National Academies Press: OpenBook

Improving the Quality of Care in Nursing Homes (1986)

Chapter: 4. Monitoring Nursing Home Performance

« Previous: 3. Regulatory Criteria
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 104
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 105
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 106
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 107
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 108
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 109
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 110
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 111
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 112
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 113
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 114
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 115
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 116
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 117
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 118
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 119
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 120
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 121
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 122
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 123
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 124
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 125
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 126
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 127
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 128
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 129
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 130
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 131
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 132
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 133
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 134
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 135
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 136
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 137
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 138
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 139
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 140
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 141
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 142
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 143
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 144
Suggested Citation:"4. Monitoring Nursing Home Performance." Institute of Medicine. 1986. Improving the Quality of Care in Nursing Homes. Washington, DC: The National Academies Press. doi: 10.17226/646.
×
Page 145

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Monitoring Nursing Home Performance THE ISSUES The federal and state governments share responsibility for quality assurance in nursing homes. The performance criteria are federal, but the federal government has delegated to the states responsibility to inspect nursing homes using these criteria and to certify their eligi- bility to participate in the Medicaid program. For the Medicare program, state governments inspect the facilities on behalf of the federal government and make certification recommendations to the federal government; the certifi- cation decisions are made by the HCFA. The federal government has authority in both the Medicaid and Medicare programs to conduct independent inspections of certified nursing homes to audit the states' certification activi- ties. The federal government also can decertify subs- tandard facilities. The federal conditions and standards were designed for use by state surveyors in inspecting nursing homes. The survey process is supposed to identify and measure performance deficiencies that result in poor-quality care and should produce documentation of the deficiencies that will support the government's case in contested enforcement actions. 104

MONITORING NURSING HOME PERFORMANCE / 105 State monitoring of the performance of nursing homes now includes three types of activities: 1. Nursing home inspections (surveys) are conducted at least once a year by staff of the state health facilities licensure and certification agency to determine the extent of compliance of facilities with federal conditions and standards. 2. "Inspection of care" (IOC) is conducted either by the state Medicaid agency, the state health facilities licensure and certification agency, or a professional review organization. By law, inspection of the care provided to every Medicaid recipient must be done annually. It is conducted for two purposes: (1) utili- zation review, to be certain that the resident is eligible for nursing home care and is placed in the right level of care; and (2) quality of care, to be sure each resident is receiving appropriate care of adequate quality. 3. Ad hoc complaints submitted by residents, their families, or ombudsmen or other third parties are also investigated. Complaints frequently concern possible violations of federal conditions and standards or other regulatory requirements. Monitoring the performance of thousands of nursing homes for quality assurance purposes has been difficult to carry out effectively and reliably. The first set of problems stems from the inadequacies of the criteria and of the survey process used to determine the quality of care being provided. The problems with the current criteria are discussed in Chapter 3 and major changes are recommended to make them more resident-centered and outcome-oriented. The first set of issues discussed in this chapter covers the inadequacies of the current survey process. Changes are recommended that follow from the new conditions and standards recommended in Chapter 3. A second set of issues concerns federal-state and intrastate role relationships. Four specific issues are discussed: the relationship of inspection of care to the survey process, the relationships of the survey process to

106 / NURSING HOME CARE the long-term-care ombudsman program, the elimination of the differences in federal and state certification responsibilities for Medicare and Medicaid facilities, and shifts in the responsibility for surveying state-owned facilities. The third set of issues deals with both state and federal capacity for effectively carrying out their quality assurance responsibilities. These issues include funding of federal and state survey units and the numbers, qualifications, and training of surveyors. PROBLEMS WITH THE SURVEY PROCESS The survey process has several problems that should be addressed to make it more effective: predictability, inefficiency, emphasis on paper compliance, insensitivity to resident needs, inconsistency, isolation from related monitoring processes, and variable state regulatory capacity. Predictability If the operators of a substandard facility know when it will be surveyed, they not only can clean it up and bring the records up to date, but they also may stock up, improve the menus, bring in additional personnel, and take other actions to bring the facility into temporary compliance. The committee heard anecdotal accounts in the public meetings and in case-study interviews of facilities ~. _ being notified about impending survey visits. Prior notice, either formal or informal, was the policy in some states because it made the visit easier by ensuring the presence of key personnel in the facility. Prior notice was prohibited by the HCFA several years ago, but a few states apparently still follow this policy. Even without direct notice, however, providers often can predict the timing of an annual survey visit within several weeks because certification lasts exactly 12 months and an annual survey is required by the regulations

MONITORING NURSING HOME PERFORMANCE / 107 at least 90 days before certification expires. The case studies found that some states routinely schedule visits during the same week each year. Others send in a team of auditors or the state fire marshal a specified number of weeks before the survey visit. Inefficiency All nursing homes are subjected to the same survey intensity regardless of their past record of compliance. Most state survey agencies have very limited budgets. They barely have enough staff to complete the round of annual required surveys and do not always have enough surveyors to follow up adequately on the major problem facilities. A more efficient survey process would permit them to spend more time in poor facilities and less time in good facilities. Paper Compliance Not only are the current standards focused on theoretical facility capability rather than actual per- formance, but compliance is often determined on the basis of record reviews rather than direct observation. Insensitivity to Resident Needs Nursing home residents have widely varying needs and some facilities specialize, either formally or informally, by accepting residents only of a particular type. The severely demented and those requiring active rehabili- tation are two groups of residents often cared for in separate facilities or on separate floors. The existing survey process makes no allowance for the observed diversity among patients and across facilities. At present, all SNFs are surveyed in the same manner; the same is true for all ICFs with the exception of

108 / NURSING HOME CARE intermediate care facilities for the mentally retarded, which are subject to a different set of standards. Inconsistency The case studies and survey of state licensure and certification agencies conducted by the committee revealed great variations among the states in the way they carry out the survey process. HCFA data show wide variations in the numbers and types of deficiencies typically cited from state to state. For example, the proportion of a state's SNFs having more than 25 deficiencies in 1983 ranged from O in Delaware to 100 percent in Washington, D.C. (mean = 24 percent).3 Another study found that the most common deficiencies in SNFs were very different from state to state.4 Part of the variation in findings may reflect real differences in facility characteristics from state to state, but much of the variation is probably due to differences in state agency interpretation of conditions and standards and in survey processes. In addition to state-level variations, numerous anecdotes of inconsistencies from one surveyor to the next were cited in the public meetings and case-study interviews. These inconsistencies in surveyor judgments are evidently random and appear as "noise" in national survey statistics, but they are extremely annoying to providers and confound state agency efforts to manage the survey process effectively. Isolation from Related Monitoring Processes In some states, there is little or no sharing of information or coordinated effort between the survey process and the processes for monitoring and investigating complaints, even though complaints can be an important source of information about quality problems in nursing homes. Relationships between the state survey agencies and ombudsman programs are often undeveloped or even adversarial. In addition, only 17 states combine or

