early morning, or in the evening. It also instructed states to avoid, if possible, scheduling a home’s survey for the same month as its previous standard survey. Though varying the starting time of surveys may be beneficial, this initiative is too limited in reducing survey predictability, as evidenced by our finding that 34 percent of current surveys were predictable. Arizona commented that it was unaware of any CMS guidance to avoid scheduling a home’s survey for the same month of the year as the home’s previous standard survey 59 U.S. General Accounting Office, California Nursing Homes: Care Problems Persist Despite Federal and State Oversight, GAO/HEHS-98-202 (Washington, D.C.: July 27, 1998). Page 48 GAO-03-561 Nursing Home Quality and indicated the state will now incorporate the requirement into its scheduling process. Comments from CMS and Arizona stated that the window of time for a survey to be unpredictable was limited and, as a result, little could be done to reduce predictability. CMS’s technical comments noted that many states have annual state licensing inspection requirements that would limit the window available to conduct surveys to 9 to 12 months after the prior survey, particularly since most inspections are done in conjunction with the federal survey to maximize available resources. CMS, however, was unable to provide a list of such states. None of the 10 states we subsequently contacted had state licensure inspection requirements that would explain their high levels of survey predictability.60 Arizona commented that the state’s licensing inspections are triggered by facilities applying to renew their licenses 60-120 days before their annual license expires. Due to budgetary constraints, Arizona conducts both this state and the federal survey at the same time. While not a requirement, the state strives to complete surveys during this 60-120 day period of time. Thus, nursing homes in Arizona may have some level of control over when federal surveys are conducted, particularly when the state begins complying with CMS guidance to avoid scheduling a home’s survey for the same month as its previous survey. As we reported in September 2000, Tennessee also had an annual licensing inspection requirement that contributed to survey predictability, but the state modified its law to permit homes to be surveyed at a maximum interval of 15 months.61 Since then, the proportion of predictable surveys in Tennessee decreased from about 56 percent to 29 percent. Arizona also stated that surveys had to be conducted within a 45-day window after the 1-year anniversary of the prior survey to be considered unpredictable.62 Arizona’s comments erroneously assume that a survey cannot take place before the 1-year anniversary of the prior survey. There is no prohibition on resurveying a home prior to the 1-year anniversary of its last survey, and many states do so. In fact, 60 We contacted 10 states that were included in our review and that had a significant percentage of predictable surveys—Alabama, California, Connecticut, Maryland, Nebraska, New York, Oklahoma, Tennessee, Virginia, and Washington. As shown in table 10 (see app. V), the proportion of predictable surveys in these states ranged from 29 percent to 83 percent. 61 See GAO/HEHS-00-197. 62 We considered surveys to be predictable if (1) homes were surveyed within 15 days of the 1-year anniversary of their prior surveys or (2) homes were surveyed within 1 month of the maximum 15-month interval between standard surveys. Page 49 GAO-03-561 Nursing Home Quality from October 1, 2000 through September 30, 2001, Arizona conducted 23 percent of its surveys before the 1-year anniversary. CMS provided several technical comments that we incorporated as appropriate. As arranged with your offices, 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 of this report to the Administrator of the Centers for Medicare & Medicaid Services and appropriate congressional committees. We also will 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. Please contact me at (202) 512-7118 or Walter Ochinko, Assistant Director at (202) 512-7157 if you or your staffs have any questions. GAO staff acknowledgments are listed in appendix XI. Kathryn G. Allen Director, Health Care—Medicaid and Private Health Insurance Issues Page 50 GAO-03-561 Nursing Home Quality Appendix I: Scope and Methodology Appendix I: Scope and Methodology This appendix describes our scope and methodology following the order that findings appear in the report. Nursing home deficiency trends. To identify trends in the proportion of nursing homes cited for actual harm or immediate jeopardy, we analyzed data from CMS’s OSCAR system. We compared standard survey results for three approximately 18-month periods: (1) January 1, 1997, through June 30, 1998, (2) January 1, 1999, through July 10, 2000, and (3) July 11, 2000, through January 31, 2002. Because surveys are to be conducted at least once every 15 months (with a required 12-month state average), it is possible that a facility was surveyed more than once in a time period. To avoid double counting of facilities, we included only the most recent survey of a facility from each of the time periods. The data from the two earliest time periods were included in our September 2000 report.1 We updated our earlier analysis of surveys conducted from January 1, 1999, through July 10, 2000, because it excluded approximately 300 surveys that had been conducted but not entered into OSCAR at the time we conducted our analysis in July 2000. Sample of state survey reports. To assess the trends in actual harm and immediate jeopardy deficiencies discussed above, we (1) identified 14 states in which the percentage of homes cited for actual harm had declined to below the national average since mid-2000 or was consistently below that average and (2) reviewed 76 survey reports from homes that had G-level or higher quality-of-care deficiencies on prior surveys but whose current survey had quality-of-care deficiencies at the D or E level, suggesting that the homes had improved.2 All the surveys we reviewed were conducted from July 2000 through April 2002. Our review focused on four quality-of-care requirements that are the most frequently cited nursing home deficiencies nationwide (see table 6). According to OSCAR data, 99 surveys in the 14 states conducted on or after July 2000 documented a D- or E-level deficiency in at least one of these four quality-of-care requirements. We reviewed all such deficiencies in surveys from 13 states but randomly selected 22 surveys from California, which cited the majority (45) of these deficiencies. In reviewing the surveys, we looked for a description of the resident’s diagnoses, any assessment of special problems, and a description of the care plan and physician orders 1 GAO/HEHS-00-197. 2 The 14 states are Alabama, Arizona, California, Iowa, Maryland, Minnesota, Mississippi, Missouri, Nebraska, Pennsylvania, South Carolina, Virginia, West Virginia, and Wisconsin. Page 51 GAO-03-561 Nursing Home Quality Appendix I: Scope and Methodology connected with the deficiency identified. We also looked for a clear statement of the home’s deficient practice and the relationship between the deficiency and the care outcome. Table 6: Quality of Care Requirements Reviewed in a Sample of State Survey Reports Nursing home quality of care requirements Description Necessary care and Facility must provide the necessary care and services for services each resident to attain or maintain the highest practicable well-being. Pressure sores Facility must ensure residents entering facility without pressure sores do not develop sores, unless the individual’s clinical condition indicates the pressure sores were unavoidable, and that residents with sores receive necessary treatment to promote healing, prevent infection, and prevent new sores. Prevention of accidents Facility must ensure each resident receives adequate supervision and assistance devices to prevent accidents. Maintenance of nutrition Facility must ensure each resident maintains acceptable parameters of nutritional status, such as body weight. Source: CMS’s Medicare State Operations Manual. Federal comparative surveys. In September 2000, we reported on the results of 157 comparative surveys completed from October 1998 through May 2000.3 To update our analysis, we asked each CMS region to provide the results of more recent comparative surveys, including data on the corresponding state survey. The regions identified and provided information on the deficiencies identified in 277 comparative surveys that were completed from June 2000 through February 2002.4 Survey predictability. In order to determine the predictability of nursing home surveys, we analyzed data from CMS’s OSCAR database. We considered surveys to be predictable if (1) homes were surveyed within 15 days of the 1-year anniversary of their prior survey or (2) homes were surveyed within 1 month of the maximum 15-month interval between standard surveys. Consistent with CMS’s interpretation, we used 15.9 months as the maximum allowable interval between surveys. Because homes know the maximum allowable interval between surveys, those 3 See GAO/HEHS-00-197. 4 One of the comparative surveys in our updated analysis was completed in May 2000. Page 52 GAO-03-561 Nursing Home Quality Appendix I: Scope and Methodology whose prior surveys were conducted 14 or 15 months earlier are aware that they are likely to be surveyed soon. Complaints. We analyzed the results of CMS’s state performance review for fiscal year 2001 to determine states’ success in investigating both immediate jeopardy complaints and actual harm complaints within time frames required either by statute or by CMS instructions. To better understand the results of state performance as determined by CMS’s review, we interviewed officials from CMS’s 10 regional offices and 16 state survey agencies (see state performance standards below for a description of how these states were chosen).5 We also reviewed the report submitted to CMS by its contractor, which was intended to assess and recommend ways to strengthen state complaint practices.6 Finally, to assess the implementation of CMS’s new automated system for tracking information about complaints, we reviewed CMS guidance materials and interviewed CMS officials and state survey agency officials from our 16 sample states. Enforcement. To determine if states had consistently applied the expanded immediate sanction policy, we analyzed state surveys in OSCAR that were conducted before April 9, 2002, and identified homes that met the criteria for referral for immediate sanction. We included surveys conducted prior to the implementation of the expanded immediate sanction policy because actual harm deficiencies identified in such surveys were to be considered by states in recommending a home for immediate sanction beginning in January 2000. To be affected by CMS’s expanded policy, a home with actual harm on two surveys must have an intervening period of compliance between the two surveys. Because OSCAR is not structured to consistently record the date a home with deficiencies returned to compliance, we had to estimate compliance dates using revisit dates as a proxy. We compared the results of our analysis to CMS’s enforcement database to determine if CMS had opened enforcement cases for the homes we identified. Our analysis compared the survey date in OSCAR to the survey date in CMS’s enforcement database. We considered any survey date in the enforcement database within 30 days of the OSCAR survey date to be a match. CMS officials reviewed and 5 We contacted officials in Alabama, California, Colorado, Connecticut, Iowa, Louisiana, Maryland, Michigan, Missouri, Nebraska, New York, Oklahoma, Pennsylvania, Tennessee, Washington, and Virginia. 