MONITORING NURSING HOblE PERFORMANCE / 109 coordinate the inspection-of-care reviews with the survey process even though both involve annual inspections of resident care. Variable State Regulatory Capacity The survey results and case studies of the state survey and certification agencies revealed large differences in the level of funding and staffing and in the types and deployment of personnel relative to the number of facilities. These differences result in part from the absence of a federal formula for distributing survey and certification funds and the absence of guidelines for organizing and staffing the state agencies, but they also reflect differences in state budgeting contributions and inspection policies and practices. There also are differences in state regulatory standarcls, due-process rules, court interpretations, and availability of inter- mediate sanctions. REDESIGNING THE SURVEY PROCESS These problems can be dealt with effectively by redesigning the survey process to implement the resident-centered, outcome-oriented conditions and standards recommended in Chapter 3. The new conditions and standards will require surveyors to scrutinize the care being provided and its effects on residents, rather than emphasize reviews of records, forms, and written policies as is now the case. In conjunction with new survey protocols and scoring procedures based on empirical resident-outcome standards developed from standardized resident assessment data, the new conditions and standards should improve consistency of decision-making on deficiencies, although surveyor judgment will still play an important role. Development and use of a shorter inspection procedure and use of an outcome-orienteci survey protocol will permit surveyors to identify and concentrate their efforts on facilities with problems. Also, the

1 10 / NURSING HOME CARE inspection-of-care reviews should be incorporated into the survey process, permitting more efficient use of regulatory capacity. Other problems, such as predictability, could be solved by making minor changes in standard operating procedures. The revised survey process should be resident-centered and outcome-oriented where appropriate, although it should not eliminate all concern for certain facility character- istics that relate to life safety, cleanliness, sanitary food service, basic capacity to provide proper care, or the process standards for therapeutic diets or drug administration. It should take into account the different mixes of resident characteristics and service needs (case mix) found in different facilities, spend less survey time in the better facilities and more in the poorer facilities, and decrease the predictability of survey timing. The new process outlined in this chapter would be more efficient because it would use a shorter standard survey that would permit survey agencies to spend less time on good facilities and more time on substandard providers. It would also relieve good providers from being subjected to unnecessarily intensive inspections. More important, the new survey process would be more effective because it would rely on more appropriate indicators of compliance with federal quality-of-care and quality-of-life conditions and standards than the structurally focused survey in use today. The main features of the new survey process are discussed in detail in the remainder of this chapter. The following points are covered: · consolidation of Medicaid and Medicare survey procedures, ~ two-stage survey approach, · ~ ~ case-mix ret erenclng, key indicators of quality, scoring and decision-making, survey data sources, coordination with complaint programs, consumer involvement,

MONITORING NURSING HOME PERFORMANCE / · positive incentives, and · continuous improvement of the survey process. Consolidation of Medicare and Medicaid Survey Procedures The procedures for certifying Medicaid and Medicare facilities are virtually identical. They should be consolidated. Recommendation 4-1: Medicare and Medicaid survey and certification process requirements should be consolidated in one place in the Code of Federal Regulations to promote consistency. Timing and Frequency of Surveys Although some states have experimented with flexible survey cycles, there is still no valid information on the optimum periodicity of inspections for detecting violations before they become serious. Even excellent facilities may fall out of compliance very quickly after key staff, ownership, or resident mix changes. The consensus among consumer, regulator, and provider groups is that annual surveys of nursing homes are both reasonable and necessary. The frequency and timing of standard surveys should be determined by each facility's performance history and should maximize the element of surprise. The objective is to encourage continuing compliance with the federal regulations. To ensure scheduling uncertainty, the actual interval between surveys for a particular facility might range from 9 to 15 months, depending upon past performance and its latest survey findings. Some facilities may need to be surveyed even more frequently if their performance has been exceptionally poor. This increased flexibility in the timing of surveys should not, however, lead to an effective lengthening of the average time between routine surveys across all

1 12 / NURSING HOME CARE facilities in each state. That should remain constant at 12 months. As a general principle, surveys should be unannounced and unanticipated by facilities, with the exception of followup visits to determine whether satisfactory corrections have been made. Whatever their record, all facilities should be at risk for a random, full-scale extended survey at any time. Facilities also should be surveyed within a specified period of time after key events occur that are likely to affect the quality of care and quality of life in a facility, for example, change in ownership, administrator, or director of nursing. (Surveys after changes in ownership are already required by current regulations.) A high rate of nursing staff turnover or extensive use of nursing pools also might trigger an inspection. Similarly, multiple validated complaints about a facility should warrant an immediate survey. The introduction into the survey cycle of flexibility that is tied to performance and key events should enable survey resources to be targeted to those facilities most in need of attention: problem or marginal facilities and facilities where new circumstances could adversely affect residents. Facilities that are performing well would be rewarded for their good behavior by less-intense monitoring. That will allow survey agency staff to be used for more urgent tasks. The time-limited agreement requirement that was dropped in 1981 legislation, but is still required by regulation, should be eliminated to allow the annual survey to take place as late as 15 months after the previous annual survey. In practice, the time-limited agreement provisions have not made it easier to terminate facilities, because the courts have imposed the same due-process prior hearing requirements for ., . . ~ terminating facilities with expired agreements as apply to facilities with agreements in force. For this reason, a group of providers, consumers, and regulators convened by the HCFA in 1983 to develop a consensus on regulatory changes recommended elimination of mandatory time-limited agreements.

MONITORING NURSING [IOME PERFORMANCE / 113 Recommendation 4-2: The timing of surveys should maximize the element o) surprise; the standard! annual survey should be conducted somewhere between 9 and 15 months after the previous annual survey, with the average , . , , ~ across all facilities within each state remaining at 12 months. Additional standard surveys also should take place whenever there are key events, such as a change in ownership. Independent of the survey cycle, all facilities should be required to pass rigorous life safety code and food inspections at regular intervals. Two-Stage Survey Approach After an initial audit of a sample of resident assessment records, each annual survey would begin with a short standard survey protocol. The standard survey would be designed to use "key indicators" of performance to identify facilities with poor resident outcomes that might have resulted from substandard nursing home performance. If a facility had problems on the key indicators (discussed below), it would be subjected to an extended survey protocol entailing observation and interview of additional residents to determine the extent to which staffing and other structural features of the facility, and the way care is being provided, may have caused the poor resident outcomes. The main purposes of the two-stage process are to relieve good facilities from the burden of a lengthy regular survey and to permit survey agencies to concentrate their efforts on poor and marginal facilities. The Resident Assessment Audit Surveyors would audit, by using the same resident assessment protocol the facilities are required to use, a sample of all residents to test the accuracy of the facility's assessment reports. A determination would then be made of whether the facility's resident assessment reports meet acceptable standards of accuracy. If

1 14 / NURSING HOME CARE surveyors find that a facility's assessments of resident status differ from their own by more than a predetermined rate, they would conclude that the facility's misclassi- fication of residents reflects either professional incompetence or deliberate inaccuracy. In either case, a complete extended survey of the facility would be called for. For facilities that pass the audit, a standard survey would be conducted. The Standard Survey The standard survey would use a statistically valid, case-mix-stratified sample of the residents in a nursing home. (The case-mix definitions, sampling issues, and the key indicators are discussed more fully below.) It also would measure overall facility performance through such environmental indicators as the personal grooming of residents, cleanliness, and so on. To the extent possible, the standard survey would use a short protocol that would rely on "key indicators" of performance. Among the key indicators that may be used, depending on the availability of empirical evidence, are those elements in a standard that have been shown to be highly predictive of compliance with the other elements in that standard. Key indicators also may be specific negative (although sometimes unavoidable) or positive outcomes appropriate to · ~ case-mix groupings. Use of the standard survey should enable surveyors to sort facilities into one of three categories: those that are superior or clearly adequate, those that are clearly inadequate or deficient in one or more areas of performance, and those whose performance Is ambiguous. Facilities in the superior/adequate group would normally be exempt from further review at that time, except for life safety code and sanitation inspections that will be required for all facilities and scheduled independently of the survey cycle. All other facilities will be required to undergo a partial or complete extended survey. ~.