6 Center for Health Systems Research and Analysis at the University of Wisconsin, Madison. Page 53 GAO-03-561 Nursing Home Quality Appendix I: Scope and Methodology concurred with our methodology. We then asked CMS to analyze the resulting 1,334 unmatched cases to determine if a referral should have been made.7 State performance standards. To assess state survey activities as well as federal oversight of state performance, we analyzed the conduct and results of fiscal year 2001 state survey agency performance reviews during which the CMS regional offices determined compliance with seven federal standards; we focused on the five standards related to statutory survey intervals, deficiency documentation, complaint activities, enforcement requirements, and OSCAR data entry. Because some regional office summary reports on the results of their reviews for each state did not provide detailed information about the results, we also obtained and reviewed regions’ worksheets on which the summary reports were based. In addition, we conducted structured interviews with officials from CMS, CMS’s 10 regional offices, and 16 state survey agencies to discuss nursing home deficiency trends, the underlying causes of problems identified during the performance reviews, and state and federal efforts to address these problems. We also discussed these issues with officials from 10 additional states during a governing board meeting of the Association of Health Facility Survey Agencies. We selected the 16 states with the goal of including states that (1) were from diverse geographic areas, (2) had shown either an increase or a decrease in the percentage of homes cited for actual harm, (3) had been contacted in our prior work, and (4) represented a mixture of results from federal performance reviews of state survey activities. We also obtained data from 42 state survey agencies on surveyor experience, vacancies, and related staffing issues. 7 CMS determined that for 438 of the 1,334 cases we asked it to examine, the state had indeed made a referral to CMS. In some of these 438 instances, there was no corresponding case in the enforcement database because OSCAR had a different survey date. The “survey date” variable in OSCAR is the latter of the health survey date and the life-safety code survey, while the corresponding date in the enforcement database is usually the health survey date. For others, an enforcement case was already open for the home at the time of the referral, and CMS officials did not open an additional case. There was also a small number of cases where the state agency referred the home for immediate sanction, and CMS chose not to accept the state’s recommendation. States failed to refer 711 cases that met CMS criteria for immediate referral. In addition, CMS did not analyze 155 other cases and was unable to determine the status of 30 cases. Page 54 GAO-03-561 Nursing Home Quality Appendix II: Trends in The Proportion of Appendix II: Trends in The Proportion of Nursing Homes Cited for Actual Harm or Immediate Jeopardy Deficiencies, 1997-2002 Nursing Homes Cited for Actual Harm or Immediate Jeopardy Deficiencies, 1997-2002 Nationwide, the proportion of nursing homes cited for actual harm or immediate jeopardy during state standard surveys declined from 29 percent in mid-2000 to 20 percent in January 2002. From July 2000 through January 2002, 40 states cited a smaller percentage of homes with such serious deficiencies while only 9 states and the District of Columbia cited a larger proportion of homes with such deficiencies.1 In contrast, from early 1997 through mid-2000, the percentage of homes cited for such serious deficiencies was either relatively stable or increased in 31 states. To identify these trends, we analyzed data from CMS’s OSCAR system. We compared results for three approximately 18-month periods: (1) January 1, 1997, through June 30, 1998, (2) January 1, 1999, through July 10, 2000, and (3) July 11, 2000, through January 31, 2002 (see table 7). Because surveys are to be conducted at least once every 15 months (with a required 12- month state average), it is possible that a facility was surveyed more than once in a time period. To avoid double counting of facilities, we included only the most recent survey from each of the time periods. Some of the data in table 7 were included in our September 2000 report.2 However, we updated our analysis of surveys conducted from January 1, 1999, through July 10, 2000, because it excluded approximately 300 surveys that had been conducted but not entered into OSCAR at the time we conducted our analysis in July 2000. 1 The proportion of nursing homes in Utah cited with serious deficiencies remained the same between the two time periods. 2 GAO/HEHS-00-197. Page 55 GAO-03-561 Nursing Home Quality Appendix II: Trends in The Proportion of Nursing Homes Cited for Actual Harm or Immediate Jeopardy Deficiencies, 1997-2002 Table 7: Trends in the Percentage of Nursing Homes Cited for Actual Harm or Immediate Jeopardy during State Standard Surveys, by State Percentage of homes cited for actual harm or immediate a Number of homes surveyed jeopardy Percentage point difference 1/97-6/98 and 1/99-7/00 and State 1/97-6/98 1/99-7/00 7/00-1/02 1/97-6/98 1/99-7/00 7/00-1/02 1/99-7/00 7/00-1/02 Alabama 227 225 228 51.1 42.2 18.4 -8.9 -23.8 Alaska 16 15 15 37.5 20.0 33.3 -17.5 13.3 Arizona 163 142 147 17.2 33.8 8.8 16.6 -25.0 Arkansas 285 273 267 14.7 37.7 27.3 23.0 -10.4 California 1,435 1,400 1,348 28.2 29.1 9.3 0.9 -19.9 Colorado 234 227 225 11.1 15.4 26.2 4.3 10.8 Connecticut 263 262 259 52.9 48.5 49.4 -4.4 0.9 Delaware 44 42 42 45.5 52.4 14.3 6.9 -38.1 District of Columbia 24 20 21 12.5 10.0 33.3 -2.5 23.3 Florida 730 753 742 36.3 20.8 20.1 -15.5 -0.8 Georgia 371 368 370 17.8 22.6 20.5 4.8 -2.0 Hawaii 45 47 46 24.4 25.5 15.2 1.1 -10.3 Idaho 86 83 84 55.8 54.2 31.0 -1.6 -23.3 Illinois 899 900 881 29.8 29.3 15.4 -0.5 -13.9 Indiana 602 590 573 40.5 45.3 26.2 4.8 -19.1 Iowa 525 492 494 39.2 19.3 9.9 -19.9 -9.4 Kansas 445 410 400 47.0 37.1 29.0 -9.9 -8.1 Kentucky 318 312 306 28.6 28.8 25.2 0.2 -3.7 Louisiana 433 387 367 12.7 19.9 23.4 7.2 3.5 Maine 135 126 124 7.4 10.3 9.7 2.9 -0.6 Maryland 258 242 248 19.0 25.6 20.2 6.6 -5.5 Massachusetts 576 542 512 24.0 33.0 22.9 9.0 -10.2 Michigan 451 449 441 43.7 42.1 24.7 -1.6 -17.4 Minnesota 446 439 431 29.6 31.7 18.8 2.1 -12.9 Mississippi 218 202 219 24.8 33.2 19.6 8.4 -13.5 Missouri 595 584 569 21.0 22.3 10.2 1.3 -12.1 Montana 106 104 103 38.7 37.5 25.2 -1.2 -12.3 Nebraska 263 242 243 32.3 26.0 18.9 -6.3 -7.1 Nevada 49 52 51 40.8 32.7 9.8 -8.1 -22.9 New Hampshire 86 83 79 30.2 37.3 21.5 7.1 -15.8 New Jersey 377 359 366 13.0 24.5 22.4 11.5 -2.1 New Mexico 88 82 82 11.4 31.7 17.1 20.3 -14.6 New York 662 668 671 13.3 32.2 32.3 18.9 0.2 North Carolina 407 414 419 31.0 40.8 30.1 9.8 -10.7 North Dakota 88 89 88 55.7 21.3 28.4 -34.4 7.1 Ohio 1,043 1,047 1,029 31.2 29.0 23.7 -2.2 -5.3 Oklahoma 463 432 394 8.4 16.7 20.6 8.3 3.9 Oregon 171 158 152 43.9 47.5 33.6 3.6 -13.9 Page 56 GAO-03-561 Nursing Home Quality Appendix II: Trends in The Proportion of Nursing Homes Cited for Actual Harm or Immediate Jeopardy Deficiencies, 1997-2002 Percentage of homes cited for actual harm or immediate a Number of homes surveyed jeopardy Percentage point difference 1/97-6/98 and 1/99-7/00 and State 1/97-6/98 1/99-7/00 7/00-1/02 1/97-6/98 1/99-7/00 7/00-1/02 1/99-7/00 7/00-1/02 Pennsylvania 811 788 764 29.3 32.2 11.6 2.9 -20.6 Rhode Island 102 99 99 11.8 12.1 10.1 0.3 -2.0 South Carolina 175 178 180 28.6 28.7 17.8 0.1 -10.9 South Dakota 124 112 114 40.3 24.1 30.7 -16.2 6.6 Tennessee 361 354 377 11.1 26.0 16.7 14.9 -9.3 Texas 1,381 1,336 1,275 22.2 26.9 25.5 4.7 -1.5 Utah 98 95 95 15.3 15.8 15.8 0.5 0.0 Vermont 45 46 45 20.0 15.2 17.8 -4.8 2.6 Virginia 279 287 285 24.7 19.9 11.6 -4.8 -8.3 Washington 288 279 275 63.2 54.1 38.5 -9.1 -15.6 West Virginia 130 147 143 12.3 15.6 14.0 3.3 -1.7 Wisconsin 438 428 421 17.1 14.0 7.1 -3.1 -6.9 Wyoming 38 41 40 28.9 43.9 22.5 15.0 -21.4 Nation 17,897 17,452 17,149 27.7 29.3 20.5 1.6 -8.8 Source: GAO analysis of OSCAR data as of June 24, 2002. a Differences are based on numbers before rounding. Page 57 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Survey Reports That Understated Quality-of- Care Problems Our analysis of a sample of 76 nursing home survey reports demonstrated a substantial understatement of quality-of-care problems. Our sample was selected from 14 states in which the percentage of homes cited for actual harm had declined to below the national average since mid-2000 or was consistently below that average. We identified survey reports in these states from homes that had G-level or higher quality-of-care deficiencies (see table 1) on prior surveys but whose current survey had quality-of-care deficiencies at the D or E level, suggesting that the homes had improved. All the surveys we reviewed were conducted from July 2000 through April 2002. Our review focused on four quality-of-care requirements that are the most frequently cited nursing home deficiencies nationwide (see table 6).1 In our judgment, 30 of the 76 surveys (39 percent) from 9 of the 14 states had one or more deficiencies that documented actual harm to residents— G-level deficiencies—and 1 survey contained a deficiency that could have been cited at the immediate jeopardy level. While state surveyors classified these deficiencies as less severe, we believe that the survey reports document that poor care provided to and injuries sustained by these residents constituted at least actual harm. Table 8 provides abstracts of the 39 deficiencies that understated quality problems. 1 According to OSCAR data, 99 surveys in the 14 states conducted on or after July 2000 documented a D- or E-level deficiency in at least one of the quality-of-care requirements we selected. We reviewed all such deficiencies in surveys from 13 states but randomly selected 22 of the 45 California surveys. The 14 states are Alabama, Arizona, California, Iowa, Maryland, Minnesota, Mississippi, Missouri, Nebraska, Pennsylvania, South Carolina, Virginia, West Virginia, and Wisconsin. Page 58 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Table 8: Abstracts of the 39 Nursing Home Deficiencies that Understated Actual Harm from a Sample of 76 Nursing Home Survey Reports Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor Alabama-1 Provide necessary Resident admitted to Site of gastrostomy tube Facility failed to provide proper November 2001 care and services: facility 5/15/01with a insertion became reddened care and services: daily D fractured hip; a with thick yellow-green cleaning and application of a gastrostomy tube was drainage, and had an odor, drain sponge around the inserted through the indicating signs of infection, on gastrostomy tube. abdomen into the stomach 11/7/01. Family indicated no one to maintain feeding. On changed the dressing. There is 10/9/01, resident was no documentation to show hospitalized for abdominal resident’s gastrostomy tube pain and signs of infection site was cleansed as ordered related to the gastrostomy 12 out of 16 opportunities. tube. On return to facility, physician orders state, “clean G tube site with soap and water, apply a drain sponge.” Alabama-5 Provide Resident 1 admitted to Resident 1 sustained four skin The facility failed to March 2001 supervision and facility 11/6/00 with