MONITORING NURSING HOME PERFORMANCE / 1 15 The Extencled Survey The extended survey would enlarge the sample of residents examined and increase the comprehensiveness of the standard protocol to look at compliance with all elements and all standards. It would further document poor resident outcomes and explore the extent to which structural and process features of the facility may have contributed to these outcomes. Partial extended surveys would be conducted in facilities where performance is questionable or clearly below par in particular areas, but where performance in other areas is not suspected of being substandard. Complete extended surveys would be conducted in facilities where serious or persistent questions about overall performance have been raised. For "ambiguous" facilities, a partial or complete extended survey would be conducted, depending on the nature of the ambiguity found during the standard survey. Recon~n~endation 4-3: Two new survey protocols should be designed and tested to implement the new conditions and stand ard s recommend ed in Cha pier 3. a stand ard survey and an extencled survey. Both must be based on the revised conditions of participation and standards. Case-Mix Referencing The survey protocols should take into account the differing characteristics of residents in a facility, because some key indicators of quality are more relevant to residents with certain characteristics than to others. Accordingly, it is necessary to classify residents into defined case-mix groupings, using specified criteria. A simple stratification approach using key variables important in dealing with nursing home residents is proposed in this section for initial use in the revised survey process. Eventually, however, case-mix categories should be defined on the basis of resident groupings that emerge from the resident assessment scores and on empirical evidence from the resident assessment data that

116/NURSINGHOME CARE residents in a particular case-mix grouping should have a statistically predictable distribution of outcomes (changes in status) over defined periods of time after admission. The numbers of residents in different case-mix groupings could then become the denominators in determining the prevalence of poor outcomes among relevant groups of residents (for example, the percentage of decubitus ulcers among bed-bound and chair-bound residents, the percentage of mentally confused residents under physical or chemical restraint, the percentage of residents with urinary incontinence who have indwelling catheters, and so forth). The purpose of both the standard and the extended surveys is to evaluate the appropriateness of the care and the quality of life provided to the various types of residents found in a nursing home. Ideally, one would like to include the alert and oriented residents, short- and long-stayers, the physically and mentally dependent, younger and older residents, the well- and lesser- educated, and public- and private-pay residents, to name a few categories. However, it is not feasible to stratify nursing home residents by a large number of variables for survey purposes. Initially, the case-mix categories would be based on measurements of physical and cognitive functioning and, if indicated, levels of mental depression. Most nursing home residents have either mental or physical impairments, or both. A few have neither. Residents in each of these categories have very different medical and social needs. The committee believes that the survey instruments and survey process can and should be organized to take into account these differences in resident need in different facility populations. It is time-consuming and therefore costly to have surveyors assess all residents in a nursing home. Moreover, it is not necessary to do so, because it is possible to assess a carefully selected sample of residents and obtain generalizable results.5 The survey instrument can be designed so that different sets of standards and elements may be used for different compositions of facility populations. The survey process can be referenced to resident case mix so that, for

MONITORING NURSING HOME PERFORMANCE / 117 example, different sets of criteria will be applied to evaluate quality of care and life for a bed-bound population than for a mostly physically independent population. It is important to measure the incidence of bed sores only among individuals at risk of skin breakdown because they are bed-bound or chair-fast rather than to measure the incipience of bed sores among all residents in a facility if a significant number are not at risk of acquiring bed sores. Otherwise, a facility with almost all ambulatory care residents and a 1 percent incidence of decubitus ulcers would be rated as superior to a facility with a totally bed-bound population and a 3 percent incidence of decubiti. Yet the comparison would not be valid. Accordingly, the case-mix referencing system should center around two parameters of resident condition that are central to the special care needs and vulnerabilities of nursing home residents: mental status and physical dependency. For survey purposes, there would be four major case-mix groups, each of which should be defined by scores on the resident assessment instrument. In addition, of the two mentally competent groups, subsets of interest are those residents who are also clinically depressed. The process for sorting and stratifying residents into these categories would be as follows: 1. The facility would give the survey team a list of the residents and the case-mix categories to which they are assigned. 2. The survey protocol would then draw samples according to a prescribed sampling algorithm from each of the case-mix groupings. The sampling algorithm, which would specify the sampling methodology (for example, randomized within case-mix grouping) and sample size (perhaps to take account of overall facility population size), would have to be specified by the HCFA. The resident assessment protocols applied to the sample residents by surveyors would require obtaining somewhat different information about sampled residents depending on their functional and mental status.

1 18 / NURSING HOME CARE · Mentally competent residents would be interviewed to determine their level of satisfaction with the care they are receiving and to provide concrete information about matters such as the availability of activities, the flexi- bility of meal and bedtime hours, and so on. · Mentally confused residents would be observed to determine whether there is excessive use of physical or chemical restraints. Audited institutional data should be used to define the numbers of residents at risk so that negative outcomes could be evaluated by the ratio of such occurrences to the numbers of at-risk patients. · For physically independent residents, the standard survey would seek information on the availability of appropriate activities. · For the physically dependent residents, the survey would place particular emphasis on specific, measurable results of poor care, such as decubitus ulcers, urinary tract infections, contractures, malnutrition, and dehydration. Audited institutional data should be used to define the numbers of residents at risk so that negative outcomes could be evaluated using the ratio of such occurrences to the numbers of at-risk residents. Recommendation 4-4: Both standard! anc/t extender! surveys should assess samples of residents stratified by standard case-mix categories. Case-nix clefinitions, and the procedures anc! sample sizes required to attain a prespecified level of precision, should be established by the HCFA. Key Indicators The standard survey would consist of a number of key indicators, that is, outcome and process measures of quality of care and quality of life that are mostly resident-centered, although some relate to facility characteristics. Many of these indicators could be drawn from existing protocols. Examples of negative indicators include excessive use of psychotropic drugs, excessive rate of adverse drug reactions, high incidence of urinary tract infections among catheterized residents,

MONITORING NURSING HOME PERFORMANCE / 119 development of avoidable decubitus ulcers among physically dependent residents while in the nursing home, dehydration, contractures, avoidable declines in functional status, and unexplained weight changes. Examples of positive indicators are pain control and increased functioning in residents with angina, lowering of blood pressure in hypertensives, residents wearing street clothes, and service of palatable food. Some of the key indicators would come directly from the resident assessment data, especially those resident outcomes based on change over time, as shown in medical and other types of facility records. But most would be measured by the inspectors through direct observation, interviewing, and assessment of the case-mix-referenced sample of residents and of the facility environment. Several states have attempted to refine the federal certification process to save money and concentrate scarce regulatory resources on facilities demonstrating poor care.6 These states include New York,7 Massachusetts,8 Wisconsin,9 Colorado, A and Illinois. ~i Ohio has a resident assessment system for reimbursement purposes that focuses on resident needs and service provision,~2~3 and Iowa has developed an outcome-oriented licensure survey that focuses on selected domains of quality.~4~~6 Evaluations of some of these survey systems that focus on key indicators of quality of care and quality of life, such as New York's sentinal health events, indicate they are at least as successful in detecting serious deficiencies in the quality of nursing home care as the current certification surveys.~7~8 Thus it seems possible to develop key indicators, many of them drawn from or modifications of existing protocols, that are resident-centered and oriented toward appropriate--and away from inappropriate--resident outcomes and care processes and that can differentiate between facilities on the quality of resident care and quality of life they provide. An example of a key indicator of potentially poor nursing and dietary care would be considerable weight loss (for example, 5 or more pounds) within 30 days (as determined from medical records and observations of