tears on right arm and leg and consistently reassess for devices to prevent diagnoses of stroke, multiple bruises to both legs preventive measures to accidents: D pressure sores, and from 1/16/01 to 3/21/01. address the problem of skin kidney failure. On tears and bruises for both 11/16/00, resident was residents. Staff were unable to noted to have abrasions provide documentation of and bruises. preventive interventions. Resident 2 was admitted Resident 2 sustained seven to the facility 11/23/98 with skin tears and bruises to legs anemia, depression, from 12/29/99 to 10/9/00. urinary incontinence, and a history of falls. She was identified as having a problem with skin tears and bruising.c Page 59 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor Arizona-3 Ensure prevention Resident admitted to On 7/5/00, it was noted that the The necessary services and July 2000 and healing of facility 08/24/99 with heart resident had developed a stage care to promote healing and pressure sores: D failure, high blood IV pressure sore. prevent worsening of existing pressure, paraplegia, and pressure sore were not a stage II pressure sore on provided. Even after the lower back.d Pressure sore pressure sore progressed to remained a stage II until stage IV and a physician May 2000, when wound ordered that the resident be was documented to be a turned every hour, the staff stage III. failed to turn the resident as directed. Surveyor observed resident lying on her back for 2 or more hours. Resident stated that frequently she was turned only twice in 8 hours. Charge nurse did not know physician had ordered resident to be turned every hour. Arizona-3 Ensure adequate Resident 1 admitted to the Resident 1 fell four times and Facility staff failed to July 2000 supervision to facility 4/7/00 with sustained skin tears, abrasions, implement a plan of care that prevent accidents: diabetes, partial paralysis and lacerations. called for identifying resident D of left side, and inability to as a fall risk by placing a star speak. Resident also had on his door by his name. No a history of spinal other preventive measures fractures, and a fall were identified, and surveyor prevention plan was observed no star next to developed on 4/15/00. resident’s name outside his door. Resident 2 admitted to the Resident sustained 12 falls Although resident was facility 12/10/97 with from 2/18 to 7/8/00 with identified as at risk for falls in a dementia, painful joints, lacerations of extremities and care plan of 4/22/00, the and visual problems. A head requiring suturing and facility staff failed to develop 7/13/00 assessment with other cuts and bruises. approaches to prevent falls indicated resident was even though the resident cognitively impaired and continued to fall and injure had a mental function that herself. varied throughout the day. She was also identified as a wanderer. California-2 Ensure prevention Resident 1 with leg Resident 1 developed a The surveyor found that the September 2000 and healing of contractures (permanent reddened open area .3 cm. in facility did not identify, pressure sores: D tightening of muscle, diameter, (stage II pressure document, or provide tendons, ligaments, or skin sore) on left lower back by intervention to prevent this that prevents normal 9/23/00. facility-acquired pressure sore. movement) was noted to The reddened area noted was have a small reddened not documented in the medical area on left lower back on record 9/20-9/22/00. 9/20/00. Page 60 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor Resident 2 was admitted Resident 2 developed a stage II The facility developed a to facility on 2/2/00. Family pressure sore. nursing care plan for identified resident as prevention of pressure sores having a “skin problem” on and turning the resident every 9/17/00. 2 hours on 9/8/00. The family identified a stage II pressure sore on 9/17/00. The surveyor found no evidence that the care plan was implemented at time of survey. Resident 3 admitted to Seven days after admission, The facility failed to prevent a facility 9/20/00 with resident 3 was noted to have rapid decline in resident’s diagnoses of multiple four stage II pressure sores on condition and occurrence of sclerosis, bilateral right and left shoulder blades facility-acquired pressure fractures of the femur, and and right buttock and three sores. Staff said they were obesity. Resident was stage I pressure sores on the unable to turn resident (a unable to turn herself in left buttock. larger bed and mattress were bed; physician not provided, which would documented resident had have facilitated turning). No no areas of skin pressure-relieving devices and breakdown and ordered staff assistance in getting out resident to be up in a of bed were provided. In the 7 wheel chair two to three days after admission, the times a day. resident was out of bed only once, at which time the pressure sores were discovered. California-2 Maintain nutritional Resident admitted to Resident’s weight was recorded Facility failed to provide a September 2000 status: D facility 7/7/00 with a as 77 pounds 1 month after comprehensive nutritional diagnosis of failure to admission. Resident sustained assessment to meet resident’s thrive and a recorded a severe loss of 12 pounds (13 nutritional needs in order to weight of 89 pounds. percent) between July and maintain body weight. August. California-5 Provide Resident was identified as Resident fell while walking Facility failed to develop and February 2001 supervision and at high risk for falls in 5/00. unassisted on 6/21/00 and implement a fall prevention devices to prevent again on 2/22/01, fracturing his plan when resident was accidents: D right hip each time. identified as being a high risk for falls and after the first hip fracture. Page 61 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor California-6 Provide Resident admitted to Resident wandered to an area Facility failed to provide May 2001 supervision and facility on 2/12/01with 100 yards from facility near two supervision and devices to devices to prevent dizziness, fainting, poor busy intersections on 3/26/01 prevent accidents even after accidents: D vision, and cognitive and again on 5/19/01. resident was found wandering impairment. Care plan of outside the facility on 2/20/01. 2/20/01 identified resident According to CMS, the failure of The facility did not immediately as a wanderer and at risk a facility to provide supervision implement procedures cited in for falls. Interventions of a cognitively impaired the care plan to supervise the suggested were visual individual with known risk for resident and prevent accidents checks every 2 hours and wandering is considered failure and wandering, nor did the involvement of resident in to prevent neglect and places facility implement existing facility activities. On the resident in immediate facility policies to prevent 2/20/01 at 9:30 pm jeopardy for death or serious wandering and injury. resident was found injury during such an incident. wandering outside on the patio and had fallen and sustained abrasions. California-8 Ensure prevention Resident admitted to Resident sustained a facility- Facility failed to ensure June 2001 and healing of facility in 1996 with stroke, acquired stage IV pressure necessary treatment and pressure sores: D paralysis of lower right sore of the right ankle service to promote healing and side, and senile dementia. measuring 7 cm. by 5 cm. prevent infection of the Physician orders of 4/5/01 pressure sore. Surveyor called for an air mattress. observed on 6/20 and 6/21/01 Assessment of 4/24/01 that there was no air mattress noted resident had a stage on resident’s bed and on IV pressure sore on the 6/20/01 that inappropriate right outer ankle. On technique was used in 5/17/01, physician ordered changing the dressing on the cleansing of the wound resident’s ankle. with saline and an anti- infective solution, dressing it with soft protective gauze. California-8 Ensure Resident admitted to Resident weighed 98.4 lbs and Facility failed to ensure that June 2001 maintenance of facility in 1990 with a experienced a severe weight the resident maintained nutritional status: diagnosis of stroke and loss of 13 pounds (12 percent) adequate nutrition. It did not D inability to speak. A 3/7/01 in 3 months. monitor the amount of assessment noted erosive nutritional supplements gastritis, anemia, and consumed by the resident and weight of 111 lbs. The inconsistently recorded county was the weights, often without conservator and requested associated dates. It did not maximum treatment. notify the physician of the Resident was placed on resident’s weight loss. an enriched pureed diet with supplemental feedings three times daily. Page 62 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor California-9 Provide Resident 1, 48 years old, Resident fell when trying to The facility failed to supervise December 2000 supervision and admitted to facility after a climb over side rails, sustaining the resident and prevent devices to prevent stroke with incontinence, a laceration to his head. accidents from occurring: staff e,f accidents: B inability to speak, right- failed to accurately assess side paralysis, and resident’s safety needs and functional use of his left inappropriately assumed side. Resident resident needed full side rails communicated by signs on the bed. and sounds. Resident 2 had a history of On 3/29/00, resident climbed The facility failed to provide a right hip fracture, chronic over the bed side rails and was supervision and appropriate weakness in both legs, found on the floor at the foot of interventions to prevent this and dementia. Resident his bed with both side rails in resident’s fall. According to the had a physician’s order the up position. Seven hours surveyor, there were no orders (9/16/99) for soft belt later, an x ray was taken and for restraints in bed and no restraints when in found that resident had a indication that all reasonable wheelchair to prevent “minimal impaction fracture” of efforts had been made to resident from getting up the left hip. safeguard the resident from from wheelchair without additional injuries. assistance. Because restraints, including side rails, can pose a serious health and safety risk to nursing home residents if used improperly, CMS requires that restraints should only be used when other, less severe alternatives fail to address a resident’s medical needs, and the benefits outweigh the potential risks. In such cases, the nursing home must ensure that any restraints are used safely and properly. Page 63 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor California-9 Ensure Resident was readmitted A stage IV pressure sore on Facility was slow to implement December 2000 maintenance of (6/11/00) to facility right heel was noted on the dietician’s nutritional following the removal of a 7/27/00. recommendations of 6/15/00 status: D hip prosthesis and a for caloric, protein, and water surgical incision that intake necessary for wound became infected with a healing. Diet ordered on fungus, resulting in a large 6/20/00. On 6/24/00 resident gaping wound. Resident was admitted to the hospital was unable to swallow for care of gastrointestinal following a stroke and was bleeding and found to need fed via a nasogastric tube. nutritional supplements to address gastrointestinal bleeding and promote wound healing. Resident was readmitted to facility on 6/29/00. Following readmission, the facility also failed to implement both the hospital’s and its own dietician’s recommendations for increased protein, calories, and water to