120 / NURSING HO3lE CARE residents). If a given percentage of residents experience such weight loss, the extended survey would examine records for acceptable reasons for weight loss (diagnosis of cancer, treatment of obesity, recent physical activity level changes, and so on). ~- ~ · . . . · ~. the surveyors also WOU1d examine tne current cometary program (caloric intake); observe residents for treatable conditions (poor or missing teeth, depression); observe meal presentation (temperature and taste of food): observe and interview ,, residents regarding eating habits, need for assistance devices or staff assistance, and food preferences; and investigate nursing staff levels and policies regarding food supplementation and nursing assistance in eating. Appendix F contains other examples of key indicators and followup procedures in the extended survey. Recommendation 4-5: The standard survey should rely on "key ind icators" of quality of resid ent life and care that would be prescribed by the HCFA. These key indicators would measure poor resident outcomes and other resident and facility conditions that night be caused by noncompliance wits' the federal conditions and standards and should be investigated f~crt1~er by the survey agency. Triggering an Extended Survey: Scoring and Decision-Making After the number of poor outcomes or inadequate care procedures involving each key indicator is determined, the survey team would have to decide whether there are enough poor cases to warrant an extended survey. For the time being, these essentially normative decisions would have to be based on the judgment of the survey team or the policies of each state agency. In the long run,-however, it should be possible to have data on national or regional norms, controlled for case mix, for each key indicator, that can be used as the bases for these decisions. For each category of resident sampled in the standard survey, facility performance would be evaluated using key indicators. The contents of the particular instruments

MONITORING NURSING HOME PERFORMANCE / 121 and protocols used should differ for the four categories of residents and should be tailored to those aspects of care and quality of life that are most relevant and appropriate for each group. The proportion of the protocols for these different classes of residents that are based on interview, observation, and record reviews will also vary with the functional and mental status of the resident being evaluated. For the sample of mentally competent residents, for instance, the standard survey should include an interview protocol that is designed to determine their level of satisfaction with the quality of care they receive and with their quality of life. Also, their views of the facility's performance, including such things as its flexibility in matters of rising and retiring, its arrangements for privacy, and consideration of food preferences in meal planning. The interview might also include the residents' perception of staff attitudes toward and treatment of the mentally impaired residents. For the subset of mentally competent residents who are clinically depressed, the protocols will be designed to determine the adequacy of the facility's diagnosis and treatment of the condition. For mentally impaired residents, the protocols will involve more observation and will focus on the appropriateness of their care and the nature of the activities program provided. The incidence of such undesirable practices as excessive use of physical or chemical restraints would be checked relative to the number of facility residents at risk for such practices. Cases of neglect and of verbal and physical abuse should be determined. (The complaint files may be a source of such data.) For physically dependent residents, protocols will include a review of the presence of potentially avoidable negative outcomes such as decubitus ulcers, urinary tract infections, contractures, malnutrition, and dehydration. Surveyors will use the audited data supplied by the facility to estimate the incidence of these and other such undesirable outcomes in relation to the numbers of residents at risk for such outcomes. These incidence data, as well as findings about the care provided to

122 / NURSING HO3lE CARE sampled residents, will be used by surveyors to reach decisions about the adequacy of facility performance. The number of physically independent and mentally unimpaired residents should be small in most nursing homes. The focus of the survey protocols for them should be on issues relevant to their functional and mental or emotional status, the availability of activities, alleviation of pain, and maintenance of or slowing of deterioration of function in their activities of daily · · . living. Pass/fail and other scoring criteria for facility performance in each area, and for each category of resident sampled, should be established in advance. At the beginning, however, scoring will have to be more discretionary until analyses of the data base from the residents' assessments reveal the population-based outcome norms for each key indicator. When the data become available, policy-level normative decisions on cutoff or pass/fail scores can be made. Facilities that score below the cutoff point in any area will then receive an extended survey for those areas in which their performance appears to be questionable or deficient. Whenever there are doubts or ambiguities about the adequacy of a facility's performance on the standard survey, a partial or complete extended survey will be conducted to clarify the situation. Citations for noncompliance will be made only after completion of an extended survey. A facility that required no extended survey following the standard survey would thus pass inspection and be exempt from routine review for a certain period of time. Careful consideration will have to be given to deciding how many instances of poor care or negative but avoidable outcomes should constitute failure and trigger an extended survey or citation for noncompliance. A single instance of resident abuse or serious neglect might constitute grounds for an extended survey and citation. For other undesirable outcomes the absolute number or percentage of residents manifesting the condition that should constitute "failure" will need to be determined.

MONITORING NURSING HOAGIE PERFORMANCE / 123 Reconzn~enciation 4-6: Facilities that perform poorly on key indicators of quality of resident care or life should be subjected to a full or partial extended survey, depending on the range of problem areas discovered. The purpose of the extended survey is to determine the extent to which the facility is responsible for the poor outcomes due to noncompliance with the federal conditions and stand arcis. Survey Data Sources As already noted, the standard survey would rely primarily (but not exclusively) on outcome and process indicators of facility performance. Extended surveys, whether partial or complete, would sample more residents and include more structural and process indicators than the standard survey. Both the standard survey and extended surveys should be based primarily on observation of and interviews with residents and staff. Examination of facility records and written procedures would be secondary. Information may also be solicited from sources outside the facility, such as ombudsmen, community organizations, and residents' families and friends. Instruments used for the extended survey, whether partial or complete, will be designed to elicit information that is more detailed, comprehensive, and intrusive in nature. In keeping with the committee's view that the residents themselves should be the focus of attention, and that resident status is best determined by direct contact between the surveyor and the resident, facility records will be used to validate observations, to check the accuracy and thoroughness of professional evaluations and facility tracking of resident needs and interventions, and to help locate the source of problems or weaknesses in facility performance. There are potential difficulties in interviewing nursing home residents. First, there are questions as to the reliability and validity of the findings based on such

124 / NURSING HO3lE CARE interviews, given the nature of the population being interviewed (the frail elderly) and their circumstances (as a somewhat "captive" population in an institutional setting). In addition, there may be scheduling and logistical problems, as well as ethical issues, associated with interviewing a physically and psychologically vulnerable population. Despite these difficulties, there is evidence that such interviews are useful and produce valid information. The Iowa licensure surveyors, for example, routinely interview residents about feelings of comfort and social adjustment, sense of freedom, perception of fairness in terms of their treatment by staff, feeling of security, and enjoyment of food. They are convinced of its usefulness. Interviews with a sample of nursing home residents who are mentally competent, willing, and able to be interviewed without ~ undue physical or psychological strain, can yield important information about the day-to-day performance of the facilities and the residents' satisfaction with the quality of care and life they experience.~9320 Recommendation 4-7: Quality assessment in the survey process should rely heavily on interviews with, and observation of, residents and staff, and only seconclarily on "paper compliance," such as chart reviews, official policies anc! procedures manuals, and other indirect measures of actual care given and resident outcomes. Coordination with Complaint Programs Although complaints can be an important source of information about substandard conditions and form the basis for potential enforcement actions, federal guidelines about complaint handling are general (State Operations Manual, Section 3500), and sufficient staff for adequate complaint handling is not always available to state licensure and certification agencies. Each state agency has its own way of handling complaints. The committee's survey found that 46 (of 47) state agencies conduct complaint investigations. Forty-one reported that their state had a statutory complaint and abuse reporting