encourage wound healing. California-10 Provide Resident admitted to Resident fell while attempting to Facility failed to provide May 2001 supervision and facility with diagnoses of get out of bed and lacerated left supervision and devices to devices to prevent dementia and Alzheimer’s elbow. prevent accidents. Specifically, accidents: D disease and a history of resident was put to bed falls, confusion, and without a restraining belt. unsteady gait. Resident identified as high risk for falls and had a physician’s order for a restraining belt when in bed. California-11 Provide necessary Resident admitted to the Resident admitted to hospital Staff failed to implement the May 2001 care and services: facility in 1999 with for “several days” to relieve a care plan. On 5/23/01 the D dementia and neurological fecal impaction. surveyor noted the resident disorders. Resident was crying out, moaning, receiving an antipsychotic grimacing, and moving her medication that has a side arms and legs about. Last effect of constipation. Care bowel movement recorded plan of 1/04/01 called for was on 5/19/01. The charge (1) providing liquids, nurse administered Tylenol roughage, and exercise, with codeine for what she (2) monitoring for believed was an earache at 10 abdominal distention, pain, a.m. Resident continued to cry cramps, nausea, and out and the charge nurse vomiting, and (3) checking called the physician who had for impaction every 3 days. the resident transferred to a hospital emergency room. Page 64 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor California-11 Provide Resident was admitted Resident sustained a 9 cm. skin Facility failed to develop skin supervision and 4/25/01 with acute kidney tear to the lower left leg on tear prevention plans. Staff did devices to prevent failure and emphysema 4/28/01 and two 3 cm. skin not fully investigate causes of accidents: E and was one of five tears below the left knee on the tears and did not know residents identified as 5/3/01. Four other residents how to prevent skin tears. The being at risk for skin tears; also sustained multiple skin staff development director all five developed skin tears to their extremities and stated that she had never tears. A care plan for hip. provided instruction for the potential for skin certified nurse aides on breakdown and treatment prevention of skin tears. of the skin tears was developed. California-14 Ensure prevention Resident admitted to Resident’s pressure sore Facility staff failed to promote March 2001 and healing of facility 1/26/01 following a progressed to a stage II by healing or prevent worsening pressure sores: D stroke, with inability to 2/28/01 and a stage III on of pressure sore by failing to swallow, a gastric tube in 3/7/01. employ the appropriate sheets place for feedings, and a that are used in conjunction stage I pressure sore on with the low-air-loss, pressure right hip. sore mattress, thereby negating the pressure-relieving benefits of the mattress. California-16 Ensure prevention Resident admitted to Resident developed a new Facility staff did not prevent April 2001 and healing of facility 11/16/98 with stage II pressure sore on the development of a facility- pressure sores: D dementia, anemia, 4/26/01. acquired pressure sore. irregular heartbeat, Specifically, the surveyor diabetes, high blood observed on 4/24/01 that the pressure, and difficulty in staff did not turn resident every swallowing. 2 hours as directed by the care plan, and left her in the same position for as long as 8 hours. California-18 Provide necessary Resident admitted to the Resident was observed Facility staff failed to assess April 2001 care and services: facility with a steel plate screaming and writhing in the resident’s pain levels after E implanted in her back unrelieved pain for greater than decreasing her Methadone. following a fracture. an hour. They did not do an in-depth Nursing care plan called pain assessment at any time for comfort measures for after admission. The surveyor back pain, such as observed the staff ignoring the heat/cold application, resident’s cries for help and therapeutic touch, and relief, which continued until the staying with resident when surveyor intervened. she was in distress. Resident also had an order for Methadone 20 mg. that had been reduced to 2.5 mg. Page 65 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor California-19 Provide necessary Resident admitted to As a result of the facility’s Facility staff did not reassess June 2001 care and services: facility on 3/97 with stroke, failure to address the resident’s this resident’s pain level and D one-sided paralysis, and pain, the resident refused the need for stronger pain relief. moderate contractures of splints used to control the upper and lower contractures and the extremities. Resident took contractures worsened, leading Tylenol four times a day to greater pain. since 2/98 for pain. As his pain worsened, he began to refuse the splinting of his contracted extremities because it was too painful. California-20 Provide Resident was admitted to Resident fell and sustained Facility failed to implement January 2001 supervision and facility on 3/6/00 and abrasions to her right flank and care plan of 12/19/00 that devices to prevent identified as a high risk for hip on 12/24/00 and again on called for safety assessment accidents: D falls on 12/6/00 because of 1/7/01, sustaining a scalp and rehabilitation screening resident’s failure to laceration on the back of her related to falls. In addition, remember warnings about head. facility failed to reassess personal safety and poor resident’s safety needs and safety awareness. alternative preventive measures after the two falls, as called for by facility policy and the care plan. Physical therapy staff did not assess resident for safety needs either. There was no documented evidence that a plan was implemented to prevent future falls. California-22 Provide Resident had diagnoses of Resident fell 17 documented Facility failed to provide October 2000 supervision and diabetes, bipolar disease, times from 4/21 to 10/14/00, supervision and prevent devices to prevent and high blood pressure. when she sustained a bruising accidents. Specifically, facility accidents: D Resident was assessed as of the right eye, and a bruise staff did not provide a self- at risk for falls. and an abrasion to her releasing seat belt or pressure forehead. sensitive alarm on resident’s wheelchair as recommended by the facility’s fall/risk committee. Although the MDS assessment of 9/4/00 indicated that the resident had no falls for 180 days, the resident’s medical record indicated that the resident fell at least six times in this period. Page 66 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor Iowa-1 Ensure prevention Resident 1 had diagnoses Resident’s stage II pressure Facility staff failed to provide June 2001 and healing of that included renal failure, sores healed and then appropriate treatment to pressure sores: D diabetes, and dementia. reopened repeatedly from prevent reoccurrence of Resident’s record noted 1/9/01 to 6/20/01. pressure sores, resulting in the the presence of two reappearance of pressure pressure sores, one on sores after they had resolved. 1/9/01 and the second on Specifically, the facility did not 4/1/01, between the reassess the current plan of buttocks and on the lower treatment and did not modify right back, respectively. the care plan to meet the needs of the resident. Resident 2 had a history of Resident developed an infected Facility staff failed to prevent stroke and dementia. A stage II pressure ulcer at the an avoidable pressure sore. 4/20/01 assessment note base of the right thumb. After the resident was indicated that the resident readmitted with the cast on his had no ulcers, skin arm, the staff did not assess problems, or lesions. On whether the skin around the 4/22/01, the resident fell, cast was intact for 18 days was admitted to the (4/27-5/14/01), at which time hospital for treatment of a the nurse noted a foul odor fracture of the right wrist, and a reddened thumb. and was readmitted to nursing home on 4/27/01 with a cast on the right arm, including the lower half of the hand and thumb. Iowa-2 (1) Ensure On 2/25/02, surveyor Resident developed a stage II Facility staff failed to ensure March 2002 prevention and observed resident being pressure sore that persisted that a resident with a pressure healing of transferred using a and reopened after resolving. sore received necessary pressure sores: D mechanical lift and noted treatment to promote healing an open stage II pressure and to prevent new sores from sore on the lower back. A developing. Specifically, the record review revealed a record lacked evidence of history of healing and assessment of potential causal reoccurrence of a lower- factors and interventions to back pressure sore on prevent the reoccurring several occasions from pressure sore. 7/8/01 through 2/26/02. Page 67 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor (2) Provide During the above cited Resident sustained multiple Facility failed to prevent supervision and observation of the same bruises, skin tears, and bruises and skin tear injuries. devices to prevent resident on the mechanical scrapes. The staff did not assess the accidents: D lift, the surveyor also noted cause of the injuries or bilateral purple bruises on implement protective devices, the resident’s lower legs such as padding of the lift and and later checked the wheelchair. On 2/26/02, a staff resident more fully and member stated that the noted a total of five bruises probable cause of the bruises and a scrape to the legs. A was the resident’s hitting the review of the resident’s mechanical Hoyer lift during record revealed multiple transfers and that the lift bruises, abrasions, and should be padded. skin tears going back 1 year. The surveyor observed that there was no padding on the mechanical lift. Iowa-4 Provide necessary Resident with a diagnosis Surveyor noted bruises on Facility staff failed to provide February 2001 care and of multiple sclerosis resident’s legs and saw how the necessary care and services: E required extensive resident’s legs and feet were services in accordance with assistance with transfers, twisted between the wheelchair the plan of care. Staff failed to walking, and other pedals and dragged and assess for risk of skin injury activities of daily living. bumped against the wheelchair from wheelchair transfers and Care plan of 1/19/01 on 1/30 and 1/31/01. Resident to protect resident from harm directed staff to monitor sustained multiple bruises on during transfers. Staff also and record all skin both lower legs. failed to document resident’s changes. Surveyor noted bruises. multiple bruises on resident’s legs. Iowa-5 Provide necessary Resident admitted to Resident fell five documented Facility failed to properly March 2001 care and facility on 7/6/99 with times, sustaining abrasions to assess and monitor after the services: D Alzheimer’s disease, high the forehead, a bloody nose resident fell, striking her head blood pressure, and and mouth, a bump to the on all five occasions. There anemia. Resident was forehead, a broken tooth, a was no documentation of receiving a diuretic to carpet burn of the knees, and a weekly monitoring of blood reduce blood pressure and broken nose. pressure or for neurological an antihistamine for status after resident struck her itching. Both drugs can head. reduce blood pressure below normal levels, causing dizziness or a drop in blood pressure when rising to stand (orthostatic hypotension). Resident’s plan of care called for staff to monitor blood pressure on a weekly basis. Page 68 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor Iowa-7 Provide necessary Resident 1 admitted to Resident developed two stage Facility staff did not August 2001 care and services: facility on 3/2/01 with II ulcers of the foot and ankle, consistently follow the orders D history of stroke, heart one on 6/18/01 and the other and provide the necessary failure, and poor on 6/26/01, which were still care for the resident. circulation, with related present, unhealed, on 8/7/01. According to the surveyor, the rash of the legs and feet. skin and heel protectors were Assessment revealed a left off and the wheelchair was small scab on the left not padded and was causing ankle that healed by 5/01. additional erosion of the ankle Resident developed a lesions. scabbed area on right foot. The physician ordered skin and heel protectors to be worn at night on 5/29/01. Resident 2 was admitted Resident 2 experienced severe Facility staff failed to provide with lung cancer, unrelieved pain. the necessary care for this degenerative arthritis, resident to maintain comfort osteoporosis, and anxiety. measures and avoid pain. The Physician’s note of 5/16/01 care plan of 5/21 and 6/13/01 indicated that resident was did not include pain dying and would need to management. The staff did not be assessed for pain relief assess the resident’s as the disease progressed complaints of pain and need and that stronger, more for effective pain relief. effective pain relievers would be considered. As the resident began to experience increasing pain, he was given Tylenol even when pain appeared severe and unrelieved. Iowa-7 Provide Resident 1 has diagnoses Resident 1 fell 11 times and The facility failed to provide August 2001 supervision and of dementia and sustained a fractured wrist, adequate interventions to devices to prevent depression with long- and three fractured ribs, bruises, prevent accidents. The accidents: D short-term memory abrasions, and a skin tear, plus personal alarm system was deficits. Surveyor noted pain associated with all these the only safety device resident had fallen falls and injuries. employed, and there is no frequently from 2/23/01 evidence that the staff through 7/23/01 and evaluated its effectiveness and sustained serious injuries. selected other measures. Personal safety alarms selected for resident were ineffective in preventing falls. Page 69 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor Resident 2 was admitted Resident 2 fell 21 times from The facility failed to provide to facility on 8/8/00 with 1/6/01 to 6/26/01 and sustained adequate interventions to renal failure and impaired multiple skin tears, two prevent accidents. The mobility. On 4/3/01, he lacerations to the head and personal alarm unit in use for was assessed as being elbow requiring emergency this resident did not prevent mentally confused at room or clinic visits for sutures, his falls from occurring and times. Surveyor noted the multiple bruises and abrasions, there is no indication that other resident’s record stated and head injuries. safety options were that resident fell considered. frequently. The care plan and monthly summary for April identify the personal alarm unit as the safety device in use during this time (initiated 3/25/01). The resident frequently removed the unit or put it in his pocket. Maryland-1 Provide Resident admitted to Resident fell out of the The facility failed to provide August 2001 supervision and facility with multiple wheelchair, was bleeding from supervision and devices to devices to prevent diagnoses including nose and mouth, and was in prevent accidents by not accidents: D congestive heart failure, acute respiratory distress. Staffplacing safety belt around high blood pressure, and did not intervene to address resident while she was in the obesity. Resident suffered respiratory distress until wheelchair. Staff also did not from shortness of breath resident stopped breathing and provide the resident with and required oxygen at 3 her pulse stopped. At this time oxygen as ordered while she liters per minute. She also the staff began to administer was in the wheelchair. Staff had a history of falls and cardiopulmonary resuscitation did not respond in a timely and was considered a high risk (CPR). appropriate manner to for falls. Resident had a resident’s onset of respiratory physician order for a distress following the fall from quick-release belt while in the wheelchair. Staff did not wheelchair for safety. initiate CPR until resident was no longer breathing and her pulse stopped. Missouri-3 Ensure adequate Resident had diagnoses of Resident experienced another The facility failed to ensure May 2001 nutritional status: peptic ulcer disease, severe weight loss, dropping adequate nutritional status. D aspiration pneumonia, and from 126 lbs in 3/01 to 116.9 After noting resident’s weight a penicillin-resistant lbs in 4/01, a loss of 7.2 percent loss in 2/01, no care plan was infection requiring long- in 1 month. developed to address the term antibiotic treatment. weight loss. In March, the From 11/00 through 2/01, dietician recommended a resident sustained a dietary supplement, which did severe weight loss of 10 to not begin for a month. 12 percent. Page 70 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor Nebraska-1 Provide necessary Resident 1 readmitted to Over a period of 9 months, Facility failed to provide the September 2000 care and facility from hospital with a resident’s blood sugar necessary care and services services: D diagnosis of insulin- fluctuated, including frequent required to manage resident’s dependent diabetes. episodes of symptomatic diabetes. Specifically, (1) the Physician orders stated hypoglycemia (low blood sugar staff infrequently called the that the physician was to between 48 and 60) and loss of physician about blood sugars be called when resident’s consciousness. below 40, the frequent blood blood sugar fell below 40 sugar fluctuations, or the or rose above 350 (normal resident’s episodes of range is 70 to 110). symptomatic hypoglycemia, Resident received insulin (2) fluctuating blood sugars on a sliding scale (insulin were not identified as a dose based on most problem in the care plan, and recent blood sugar), and a (3) there was no assessment variety of dietary of the resident’s diabetes, interventions. appropriate diet, treatment effectiveness of hypoglycemic episodes, and administration of insulin on a sliding scale. Resident 2 with diagnoses This terminally ill resident Facility staff did not provide of emphysema, suffered with unrelieved pain the necessary care and Parkinson’s disease, and for at least 4 months. services to this resident. The osteoarthritis was staff did not assess or respond receiving hospice services. to the resident’s continuing Resident experienced complaints of pain and noted increasing pain on a daily in the record that the resident basis, unrelieved by was demanding and regular Tylenol, a manipulative. Nor did they tranquilizer, and an monitor the effectiveness of antipsychotic drug specific the medications administered, for schizophrenia and resulting (according to the mania. Resident obtained surveyor) in the resident’s short-term (2.5 hours) voicing thoughts of suicide. relief from Tylox (Tylenol and oxycodone for pain relief and sedation). Page 71 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor Nebraska-3 Ensure prevention Resident was readmitted Resident developed a stage III Facility failed to ensure that a September 2001 and healing of to facility 5/24/01 with pressure sore on the right heel resident did not develop a pressure sores: D diagnoses of stroke, with thick green drainage and pressure sore in the facility. diabetes, and one stage II foul odor. Specifically, the facility staff pressure sore of the lower failed to recognize the back and one stage I challenge the resident had in pressure sore between the moving in bed because of the buttocks. Resident was right-sided paralysis. In totally dependent on staff addition, they were slow to use for bed mobility because of a pressure-reducing mattress. a right-sided paralysis and When the mattress was placed developed pressure sores on the bed the staff did not of both heels that were discontinue use of the fleece- noted on 6/3/01 and lined protection booties and identified as stage II on continued use for 3 weeks, 7/24/01. A pressure- which negated the pressure- reducing mattress was reducing effects of the added to the care plan on mattress. 9/4/01. Pennsylvania-3 Ensure prevention Resident had a left hip In addition to the stage II Facility failed to prevent the May 2001 and healing of fracture and was identified pressure sore of the foot, development of pressure pressure sores: D as high risk for skin resident developed a second sores. Specifically, the boot, breakdown on 12/18/00. A stage II facility-acquired which was left on continuously, stage I pressure sore of pressure sore on 4/10/01. contributed to the development the left heel was noted on of the pressure sore identified 3/7/01 and by 3/14/01 it on 4/10/01. In addition, the had progressed to stage II. dietician did not note the A special boot to keep left existing original pressure sore heel elevated was not and wrongly assumed the applied until 3/21/01 and resident had no extra need for was then left on protein. The need for continuously. A second additional protein in the diet stage II pressure sore was was confirmed by laboratory noted on the left outer foot tests indicating the resident’s 4/10/01. The boot was protein levels were below the discontinued on 4/11/01. A normal range. nutrition assessment on 3/27/01 indicated resident’s skin was intact and recommended no increase in protein in the diet. Page 72 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor Pennsylvania-3 Provide Resident had piriformis Resident developed a second- Facility staff failed to provide May 2001 supervision and syndrome (compression of degree burn of the right supervision and prevent injury. devices to prevent the sciatic nerve by the buttock, which blistered and During a routine check on accidents: E piriformis muscle) with a was still healing after a month. 1/9/01, the facility found that physician’s order for the temperature on the physical therapy using hydrocollator pack was 11 stretching exercises and degrees above the heat application. Physical manufacturer’s recommended therapy used a temperature. On 4/16/01 the hydrocollator pack to hydrocollator pack was applied provide moist heat to the resident’s right buttock. treatments.g Resident said that he told the therapy staff that the pack was getting too hot and the pack was removed. Facility staff did not check the water temperature after the incident. Resident 2 had diagnoses Resident 2 fell nine The facility failed to ensure that included dementia, documented times and, as a adequate supervision and poor vision, and result of these falls, sustained a assistance devices to prevent Parkinson’s disease and skin tear, a laceration requiring accidents. According to the was assessed as a transfer to the hospital for surveyor, there was no moderate risk for falls on treatment, and a dislocated hip evidence that the facility had 12/29/00. The MDS requiring another hospital visit. implemented effective significant change interventions to avoid the risk assessment of 1/24/01 of such accidents for the and the 4/9/01 quarterly resident. The surveyor noted review noted a history of that this at-risk resident’s room falls, impaired decision was too far from the nurses’ making, and the need for station, making observation assistance for transferring difficult. and walking. The records noted interventions found to be ineffective continued to be used. Pennsylvania-9 Provide A dependent resident with Resident sustained eight skin Surveyor stated that the facility May 2001 supervision and cognitive impairment was tears on 6/27/00, 7/24/00, failed to ensure that the devices to prevent assessed as at risk for 7/31/00, 8/16/00, 9/20/00, necessary safety measures accidents: D falls and skin tears. 