MONITORING NURSING HOME PERFORMANCE / 125 system, 34 operated by the survey agency and 7 by another agency. Of the survey agencies, 10 had separately staffed complaint units; the others used regular surveyors. Complaints are a potentially important source of infor- mation about compliance between annual certification inspection visits. There should be HCFA guidelines for analyzing and reporting complaints. The complaint procedure should include criteria for deciding whether to conduct an investigation, whether to conduct an onsite visit, how to schedule followup visits, and when to cite deficiencies. Complaint histories might also be used to decide whether to initiate an earlier survey or go directly to an extended survey or both. State licensure and certification agencies should be required to work out cooperative agreements for the reporting and handling of complaints with their state ombudsman program and the Medicaid fraud unit as well as with any state-mandated patient abuse or complaint programs. (The Medicaid fraud unit often obtains information relevant to a facility's compliance with licensure and certification standards.) Recommendation 4-8: The HCFA should require states to have a s pecif ic proced ure and su ff icient sta If to properly investigate con~plaints. Consumer Involvement The principle of resident-centered standards is furthered by direct resident interviews. Testimony at the public meetings and at the consumer issues workshop conducted by the committee in December 1984 noted that in most states current survey procedures do not require communication with residents before, during, or after the survey. Residents are an important source of information to surveyors. If surveyors seek information directly from residents and inform them at the conclusion of a survey of deficiencies and plans of correction, survey objectives are more likely to be achieved. Information provided to residents at the conclusion of a survey should not abridge the confidentiality of individual residents' care or

126 / NURSING HOME CARE records. The recommendation below also provides a means for inviting further information without compromising the unannounced survey. It is intended to facilitate effective communication between residents and surveyors. Recommendation 4-9: The HCFA sI'ould incorporate in its survey operations manual the following adclitional procecIures to be followed by surveyors in addition to interviews with those residents sampled for the survey protocols: · At the beginning of the survey surveyors should meet briefly with members of the facility s resident council or with a group of willing and capable resiclents to elicit general information about services and resident satis- faction as well as to identify any areas of particular concern. · Resident re preventatives shouic! participate in the part of the exit conference where deficiencies are cited and the plan of correction is discussed. · At the close of the survey the following notice shouIc! be posted in a location accessible to residents anc! visitors: The (state s''rvev ~oencvJ completer! its regular certification survey of (facility name) on (date). Anyone wishing to provide additional information may contact the (state s'~rvev agency) before Elate). (A d d ress J (Phone ) Positive Incentives Facilities that pass the abbreviated survey will receive regulatory relief by not having to submit to further inspections until the next annual survey except for being subject to a random extended survey (and unless there is an ownership or other change requiring a new survey or a pattern of complaints triggers a new survey). Surveyors and state agencies should be encouraged not to limit their comments to noting deficiencies, but to praise good or outstanding performance when they see it, both

MONITORING NURSING HOME PERFORMANCE / 127 privately and publicly. The use of letters of recommen- dation for outstanding performance (perhaps for no deficiencies for two or more consecutive inspections) should be explored. If surveyors find some good or outstanding aspects of a facility along with some deficiencies, they should not fail to note both at the exit conference. The effect on staff and management attitudes and morale is certain to be positive. The HCFA should introduce these concepts into its manuals and its training program. Recommend ation 4-10: In abolition to exempting goon' facilities from extencled surveys, ways should be explored to commend superior performance. Continuing Improvement of the Survey Process An effective and efficient survey process, like the conditions and standards it applies, cannot remain static. Survey procedures must be adapted to changes in the characteristics of residents (for example, increasing age and disability) and of nursing homes (size, staffing). More important, they must be updated as knowledge increases about the conditions and problems of nursing home residents, and with improvements in care techniques. Administration of the survey process also must be monitored and evaluated to improve consistency and en ~ 1clency. The development of new and better methods to assess quality of care and nursing home performance should be encouraged. Reliability and Validity of Instr''~q~e,~ts and Procedures The survey procedures should be designed to implement the new resident-centered, outcome-oriented conditions and standards and should be revised regularly as the conditions and standards evolve in the light of new knowledge and other changes (in resident, facility,

128 / NURSING HO3lE CARE or staff characteristics, and so on). Such changes will be especially common in the early years as the new system is implemented and the resident assessment data base develops. Recon~n~endation 4-11: The new survey protocols, including the forms, procedures, and guidelines used by surveyors, should be designed in accordance with the revised and amend ed cond itions and stand ard s recon~n~end eel in Cha pier 3, and they should be revised as the conditions and standards are changed in the future. It is important that the survey instruments and procedures be tested so that when used by properly trained surveyors they produce consistent and reliable findings. Recommend ation 4-12: All survey protocols (instruments and procedures) should be tested so that they are capable of yielding reliable and consistent results when used by properly trained surveyors anywhere. Survey findings must be valid and reliable as well as consistent--they should be capable of determining the extent to which a facility is in compliance with the conditions and standards of participation. This is particularly important for the standard survey which relies upon obtaining data from a sample of residents. To assure the validity of the standard survey, extended surveys should be taken in a random sample of facilities each year and the results compared with the findings of standard surveys of the same facilities. In addition to providing data for improving the conditions and standards, these surveys would provide a check on how well the two-stage survey process is working and should induce facilities to stay in compliance with all regulatory requirements, not just those that might be checked by the standard survey. Recommendation 4-13: A sample of facilities should be subject to an extender! survey each year. Information from this sample should be used to validate and improve the standard survey.

MONITORING NURSING HO3IE PERFORMANCE / 129 Consistency of Survey Results A major criticism of the survey process by providers of long-term care has been the inconsistency of surveyors' interpretations of their findings on what constitutes acceptable or deficient performance. The results of a survey are likely to be dependent on the professional and personal values and biases of the individual surveyors. It is therefore essential that state survey agencies make a serious effort to increase consistency of interpretation and decision-making by surveyors. It should be possible to improve surveyor consistency by means of better training, monitoring, and evaluation of surveyor performance as well as better design of survey instruments and procedures. Such monitoring and training are done only in a few states, but such activities are essential to ensure consistency in translating survey findings into judgments of nursing home compliance with conditions and standards. The importance of adequate training for surveyors to achieve consistency cannot be overemphasized. Such training should focus on the development among surveyors of a common language for describing what is observed during the course of a survey and the conclusions that are reached, techniques of eliciting relevant and useful information while surveying a facility, and methods and common points of departure for discussing a facility's performance and problems with its management, among other things. This training should not only increase the reliability and consistency of surveys, but also enhance the credibility of surveyors as a group with facility managers. Recommendation 4-14: The HCFA should require the state agencies to in~plen~ent a program to develop and support consistent and reliable surveys. This program should be based on effective training and monitoring of surveyor performance to reduce inconsistency.

130 / NURSING HOME CARE PaCS: A NEW HCFA SURVEY PROTOCOL In 1984, the HCFA began to test a new resident-centered survey process that focuses on the provision of services and resident outcomes. It was named PaCS (for Patient Care and Services). The HCFA developed PaCS to redirect the survey process from emphasizing facility structure and theoretical care-giving capacity toward evaluating the actual delivery of care and its outcomes. The PaCS process was based on the preliminary results of a series of state experiments with demonstrations of modified survey processes (for example, the final evaluations of experiments and demonstrations in New York, Wisconsin, and Massachusetts, Washington, and Iowa. A PaCS survey encompasses- 1. evaluation through direct observation of certain aspects of the physical environment, including cleanliness, space, equipment, infection control, and disaster preparedness; 2. detailed review of care provided to a sample of residents, through observation, interviews, and medical record reviews; 3. evaluation of meals, dining, and eating assistance by observing meal service; and 4. observation of drug administration for a sample of residents. Currently, PaCS is being evaluated experimentally in three states--Connecticut, Rhode Island, and Tennessee. In addition, all other states have been asked by the HCFA to administer PaCS in a small number of SNFs with good compliance histories. More recently, the PaCS survey process has become the HCFA's response to the court's decision in Smith v. Heckler that the HCFA produce a more effective regulatory process for assuring adequate quality of care in nursing homes. The HCFA plans to evaluate the PaCS experiments, make any needed modifications in the process and implement it as early as April 1986. The committee has reviewed the PaCS forms and

MONITORING NURSING HOME PERFORMANCE / 131 accompanying guidelines and has heard from state survey agency officials who have used the new survey process. There is general agreement that PaCS is a significant improvement over the traditional survey process, primarily because it focuses on resident outcomes rather than facility capacity and record reviews. In concept, it resembles the recommendation for a standard survey protocol made earlier in this chapter. It is a step in the right direction, but much additional work remains to be done before PaCS could become a valid and reliable resident-outcome-oriented survey protocol. Five major problems exist with its present form. First, PaCS is being implemented without changing the conditions and standards, which remain oriented toward facility and capability and do not include quality-of-life factors. The conceptual problems of reliably relating findings from resident-centered data to compliance with structural standards have not been addressed. Major changes are necessary in the conditions and standards to make them more resident-centered and outcome-oriented before an effective survey process can be designed and implemented. Second, PaCS has not designed a formal protocol for sampling of residents for detailed reviews of care- giving. There is no requirement to stratify by case mix, for example, nor recognition of the requirements of valid sampling. The PaCS process leaves it to the surveyors to select a sample of residents and to decide, relying exclusively on surveyor judgment, on the proportion of undesirable outcomes that are beyond the facility's control, those being appropriately handled, and those that are due to oversight or neglect. It is essential to incorporate statistically defensible sampling procedures to achieve valid, consistent, and reliable findings that can be sustained in enforcement proceedings. Third, the PaCS survey process still relies on unguided surveyor judgment to make the important decisions of whether care problems demonstrated by a facility constitute deficiencies. PaCS does not have guidelines with criteria for making these decisions, but leaves them to unguided surveyor judgment. The PaCS experiments should therefore be carefully analyzed for interrater reliability in use of the instrument. As noted earlier in

1 32 / NURSING HO3lE CARE this chapter, although many key decisions must continue to be left to the professional judgment of surveyors, many aids to guide such judgment must be built into the survey system. Fourth, current PaCS procedures do not require the facilities to maintain standar~i resident assessment data. A sound quality assurance system for nursing homes has to rely on standard resident assessment data both for reliable case-mix groupings and for tracking changes ir; resident outcomes. This was the reason for recommending that a standard resident assessment procedure be added as a condition of participation. Fifth, PaCS does not integrate the PaCS survey with the inspection-of-care (IOC) function. The committee has recommended that IOC be combined with a resident-centered, outcome-oriented survey process to make it more efficient by preventing duplication. This depends on implementing the resident assessment program. The recommendations made in Chapter 3 and those made in this chapter constitute an integral package. PaCS, properly developed to take account of the problems enumerated above, could become the standard survey protocol discussed earlier. But as currently envisioned it is not conceptually or operationally part of a comprehensive revision of the nursing home regulatory system. It does not incorporate many of the other key changes in the nursing home performance criteria, in the survey process, and in the enforcement process that are necessary to make significant improvements in the regulatory system. INCREASING STATE REGULATORY CAPACITY To facilitate the attainment of regulatory goals, the federal government should help the states to increase their capacity to conduct effective, reliable surveys of nursing homes by providing the state survey agencies with enough resources (funds, training programs, and research results) to help achieve more adequate and consistent application and enforcement of federal standards.

MONITORING NURSING HOME PERFORMANCE / 133 Federal Funding of Certification Activities Funding for Medicare certification activities comes from the Medicare trust funds. Historically, although states have submitted estimated budgets each year to their regional offices, they have received Medicare allocations incrementally larger than their previous year's budget. In 1981 the Medicare certification budget was cut from $30 million to $25 million and, in 1982 it was cut to less than $14 million (see Table 4-1~. When Congress restored the funding in 1983, the HCFA tried to reallocate the funding among states more in accordance with workload. The costs of average long-term care and other health facility surveys are estimated on the basis of 1980 expenditures, and are updated annually with an inflation factor. The figure is multiplied by the number of facilities in each region to determine its allocation. But each regional office uses a different allocation method to distribute the funds among its states. For Medicaid, Title XIX authorizes the HCFA to match whatever the state spends in certain approved categories. From 1965 to 1972 the matching ratio was 75 percent federal to 25 percent state funds. In 1972 the law was changed to authorize 100 percent federal funding of the salaries, travel, and training of state surveyors. In 1980, Congress reduced the matching ratio back to 75 percent federal. Federal funding of the survey and certification program is modest in total amount--less than $70 million in fiscal year 1984--or about 0.6 percent of total federal Medicare and Medicaid expenditures for nursing home services. It is not distributed entirely according to a formula based on consistent criteria. The states are still feeling the effects of major cuts in the funding of Medicare surveys imposed in 1981-1982. The amount the states themselves contribute for the licensure part of the survey process varies greatly. As a result, the number of surveyors and the number of inspections (and their intensity as shown by average person-days of surveyor time in a facility) vary significantly from state to state. The committee's survey of state licensure and certification agencies found that the average expenditures

134 / NURSING HO3lE CARE TABLE 4-1 HCFA Expenditures for State Survey Agency Activities (in millions of dollars) · ~ Fiscal Year Medicare Medicaid Total 1977 $23.6 $33.2 $56.8 1978 24.9 36.2 61.1 1979 25.3 34.4 59.7 1980 27.4 38.4 65.8 1981 24.6 34.2 * 58.8 1982 13.6 31.8 45.4 1984 35.6 32.2 67.8 SOURCE: The Health Care Financing Administration 1984. Federal (Medicaid) matching for surveyor salaries, travel, and training was cut from 100 to 75 percent in 1980.

MONITORING NURSING HO3lE PERFORMANCE / 135 for nursing home licensure and certification surveys vary from $1,296 to $13,018 per nursing home (median = $4,700~. This variation apparently has more to do with the historical funding base provided by the HCFA and the willingness of a state to add state funds than to the current workload level (total number of facilities, adjusted by bed size), or mix (mostly skilled vs. mostly intermediate facilities, mostly large size vs. small, high vs. low Medicaid admission criteria, and so on). The recommendations made earlier in this report for a resident assessment system and a new survey process and procedures will require extensive training for all surveyors, training of nursing home staff, and improved and better supervision of surveyors by state licensure and certification agencies. This will require larger budgets for the state licensure and certification agencies. To facilitate cooperation by the states in introducing the new survey process and the resident assessment system and enhancing their survey staff supervisory capabilities, the Congress should once again authorize 100 percent federal support for state survey and certification activities (in nursing homes). This authority should be extended for 3 years to facilitate installation of the new system. After 3 years, the matching ratio should be reviewed and a permanent ratio involving some state participation reinstated. Recommendation 4-15: Title XIX of the Social Security Act should be amended to authorize 100 percent federal funding of costs of the nursing home survey and certification activities of the states. This authority should be extended for 3 years, after which time a federal-state matching ratio should be reestablished. The HCFA should develop a standard formula for distributing funds to the states under this authority so that each state is funded on an equal basis in proportion to its federal certification workload.