10/24/00, 1/8/01, and 1/27/01. and/or devices were Interventions to prevent implemented and failed to falls listed in the care plan adequately assess the included use of personal ongoing use of these devices alarms, protective sleeves, given their ineffectiveness in and padded side rails. preventing falls and skin tears. Page 73 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor Virginia-1 Provide necessary Resident admitted to This resident suffered with The facility did not provide August 2000 care and facility for pain severe pain that was necessary care and services services: D management associated incompletely relieved by the to manage this resident’s pain. with spread of cancer to use of Percocet. The longer Resident did not receive any of the spine. Resident had acting Oxycontin was never the longer-acting Oxycontin physician orders for used. and received only 10 doses of Oxycontin every 12 hours the Percocet during the 6 days for long-term pain relief, as he was in the facility. He was needed, and Percocet not offered pain relief in the every 4 hours for any morning when he was being additional pain, as needed. turned and bathed. Monitoring Staff noted resident lay of medication effectiveness very still in bed and was incomplete. Percocet was seldom asked for pain given, on average, once a day. medication but that it was obvious he was in a lot of pain whenever he was turned or touched. Resident’s daughter said her father was in constant pain and was depressed. Virginia-2 Provide necessary Resident was admitted to Resident sustained fractures of The facility failed to provide March 2001 care and facility 11/4/97, with the eighth and ninth ribs with the necessary care and services: D diagnoses of stroke, fluid in the left lower lobe of the services to provide prompt depression, and delusions. lung demonstrated by x ray. treatment of the resident’s An MDS of 11/9/00 chest injury. Specifically, the indicated the resident was facility failed to take cognitively impaired and appropriate action to assess required lift transfer. On and provide the necessary 12/27/00 the nurse noted a care for this resident’s injury large area of bruising on for 11 days. The results of an the left chest and left investigation implicated the lift underarm with swelling used to transfer the resident to around the rib cage. On and from the bed. 1/6/01 resident began to experience shallow breathing. Physician ordered a chest x ray if resident’s breathing difficulties continued. Page 74 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor Virginia-2 Ensure prevention Resident 1 admitted to the Resident developed three open The facility failed to prevent March 2001 and healing of facility with diagnoses of pressure sores of the buttocks, the development of facility- pressure sores: D Alzheimer’s disease, evident 2 days after the MDS acquired pressure sores. The anemia, depression, and assessment. One of the staff did not obtain timely joint pain. No pressure pressure sores was a stage III. alternative treatments and sores were noted on the interventions to promote admission assessment healing of early pressure form. The care plan on sores. 2/22/00 noted the resident was incontinent of bowel and bladder and at risk for pressure sores. Resident’s blood protein was low. The most recent MDS (2/23/01) indicated no pressure sores but noted the resident was losing weight, 5 percent or more in the past 30 days (1/24/01- 2/23/01). Resident 2 admitted to Resident developed an open Staff failed to obtain timely facility on 12/24/00 with stage III pressure sore with alternative treatments and diabetes, stroke, prostate yellow drainage. interventions to promote cancer, requiring limited healing upon worsening of assistance for activities of these sores from1/18/01 daily living, and incontinent through 3/1/01. Specifically, of bowel and bladder. As the staff continued to treat the of 12/31/00 resident had pressure sores without an unhealed surgical evaluating the effectiveness of wound of the back, two the treatment. stage IV pressure sores of the right and left heels, and an excoriated (stage I) buttock. After a brief hospitalization, resident was readmitted to facility and the clinical record on 2/26/00 described the buttock sore as a stage II pressure sore. Treatment with a sealed dressing continued. Page 75 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems Requirement and State and date scope and Resident description and Actual harm to resident Deficiencies in care cited by a b of survey severity cited relevant diagnoses documented by surveyor surveyor Virginia-4 Provide necessary Resident was an 81-year- Resident sustained a Facility failed to provide March 2001 care and services: old admitted to the facility nondisplaced fracture of the left necessary care and services. D on 8/17/90 with psychoses wrist and suffered unnecessary The facility failed to assess and hypothyroidism. pain. and investigate the source of Recent assessment the resident’s pain. Nurses’ (1/22/01) indicated long- notes indicate no apparent and short-term memory injury after fall. On 9/15/00 at loss and moderate 6:30 p.m., resident complained dependency for activities of pain in left arm. There was of daily living. Care plan bruising on wrist and thumb, identified resident as at and the arm was swollen and risk for falls. A list of tender to touch. According to preventive measures was the surveyor, there was a provided. On 9/14/00 at delay in seeking more 7:30 p.m., resident fell and aggressive treatment or complained of pain all service, as evidenced by the over. fact that an x-ray was not obtained until 37 hours after the resident’s fall. Source: State nursing home survey reports. a To more easily distinguish among multiple surveys from the same state, we assigned consecutive numbers to each state’s surveys. b The resident description and relevant diagnoses are limited to the information provided by the surveyor. In some of the surveys, no background or diagnostic information was provided. c Skin tears and multiple bruises are serious and painful injuries for older individuals and should not be considered in the same context as cuts and bruises sustained by healthy and younger adults. A skin tear is a traumatic wound occurring principally on the extremities of older adults as a result of friction alone or shearing and friction forces that separate the top layer of skin from the underlying layer or both layers from the underlying structures. A skin tear is a painful but preventable injury. Individuals most at risk for skin tears are those with (1) fragile skin, (2) advanced age, (3) assistance devices (wheelchairs, lifts, walkers), (4) cognitive and sensory impairment, (5) history of skin tears, and (6) total dependence for care. In addition, treatment of bruises and skin tears for elderly residents of a nursing home is frequently complicated by diabetes, poor circulation, poor nutrition, and medications with blood thinning effects. See Sharon Baranoski, “Skin Tears: Staying on Guard Against the Enemy of Frail Skin,” Nursing 2000, vol. 30, no. 9, 2000. d Stages of pressure sore formation are I—skin of involved area is reddened, II—upper layer of skin is involved and blistered or abraded, III—skin has an open sore and involves all layers of skin down to underlying connective tissue, and IV—tissue surrounding the sore has died and may extend to muscle and bone and involve infection. e The following two resident incidents were cited at the B level for scope and severity, which means the surveyor found that both injuries were unavoidable and that the nursing home was in substantial compliance with the requirements. f These two citations involve two residents, one cognitively competent and the other with dementia, who were injured because side rails were in place on their beds. Numerous reports have cited the danger of side rails. Residents trying to get out of bed over the rails have injured themselves by falling. Other individuals have been caught between the bed rails and the mattress or have caught their heads in the rails. Some of these injuries resulted in death. Page 76 GAO-03-561 Nursing Home Quality Appendix III: Abstracts of Nursing Home Survey Reports That Understated Quality-of- Care Problems g A hydrocollator pack is a canvas bag containing a silicone gel paste that absorbs an amount of water 10 times its weight. The pack is placed in a heated water container, set at a temperature above 150° F. When ready, it is placed in a protective dry terrycloth wrap and applied on top of the area where the individual is experiencing pain. Lying or sitting on the pack negates the insulating effect of the terrycloth and the individual may be burned. Page 77 GAO-03-561 Nursing Home Quality Appendix IV: Information on State Nursing Appendix IV: Information on State NursingHome Surveyor Staffing Home Surveyor Staffing Table 9 summarizes state survey agencies’ responses to our July 2002 questions about nursing home surveyor experience, vacancies, hiring freezes, competitiveness of salaries, and minimum required experience. Table 9: State Survey Agency Responses to Questions about Surveyor Experience, Vacancies, Hiring Freezes, Competitiveness of Salaries, and Minimum Required Experience Surveyors with 2 years or less Surveyor Surveyor hiring RN surveyor Minimum required experience positions vacant freeze in effect as salaries are experience for RN a State (percent) (percent) of mid-2002 competitive surveyors (years) Maryland 70 9 Yes Yes 0 to 2 Oklahoma 67 4 Yes Yes 0 to1 New Hampshire 60 12 Yes No 2 Florida 55 8 No No 0 Idaho 54 0 Yes No 1 Washington 54 0 No No 2 California 52 6 Yes Yes 1 Georgia 51 14 No No 3 Kentucky 51 17 No Yes 4 District of Columbia 50 9 Yes Yes 3 Utah 50 8 No No 2 Louisiana 48 6 Yes No 2 to 3 Alabama 48 10 No No 0 Tennessee 45 18 No No 3 Maine 42 9 Yes No 5 Hawaii 40 17 No No 2-½ New York 40 4 Yes No 1 to 2 Missouri 36 11 No No 2 Oregon 34 12 Yes No 5 Arkansas 33 20 No No 2 North Carolina 33 18 No No 4 b b Texas 32 20 No No 1 New Mexico 30 34 No No 3 New Jersey 30 23 Yes No 3 Nebraska 29 6 No No 1 to 2 Connecticut 29 1 Yes Yes 4 Alaska 29 22 No No 2 Wisconsin 25 15 No No 0 Colorado 24 17 No No 1 Virginia 21 5 No No 0 Indiana 20 18 No No 1 Arizona 20 24 Yes No 2 South Dakota 18 0 No Yes 2 Ohio 17 5 No Yes 0 Page 78 GAO-03-561 Nursing Home Quality Appendix IV: Information on State Nursing Home Surveyor Staffing Surveyors with 2 years or less Surveyor Surveyor hiring RN surveyor Minimum required experience positions vacant freeze in effect as salaries are experience for RN a State (percent) (percent) of mid-2002 competitive surveyors (years) Michigan 17 5 Yes No 0 C Kansas 17 4 No No Massachusetts 16 14 Yes Yes 1 to 3 Pennsylvania 15 7 No Yes 1 Rhode Island 9 13 No Yes 1 Illinois 5 5 Yes Yes 2 to 3 Iowa 4 0 Yes No 5 Minnesota 0 17 Yes No 3 Source: State survey agency responses to July 2002 GAO questions. a Nine states did not respond to our inquiry—Delaware, Mississippi, Montana, Nevada, North Dakota, South Carolina, Vermont, West Virginia, and