136 / NURSING HO3lE CARE State Surveyor Qualifications Federal regulations and the State Operations Manual are very general regarding survey agency staffing levels and qualifications. In practice, there are significant variations in the experience and educational backgrounds of the surveyors and the composition of the survey teams in each state, for example, how many nurses, generalists or sanitarians, and other specialists such as pharmacists, nutritionists, physicians are on the teams. Nationally, about half are nurses, a fifth are sanitarians, and most of the rest are engineers, administrators, and generalists 23,24 Surveyors come from a variety of backgrounds, and few have previous nursing home or long-term-care experience. Federal guidelines for survey staff composition permit states a great deal of latitude, and the HCFA's data on surveyors indicate that some states are not staffed adequately to conduct surveys that are more oriented to resident care. For example, at least one state had no nurses on its survey staff in 1982.23 In 1983, eight states had only one or two licensed nurses on staff.24 Recommendation 4-16: The HCFA should revise its guidelines to make them more specific about the qualifications of surveyors and the composition and numbers of survey team staff necessary to conduct adequate resiclent-centeredt, outcome-orientecl inspections of nursing homes. At a minimum, every survey team should include at least one nurse. For use on extended surveys, the survey agency should have specialists on staff (or, in small states, as consultantsJ in the clisciplinary areas coverer! by the conditions and stanalarcis (for exur,~ple, pharmacy, nutrition, social services, and activitiesJ. Federal Training Support Federal training requirements are minimal and federal training programs were cut back substantially in 1980-1981 because of budget constraints. According to the case

MONITORING NURSING HOME PERFORMANCE / 137 studies, the states vary greatly in the scope of their training efforts. Three-quarters of the surveyors had at least 10 hours of in-service training in 1982, but one-quarter had less than 10 hours and, of those, a third had none.24 Recommendation 4-17: Federal training efforts and support of state-level training programs shout/ be increased, especially during the period! of transition to the new survey process, and cluring the implementation of the new resident assessment condition of participation. Dissemination of Research and Evaluation Results Information about survey operations and their results are inadequate at the state and federal levels.4 Evaluation of the new survey system will depend on the availability of performance data. At the same time, the federal government should continue to sponsor experiments in improving the survey process.~2~22~25 The federal government- should disseminate the results of experiments sponsored by it or the states to the other states. Recommendation 4-18: National data about survey operations anc! results, and from any experiments and demonstrations sponsored! by the HCFA or the states, should be collected, analyzed, and disseminated by the federal government to facilitate continued improvement in survey method s. Federal Oversight and Sanctioning Responsibilities The HCFA regional offices have not been able to carry out their monitoring responsibilities effectively in part because of inadequate resources and procedures. Regional office personnel devoted to certification work totals about 300, or about 30 per regional office.

1 3 8 / NURSING HOME CARE The HCFA has three ways to judge state survey agency performance, other than paper reviews of survey documents. They are- 1. Validation surveys. Theoretically, the federal surveyors are supposed to conduct validation surveys of a 5 percent sample of nursing homes assess state survey performance. In practice, this goal rarely has been attained. After the number of federal surveyors was cut from 100 to 70 in 1981, the sample size was reduced to 3 percent. Moreover, the validation surveys are often not performed until several months after the state survey, making it difficult to prove that the state overlooked or misinterpreted deficiencies found by federal surveyors. The new outcome-oriented conditions and standards and the new survey process should make it possible to judge state performance in a more reliable and consistent way. This will undoubtedly require an increase in the number of federal surveyors. 2. Complaint investigations. Complaints pertaining to possible violations of federal requirements are usually referred to the appropriate state survey agency for investigation but they may be conducted directly by federal surveyors. In some cases, this should stimulate a "look behind" survey. 3. Look behind. The HCFA has long had the authority to review state survey and certification decisions and to deny federal Medicaid reimbursement to a facility that is improperly certified by a state survey agency. Technically, under this "old look-behind" provision, the HCFA did not have the authority to decertify Medicaid-only facilities, only the authority to recover from a state any federal funds paid to a certified facility on the grounds that the state had not followed correct procedures. In 1981 the Omnibus Budget Reconciliation Act gave the HCFA direct authority to cancel the agreement between the Medicaid agency and the facility for not meeting federal standards, as determined by an onsite survey by a federal team. This is called "new look-behind." However, it requires a full evidentiary hearing before an administrative law judge before the effective date of

MONITORING NURSING HOME PERFORMANCE / 139 termination (if the deficiencies do not pose an immediate and serious threat to patient health and safety). Termination can be further delayed pending appeal to an appeals council, and judicial review. This requirement for a prior hearing before an administrative law judge, except in cases of immediate and serious threats to residents, is not now required for other termination procedures under Medicare and Medicaid law. In the view of HCFA officials, it lessens the effectiveness of the new procedure. There have been several problems with federal oversight. First, in recent years, insufficient numbers of federal surveyors have precluded surveys of 5 percent of nursing homes as called for in federal procedures. Second, the nursing homes surveyed in each state are not for the most part randomly selected; most are selected because there has been a complaint or a pattern of complaints about care in the homes. Third, the lack of timeliness of these surveys further reduces their value for evaluating state survey performance. They often take place weeks or months after the state visit and thus do not constitute a limited check on the reliability of the state's results. Fourth, the HCFA is very limited in what it can do to states that do not carry out their federal surveying responsibilities. It does not have effective sanctions, short of terminating its agreement with the state (which has never been done), to use against states that underenforce or wrongly interpret federal standards. An intermediate sanction, such as reducing the amount of Medicaid matching funds, is needed. Recommendation 4-19: The HCFA should increase its capabilities to oversee state survey and certification of nursing homes and to enforce federal requirements on states as well as facilities by · adding enough additional federal surveyors to each regional office to ensure that the random sample of nursing homes surveyed each year in each state is large enough to allow reasonable inferences about the adequacy of the state's survey and certification activities;

140 / NURSING HO3IE CARE · scheduling "look-behind" surveys so that valid con~pc~risons can be node of the findings of federal anc! state surveys; arid · amending Title XIX of the Social Security Act to authorize the HCFA to withhold a portion of Medicaid matching funcis front states that perform inadequately in their survey and certification of nursing homes. ORGANIZATIONAL CHANGES Incorporation of Inspection of Care in the Survey Process Federal law and regulations currently require each state Medicaid agency to conduct at least one "inspection-of- care" (IOC) review of all patients annually to determine the appropriateness and quality of care given to recipients. The inspection of care involves a look at the care given to every Medicaid resident. It is done by a team of nurses and social workers, often with access to physician consultants. Traditionally, this inspection-of- care process has been performed independently of the facility surveys in all but a few states. Federal guidelines for IOC are general, and inspection-of-care programs differ widely in the way they are conducted, the size and qualifications of the inspection teams, and the scope of the review. Many focus on level-of-care determinations rather than quality-of-care problems and do not have resident assessment tools and techniques adequate to determine quality of care for regulatory purposes. In the past few years, some states have combined their inspection-of-care and survey staffs, usually for budgetary reasons. In some states, the processes are fully integrated--done by the same team on the same visit. In others, they are done separately, but the information derived from the two processes is shared. The responsible agencies regularly take joint action in some states. In most states, however, the two processes operate in isolation from each other.26~27

MONITORING NURSING HOME PERFORMANCE / 141 In the 47 states responding to the committee's survey, 17 licensure and certification agencies were also responsible for inspection-of-care reviews. In nine of the states, the same team conducted both IOC and the certification surveys on the same visit; in the other eight states, IOC was conducted by a different team or on a different visit? or both. Of the 46 states answering the question of whether IOC should be integrated with the survey process, 32 said they should be done by the same team or at the same visit or both. Another seven thought they should be separate functions under the same supervisor. Only seven advocated keeping them as separately administered functions. Inspection of care, as it is currently conducted in most states, provides resident-centered quality-of-care information that is not always available to or used by the certification surveyors. The survey and IOC should be combined because they are somewhat duplicative and IOC findings would help in the assessment of compliance with resident care standards. Combining IOC with the recommended new survey process would require a statutory change to permit reviews of a sample rather than of all residents. The transfer of IOC also will affect utilization review and control responsibilities. Currently, the regulations governing IOC are included under the general subject of utilization control. These regulations require each state Medicaid program to have a surveillance and utilization control program to (1) guard against unnecessary or inappropriate use of services, (2) minimize excess payments, and (3) assess the quality of those services. Utilization control must include for each recipient a physician's certification and periodic recertification of the need for nursing home care, a medical evaluation and a rehabilitation plan for admission, and a discharge plan. In addition, there must be a utilization review (UR) plan for each facility that includes periodic reviews of each recipient's need for continuing stay in a nursing home, medical care evaluation studies, and discharge plan reviews. The state cannot receive the full federal share of payments for Medicaid services provided in a facility that does not have a proper utilization review program.

142 / NURSING HOME CARE In most states, utilization reviews, including the continuing stay reviews, are done by facility-based UR committees. The annual IOC visit, with its 100 percent review of Medicaid recipients, is the means by which the Medicaid agency monitors the performance of the UR committees. With consolidation of ICFs and SNFs, UR committees will be required to determine the need for continued nursing home care. The annual IOC has been used for this purpose. Accordingly, if IOC is transferred, the survey agency would need to perform this audit function for the Medicaid agency. The effort should be directed at a sample of residents most likely to be discharged. This function would be greatly facilitated by the availability of the standard resident assessment data. The placement of residents in the nursing home could be checked at the time of the standard survey and reported to the Medicaid agency. If the placement decisions for the sample are wrong in too many cases, a review of all residents could be triggered. Recommendation 4-20: The inspection-of-care function s1~o''1d be carried out as part of the new resiclent-centered, outcome-oriented survey process. But individual resident reviews should be required for a sample of residents (private-pay as well as Medticaid) rather than for all residents (although individual states may elect to continue 100 percent reviews). Restructuring of State and Federal Roles and Responsibilities The federal and state role relationships in nursing home regulation must be clear and workable, because the two levels of government share the responsibility for maintaining the federal quality standards in nursing homes participating in the Medicare and Medicaid programs. In the past, federal statutes have given principal responsibility to the states for determining whether participating nursing homes comply with federal health and safety standards. The states do this by conducting onsite inspections and complaint investigations in all facilities

MONITORING NURSING HOME PERFORMANCE / 143 participating in Medicare and Medicaid. They certify th compliance or noncompliance of these facilities. In the case of Medicaid-only facilities, which account for 61 percent of the participating facilities and 53 percent of the beds,28 the state Medicaid agency makes the final decision to enter into a provider agreement with a certified facility. In the case of Medicare-only or Medicare and Medicaid facilities, however, it is the HCFA regional offices that make this decision. In both instances, the federal government's primary responsibility should be to monitor and assist the states in the performance of their jobs. One result of this difference in certification responsibilities for Medicare and Medicaid facilities has been federal preoccupation with Medicare SNFs and relative state autonomy over Medicaid-only facilities. Another result is state Medicaid certification of state-owned nursing homes and hospitals. It is a potential conflict of interest for a state to survey its own institutions. It puts the survey agency in the position of criticizing the performance of a sister agency (often in the same department) and, if it requires major state expenditures, it may come under pressure from the governor's office to modify its findings. Moreover, the survey agency is put at a disadvantage in taking a tough line with private facilities when it is widely believed that state facilities are borderline or worse. The respective roles of the federal and state govern- ments would be clarified and strengthened if the states assumed responsibility for approving certification of all (Medicare as well as Medicaid) facilities except state-owned institutions. The latter should be certified by the federal regional offices on the basis of inspec- tions by federal surveyors. The primary role of the regional offices would still be to monitor the activities of the state survey agencies and to take steps, including the use of the sanctions referred to in the previous recommendation, to ensure adequate performance. This recommendation concerning certification authority should be implemented by overhauling the so-called "1864 agreement"--the contract between the Secretary of Health and Human Services and each state health department to Le

144 / NURSING HOME CARE carry out Medicare surveys--assuming the following other recommendations of the committee are implemented: (1) the development and adoption of more outcome-orienteci conditions and standards anti of a new survey process to implement them, (2) provision of adequate resources and training to the states to carry out their certification responsibilities, (3) increased and improved federal monitoring of state survey performance, and (4) the adoption of federal sanctions to use against states that do not adequately apply or enforce federal requirements. Section 1864 of the Social Security Act directs the Secretary of Health and Human Services to make agreements with any "able and willing" state under which the state health department or other appropriate state agency surveys health facilities wishing to participate in Medicare and certifies whether or not they meet federal definitions, standards of care, and other requirements. In return, the secretary agrees to pay for the reasonable costs of the survey and certification activities of the state agency. Currently, 1864 agreements are open-ended in duration, but they may be terminated under certain conditions by either party. Although the HCFA has been dissatisfied with the performance of some states from time to time, it has never terminated an 1864 agreement. Because section 1864 compels the secretary to enter into agreement with any state that wants to, and does not provide for alternative sponsorship of survey activities, the HCFA has not had much leverage with states that do not strictly comply with federal requirements. The HCFA implemented a revised 1864 agreement on July 1, 1985, in an attempt to hold the states more accountable. It should continue this effort to clarify the respective roles of the federal and state levels in conjunction with the other major recommendations cited above, that is, implementation of a resident-centered, outcome-oriented standards and survey process and increased resources at the federal and state levels. It should be noted that the federal cost savings resulting from the elimination of the paper reviews of the certification packages in the regional offices should

MONITORING NURSING HOlIE PERFORMANCE / 145 offset in part the higher costs of the expanded federal oversight function called for in the last recommendation. Recommendation 4-21: The respective roles and responsibilities of the federal anc! state governments should be realigned as follows: · The states should be responsible for certifying all Med. icare and Med. icaid f acilities (exce pt state institutions) according to federal requirements. ~ The HCFA should monitor state performance more actively and be responsible for conducting surveys of, and certifying, state-owned institutions directly.

Next: 5. Enforcing Compliance with Federal Standards »
Improving the Quality of Care in Nursing Homes Get This Book
×
Buy Paperback | $108.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

As more people live longer, the need for quality long-term care for the elderly will increase dramatically. This volume examines the current system of nursing home regulations, and proposes an overhaul to better provide for those confined to such facilities. It determines the need for regulations, and concludes that the present regulatory system is inadequate, stating that what is needed is not more regulation, but better regulation. This long-anticipated study provides a wealth of useful background information, in-depth study, and discussion for nursing home administrators, students, and teachers in the health care field; professionals involved in caring for the elderly; and geriatric specialists.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!