Wyoming. b Texas indicated that although there was no hiring freeze or layoffs, the survey staff was reduced by 107 positions through attrition from September 1, 2001, through June 1, 2002, in light of state funding changes and agency cuts. As of mid-2002, Texas was authorized 215 nurse surveyors and had 42 positions vacant. c Kansas requires independent experience in professional health care, but does not specify a time period for that experience. Page 79 GAO-03-561 Nursing Home Quality Appendix V: Predictability of Standard Appendix V: Predictability of Standard Nursing Home Surveys Nursing Home Surveys Our analysis found that 34 percent of current nursing home surveys were predictable, allowing nursing homes to conceal deficiencies if they choose to do so. In order to determine the predictability of nursing home surveys, we analyzed data from CMS’s OSCAR database (see table 10). We considered surveys to be predictable if (1) homes were surveyed within 15 days of the 1-year anniversary of their prior survey or (2) homes were surveyed within 1 month of the maximum 15-month interval between standard surveys. Consistent with CMS’s interpretation, we used 15.9 months as the maximum allowable interval between surveys. Because homes know the maximum allowable interval between surveys, those whose prior surveys were conducted 14 or 15 months earlier are aware that they are likely to be surveyed soon. Page 80 GAO-03-561 Nursing Home Quality Appendix V: Predictability of Standard Nursing Home Surveys Table 10: Predictability of Current Nursing Home Surveys, by State Homes surveyed within 1 Number of active Homes surveyed within 15 month of 15-month homes with a current Predictable surveys days of 1-year anniversary maximum interval of State and prior survey (percent) of prior survey (percent) prior survey (percent) Alabama 225 82.7 5.8 76.9 Oklahoma 354 71.5 0.6 70.9 South Carolina 174 67.8 6.9 60.9 Nebraska 226 59.7 3.1 56.6 Utah 91 52.7 1.1 51.6 Montana 103 52.4 8.7 43.7 Georgia 357 52.4 0.6 51.8 Hawaii 44 52.3 13.6 38.6 New York 663 52.0 14.8 37.3 Idaho 84 50.0 4.8 45.2 New Mexico 80 43.8 13.8 30.0 Delaware 42 42.9 31.0 11.9 California 1,324 41.2 9.5 31.7 Nevada 45 40.0 24.4 15.6 Arizona 138 39.9 21.0 18.8 New Jersey 359 39.0 18.7 20.3 Oregon 142 38.0 14.1 23.9 Maryland 246 37.0 20.7 16.3 Massachusetts 497 36.2 17.3 18.9 Arkansas 239 35.6 27.6 7.9 Virginia 275 35.3 30.5 4.7 Iowa 457 34.6 31.1 3.5 Nation 16,332 34.0 13.0 21.0 Kentucky 303 33.7 10.6 23.1 Ohio 973 33.6 3.0 30.6 North Dakota 85 32.9 28.2 4.7 Vermont 43 32.6 11.6 20.9 New Hampshire 83 32.5 12.0 20.5 South Dakota 111 32.4 18.9 13.5 Wisconsin 404 32.4 19.6 12.9 Washington 268 32.1 22.4 9.7 Florida 718 32.0 9.3 22.7 Mississippi 187 31.6 2.1 29.4 Rhode Island 96 31.3 12.5 18.8 Connecticut 253 30.8 15.8 15.0 Wyoming 39 30.8 10.3 20.5 Indiana 550 30.7 14.4 16.4 Tennessee 324 29.0 6.2 22.8 Louisiana 315 28.6 19.0 9.5 Texas 1,122 27.2 15.7 11.5 Page 81 GAO-03-561 Nursing Home Quality Appendix V: Predictability of Standard Nursing Home Surveys Homes surveyed within 1 Number of active Homes surveyed within 15 month of 15-month homes with a current Predictable surveys days of 1-year anniversary maximum interval of State and prior survey (percent) of prior survey (percent) prior survey (percent) Colorado 222 26.1 9.0 17.1 Pennsylvania 757 26.0 24.0 2.0 Kansas 369 25.2 13.6 11.7 Missouri 531 25.0 11.9 13.2 Maine 121 24.8 8.3 16.5 Minnesota 427 20.4 4.4 15.9 Alaska 15 20.0 6.7 13.3 District of Columbia 20 20.0 15.0 5.0 North Carolina 411 17.3 13.9 3.4 Illinois 849 15.2 9.7 5.5 West Virginia 138 10.9 8.7 2.2 Michigan 433 10.2 8.8 1.4 Source: GAO analysis of OSCAR data as of April 9, 2002. Page 82 GAO-03-561 Nursing Home Quality Appendix VI: Immediate Sanctions Appendix VI: Immediate Sanctions Implemented Under CMS’s Expanded Immediate Sanctions Policy Implemented Under CMS’s Expanded Immediate Sanctions Policy From January 2000 through March 2002, states referred 4,310 cases to CMS under its expanded immediate sanctions policy when nursing homes were found to have a pattern of harming residents.1 Because some homes had more than one sanction or may have had multiple referrals for sanctions, 4,860 sanctions were implemented (see table 11). Table 12 summarizes the amounts of federal civil money penalties (CMP) implemented against nursing homes referred for immediate sanction. Although these monetary sanctions were implemented, CMS’s enforcement database does not track collections. In addition, states may have imposed other sanctions under their own licensure authority, such as state monetary sanctions, in addition to or in lieu of federal sanctions. Such state sanctions are not recorded in CMS’s enforcement database. Table 11: Federal Sanctions Implemented against Nursing Homes Referred for Immediate Sanction, January 14, 2000, through March 28, 2002 a Type of sanction Number implemented CMP 2,933 Denial of payment for new admissions 1,232 Directed in-service training 345 State monitoring 192 Directed plan of correction 77 CMS approved alternative or additional state sanction 48 Termination from the Medicare and Medicaid programs 26 Temporary management 4 Denial of payment for all residents 2 Transfer of residents and closure of facility 1 Total 4,860 Source: CMS enforcement database as of March 28, 2002. a We excluded sanctions that were not implemented either because they were pending as of March 28, 2002, the date of our extract of CMS’s enforcement database, or because CMS withdrew them after imposition. 1 We use the term “cases” because some homes had multiple referrals for immediate sanctions. Page 83 GAO-03-561 Nursing Home Quality Appendix VI: Immediate Sanctions Implemented Under CMS’s Expanded Immediate Sanctions Policy Table 12: Federal CMPs Implemented under CMS’s Immediate Sanctions Policy, January 2000 through March 2002 State CMP amount Alabama $375,627.50 Alaska 0.00 Arizona 350,652.50 Arkansas 1,571,654.04 California 1,681,813.50 Colorado 1,489,100.00 Connecticut 696,350.00 Delaware 214,342.50 District of Columbia 20,000.00 Florida 1,975,375.00 Georgia 487,050.00 Hawaii 20,000.00 Idaho 37,350.00 Illinois 2,801,656.50 Indiana 1,977,685.50 Iowa 175,945.00 Kansas 415,400.00 Kentucky 1,195,177.50 Louisiana 20,000.00 Maine 184,920.00 Maryland 290,270.00 Massachusetts 1,031,445.00 Michigan 1,035,815.00 Minnesota 66,307.50 Mississippi 186,977.50 Missouri 467,157.50 Montana 0.00 Nebraska 11,207.50 Nevada 429,500.00 New Hampshire 93,350.00 New Jersey 1,543,007.50 New Mexico 222,430.00 New York 0.00 North Carolina 2,171,013.75 North Dakota 15,730.00 Ohio 3,104,870.00 Oklahoma 1,075,036.50 Oregon 15,225.00 Pennsylvania 1,250,417.00 Rhode Island 9,425.00 South Carolina 29,250.00 Page 84 GAO-03-561 Nursing Home Quality Appendix VI: Immediate Sanctions Implemented Under CMS’s Expanded Immediate Sanctions Policy State CMP amount South Dakota 0.00 Tennessee 381,432.50 Texas 7,677,219.58 Utah 37,157.00 Vermont 11,550.00 Virginia 934,425.00 Washington 0.00 West Virginia 112,160.00 Wisconsin 901,960.50 Wyoming 0.00 Total $38,794,439.37 Source: CMS enforcement database. Page 85 GAO-03-561 Nursing Home Quality Appendix VII: Cases States Did Not Refer to Appendix VII: Cases States Did Not Refer to CMS for Immediate Sanction CMS for Immediate Sanction State survey agencies did not refer to CMS for immediate sanction a substantial number of nursing homes found to have a pattern of harming residents. Most states failed to refer at least some cases and a few states did not refer a significant number of cases.1 While seven states appropriately referred all cases, the number of cases that should have been but were not referred ranged from 1 to 169. Four states accounted for about 55 percent of cases that should have been referred. Table 13 shows the number of cases that states should have but did not refer for immediate sanction (711) as well as the number of cases that states appropriately referred (4,310) from January 2000 through March 2002. Table 13: Number of Cases States Did Not Refer for Sanction, as Required, and the Number States Appropriately Referred, January 2000 through March 2002 Number of cases not Number of cases State referred as required referreda Nation 711 4,310 Texas 169 423 New York 140 22 Massachusetts 46 81 Pennsylvania 38 164 Connecticut 26 244 Washington 26 227 Illinois 24 241 Florida 21 150 New Jersey 20 56 Tennessee 20 46 Minnesota 19 68 Missouri 18 108 South Carolina 18 3 North Carolina 10 242 Arizona 9 24 Maryland 9 34 Wyoming 9 11 California 7 96 Michigan 7 284 Arkansas 6 115 Montana 6 14 Ohio 6 323 Idaho 5 31 1 We use the term “cases” because some homes had multiple referrals for immediate sanctions. Page 86 GAO-03-561 Nursing Home Quality Appendix VII: Cases States Did Not Refer to CMS for Immediate Sanction Number of cases not Number of cases a State referred as required referred Indiana 5 270 Louisiana 5 82 Oklahoma 4 53 West Virginia 4 11 Delaware 3 14 Georgia 3 81 Hawaii 3 1 Iowa 3 44 New Hampshire 3 20 Colorado 2 116 District of Columbia 2 1 Oregon 2 51 Rhode Island 2 3 South Dakota 2 18 Virginia 2 41 Wisconsin 2 61 Alabama 1 50 Kansas 1 175 Maine 1 18 New Mexico 1 19 Nevada 1 12 Alaska 0 0 Kentucky 0 75 Mississippi 0 23 Nebraska 0 30 North Dakota 0 20 Utah 0 11 Vermont 0 3 Source: CMS regional office review of cases identified through GAO’s analysis of OSCAR data and the CMS Enforcement Database. a Reflects cases entered in CMS’s enforcement database by March 28, 2002. Page 87 GAO-03-561 Nursing Home Quality Appendix VIII: HCFA State Performance Appendix VIII: HCFA State Performance Standards for Fiscal Year 2001 Standards for Fiscal Year 2001 Table 14 summarizes HCFA’s state performance standards for fiscal year 2001, describes the source of the information CMS used to assess compliance with each standard, and identifies the criteria the agency used to determine whether states met or did not meet each standard. Table 14: Overview of HCFA’s Seven State Performance Standards for Nursing Home Survey Activities for Fiscal Year 2001 Criteria for determining compliance Description Source of information with standard 1. Surveys are planned, scheduled, and conducted in a timely manner At least 10 percent of standard surveys OSCAR and state survey schedules At least 10 percent of standard surveys begin on weekends or “off-hours” begin on weekends or off-hours Standard surveys are conducted within OSCAR 100 percent of nursing homes are prescribed time limits surveyed within statutory time limits 2. Survey findings (deficiencies) are supportable State surveyors explain and properly A random sample of 10 percent At least 85 percent of the deficiencies document all deficiencies in survey reports (maximum of 40, minimum of 5) of the reviewed meet the principles of following HCFA guidance known as the state’s survey results in which certain documentation “principles of documentation” deficiencies were cited at “D” or higher levels of scope and severity 3. Surveys are fully documented and consistent with applicable laws, regulations, and general instructions Surveys are adequately conducted by state Reports generated from HCFA’s database 100 percent of standard surveys are agencies using the standards, protocols, on federal monitoring surveys adequately conducted by state agencies forms, methods, procedures, policies, and using the standards, protocols, forms, systems specified by HCFA instructions methods, procedures, policies, and systems specified by HCFA instructions 4. When states certify that nursing homes are not in compliance, they follow adverse action procedures set forth in regulations and general instructions “Immediate and Serious Threat” cases are OSCAR, Enforcement Tracking System In 95 percent of cases in which there is processed in a timely manner reports, and state agency provider immediate jeopardy or a serious threat to certification files resident health and safety, the state agency adheres to the 23-day termination process Payments are not made to nursing homes OSCAR, Enforcement Tracking System The state provides timely notice to HCFA that have not achieved substantial reports, and state agency provider (i.e., 20 days prior to the home’s compliance within 6 months of their last certification files termination date) on 100 percent of the surveys cases in which the nursing home has not achieved timely compliance 5. All expenditures and charges to the program are substantiated to the Secretary’s satisfaction The state agency employs an acceptable HCFA budget expenditure and workload More than 20 different items on the two process for charging federal programs reports reports submitted by the states are reviewed for accuracy, completeness, and timeliness and are scored as either on time or late, or met or not met for a reporting period The state agency has an acceptable OSCAR reports Numerous items submitted by the states, method for monitoring its current rate of such as quarterly expenditure reports and expenditures supplemental budget requests, are reviewed to determine if state requirements for monitoring expenditures are met, not met, or not applicable Page 88 GAO-03-561 Nursing Home Quality Appendix VIII: HCFA State Performance Standards for Fiscal Year 2001 Criteria for determining compliance Description Source of information with standard 6. Conduct and reporting of complaint investigations are timely and accurate, and comply with general instructions for handling complaints Investigate immediate jeopardy complaints Semiannual review of a 10 percent 100 percent of immediate jeopardy within 2 workdays sample of a state’s complaint files complaints are investigated within 2 days Investigate actual harm complaints within (maximum of 20 cases) 100 percent of actual harm complaints are 10 workdays investigated within 10 days Maintain and follow guidelines for the The state agency has and follows its own prioritization of all other complaints written criteria governing the prioritization of complaints that do not allege immediate jeopardy or actual harm State enters complaint data into OSCAR Semiannual on-site reviews of 20 state 100 percent of deficiencies cited in the appropriately and in a timely manner complaint survey reports sampled complaints are cited under the correct federal citation OSCAR data are reviewed quarterly for Average time to enter results of complaint timely entry investigations does not exceed 20 calendar days from completion of the case 7. Accurate data on survey results are entered into OSCAR in a timely manner Results of standard surveys are entered Semiannual review of all standard surveys The statewide average time between state into OSCAR in a timely manner based on OSCAR data agency sign-off of the certification and transmittal form and entry of the survey results into OSCAR does not exceed 20 calendar days Results of surveys are entered into OSCAR Semiannual review of a random sample of No less than 85 percent of cases reviewed accurately nursing home survey results demonstrate that data were entered into OSCAR accurately Source: HCFA’s State Performance Review Protocol Guidance for fiscal year 2001. Note: HCFA did not finalize and issue the fiscal 2001 performance standards and guidance until April 2001. Page 89 GAO-03-561 Nursing Home Quality Appendix IX: Highlights of State Compliance Appendix IX: Highlights of State Compliance with CMS Performance Standards with CMS Performance Standards Table 15 summarizes the results of CMS’s fiscal year 2001 state performance review for each of the five standards we analyzed. We focused on five of CMS’s seven performance standards: statutory survey intervals, the supportability of survey findings, enforcement requirements, the adequacy of complaint activities, and OSCAR data entry. Because several standards included multiple requirements, the table shows the results of each of these specific requirements separately. Table 15: State Compliance with Selected CMS Performance Standards, Fiscal Year 2001 CMS standard and requirements Number of states not meeting standard Survey timeliness The state begins no less than 10 percent of its standard surveys during 2 weekends or “off-hours.” (Standard 1, criterion 1) The state conducts standard surveys in prescribed times. (Standard 1, criterion 2) • The average statewide interval between consecutive standard surveys 9 is not greater than 12 months. • Each home is surveyed within 15 months of its prior survey. 17 Supportability of survey findings The state explains and properly documents deficiencies. (Standard 2) Due to complications with the review protocol, this standard was not reported. Enforcement The state properly follows termination procedures. (Standard 4, criterion 1) 3 The state notifies CMS when a nursing home has not achieved substantial 4 compliance in a timely manner. (Standard 4, criterion 2) Complaints The state investigates all complaints alleging immediate jeopardy to a 12 resident within 2 workdays. (Standard 6, criterion 1) The state investigates all complaints alleging actual harm to a resident 42 within 10 workdays. (Standard 6, criterion 2) The state has and follows guidelines for prioritizing complaints not alleging 15 immediate jeopardy or actual harm. (Standard 6, criterion 3) The state enters citations resulting from complaint investigations into 13 CMS’s complaint database. (Standard 6, criterion 4) OSCAR The state enters survey results into CMS’s database in a timely manner. 9 (Standard 7, criterion 1) The state enters survey results into CMS’s database accurately. (Standard 24 7, criterion 2) Source: GAO analysis of results of CMS Fiscal Year 2001 State Performance Standard Reviews. Note: We reviewed five of the seven CMS performance standards. See app. VIII, table 14, for a description of standards three and five, which we did not review. Page 90 GAO-03-561 Nursing Home Quality Appendix X: Comments from the Centers for Medicare & Medcaid Services Appendix X: Comments from the Centers for Medicare & Medcaid Services Page 91 GAO-03-561 Nursing Home Quality Appendix X: Comments from the Centers for Medicare & Medcaid Services Page 92 GAO-03-561 Nursing Home Quality Appendix X: Comments from the Centers for Medicare & Medcaid Services Page 93 GAO-03-561 Nursing Home Quality Appendix X: Comments from the Centers for Medicare & Medcaid Services Page 94 GAO-03-561 Nursing Home Quality Appendix XI: GAO Contact and Staff Appendix XI: GAO Contact and Staff Acknowledgements Acknowledgements Walter Ochinko, (202) 512-7157 GAO Contact The following staff made important contributions to this work: Jack Acknowledgements Brennan, Patricia A. Jones, Dan Lee, Dean Mohs, and Peter Schmidt. Page 95 GAO-03-561 Nursing Home Quality Related GAO Products Related GAO Products Nursing Homes: Public Reporting of Quality Indicators Has Merit, but National Implementation Is Premature. GAO-03-187. Washington, D.C.: October 31, 2002. Nursing Homes: Quality of Care More Related to Staffing than Spending. GAO-02-431R. Washington, D.C.: June 13, 2002. Nursing Homes: More Can Be Done to Protect Residents from Abuse. GAO-02-312. Washington, D.C.: March 1, 2002. Nursing Homes: Federal Efforts to Monitor Resident Assessment Data Should Complement State Activities. GAO-02-279. Washington, D.C.: February 15, 2002. Nursing Homes: Success of Quality Initiatives Requires Sustained Federal and State Commitment. GAO/T-HEHS-00-209. Washington, D.C.: September 28, 2000. Nursing Homes: Sustained Efforts Are Essential to Realize Potential of the Quality Initiatives. GAO/HEHS-00-197. Washington, D.C.: September 28, 2000. Nursing Home Care: Enhanced HCFA Oversight of State Programs Would Better Ensure Quality. GAO/HEHS-00-6. Washington, D.C.: November 4, 1999. Nursing Homes: HCFA Should Strengthen Its Oversight of State Agencies to Better Ensure Quality of Care. GAO/T-HEHS-00-27. Washington, D.C.: November 4, 1999. Nursing Home Oversight: Industry Examples Do Not Demonstrate That Regulatory Actions Were Unreasonable. GAO/HEHS-99-154R. Washington, D.C.: August 13, 1999. Nursing Homes: HCFA Initiatives to Improve Care Are Under Way but Will Require Continued Commitment. GAO/T-HEHS-99-155. Washington, D.C.: June 30, 1999. Nursing Homes: Proposal to Enhance Oversight of Poorly Performing Homes Has Merit. GAO/HEHS-99-157. Washington, D.C.: June 30, 1999. Nursing Homes: Complaint Investigation Processes in Maryland. GAO/T-HEHS-99-146. Washington, D.C.: June 15, 1999. Page 96 GAO-03-561 Nursing Home Quality Related GAO Products Nursing Homes: Complaint Investigation Processes Often Inadequate to Protect Residents. GAO/HEHS-99-80. Washington, D.C.: March 22, 1999. Nursing Homes: Stronger Complaint and Enforcement Practices Needed to Better Ensure Adequate Care. GAO/T-HEHS-99-89. Washington, D.C.: March 22, 1999. Nursing Homes: Additional Steps Needed to Strengthen Enforcement of Federal Quality Standards. GAO/HEHS-99-46. Washington, D.C.: March 18, 1999. California Nursing Homes: Federal and State Oversight Inadequate to Protect Residents in Homes with Serious Care Problems. GAO/T-HEHS- 98-219. Washington, D.C.: July 28, 1998. California Nursing Homes: Care Problems Persist Despite Federal and State Oversight. GAO/HEHS-98-202. Washington, D.C.: July 27, 1998. (290158) Page 97 GAO-03-561 Nursing Home Quality The General Accounting Office, the audit, evaluation and investigative arm of GAO’s Mission Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO’s commitment to good government is reflected in its core values of accountability, integrity, and reliability. The fastest and easiest way to obtain copies of GAO documents at no cost is Obtaining Copies of through the Internet. GAO’s Web site (www.gao.gov) contains abstracts and full- GAO Reports and text files of current reports and testimony and an expanding archive of older products. The Web site features a search engine to help you locate documents Testimony using key words and phrases. You can print these documents in their entirety, including charts and other graphics. Each day, GAO issues a list of newly released reports, testimony, and correspondence. GAO posts this list, known as “Today’s Reports,” on its Web site daily. The list contains links to the full-text document files. To have GAO e-mail this list to you every afternoon, go to www.gao.gov and select “Subscribe to daily E-mail alert for newly released products” under the GAO Reports heading. Order by Mail or Phone The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders should be sent to: U.S. General Accounting Office 441 G Street NW, Room LM Washington, D.C. 20548 To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202) 512-6061 Contact: To Report Fraud, Web site: www.gao.gov/fraudnet/fraudnet.htm Waste, and Abuse in E-mail: email@example.com Federal Programs Automated answering system: (800) 424-5454 or (202) 512-7470 Jeff Nelligan, Managing Director, NelliganJ@gao.gov (202) 512-4800 Public Affairs U.S. General Accounting Office, 441 G Street NW, Room 7149 Washington, D.C. 20548
Nursing Home Quality: Prevalence of Serious Problems, While Declining, Reinforces Importance of Enhanced Oversight
Published by the Government Accountability Office on 2003-07-15.
Below is a raw (and likely hideous) rendition of the original report. (PDF)