Quality of Care and Nursing Staff in Nursing Homes

Jean Johnson, Ph.D., C. McKeen Cowles, M.S., and Samuel J. Simmens, Ph.D.

Overview Of Papers Prepared For The Institute To Medicine

This paper is comprised of two sections for committee review. The first section presents a broad picture of issues related to nursing homes and staff. It describes the nursing home environment, characteristics of nursing homes using trend data since 1992, roles and education of nursing staff, the relationship of resident needs to staffing, and quality-of-care measures in nursing homes. The second section presents the results of a study undertaken specifically to provide new information related to nurse staffing and quality of care in nursing homes. Both papers need to be considered jointly since the first establishes the context for the research described in the second.

The Context Of Nursing Care In Nursing Homes
Background

The relationship between quality of care and the number and type of nursing staff in nursing homes is complex. At present there is no definitive information

Dr. Johnson is associate dean for health sciences programs, The George Washington University; Mr. Cowles is a consultant; and Dr. Simmens is an assistant research professor in the Department of Health Care Sciences, The George Washington University.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 426
--> Quality of Care and Nursing Staff in Nursing Homes Jean Johnson, Ph.D., C. McKeen Cowles, M.S., and Samuel J. Simmens, Ph.D. Overview Of Papers Prepared For The Institute To Medicine This paper is comprised of two sections for committee review. The first section presents a broad picture of issues related to nursing homes and staff. It describes the nursing home environment, characteristics of nursing homes using trend data since 1992, roles and education of nursing staff, the relationship of resident needs to staffing, and quality-of-care measures in nursing homes. The second section presents the results of a study undertaken specifically to provide new information related to nurse staffing and quality of care in nursing homes. Both papers need to be considered jointly since the first establishes the context for the research described in the second. The Context Of Nursing Care In Nursing Homes Background The relationship between quality of care and the number and type of nursing staff in nursing homes is complex. At present there is no definitive information Dr. Johnson is associate dean for health sciences programs, The George Washington University; Mr. Cowles is a consultant; and Dr. Simmens is an assistant research professor in the Department of Health Care Sciences, The George Washington University.

OCR for page 426
--> to determine what exact nursing inputs will relate to specified resident outcomes. There is information emerging, however, that will help establish a set of relationships between resident needs, staffing, and quality. There has been considerable controversy over the past years about the minimum staff required to meet resident needs. A significant tension has existed between consumer and professional groups who propose increased numbers of registered nurses (RNs) and total staff, and payers, primarily state Medicaid agencies, who are concerned about the cost to the public of paying for additional staff. It is important to recognize that while minimum staffing levels are particularly tied to Medicaid payment, Medicare is an increasingly significant payment source through expansion to subacute services. It is likely, given the current funding crisis in the Medicaid and Medicare programs, that financing for nursing home care will be increasingly problematic, especially if these programs are cut. Financing of nursing home care will require potentially a new funding structure. If money were not an issue, there would be no need to constrain staff, and every nursing home resident could have their own RN care 24 hours a day. Many states have established a minimum standard for nursing time per resident per day. Federal requirements state that an RN must be present in a facility for 8 hours a day, 7 days a week, regardless of the size of the facility. It is difficult to argue against the need for a minimum number of staff members with defined capabilities to carry out the basic functions of nursing home care. However, the notion of a minimum number applied uniformly to all facilities is complicated by the fact that nursing homes have differing case-mix populations needing different levels of nursing expertise and different amounts of care. The problem with the concept of minimum staff is that it becomes translated to mean the maximum staff for payment purposes. In addition to the actual minimum levels, there is a more general requirement that there be sufficient staff to provide the required care to residents. However, there has been little guidance as to what ''sufficient" staff actually means. The definition of sufficient has been operationalized in conflicting ways. The regulatory agencies hold nursing homes to a standard of "highest practicable level" for resident care, yet the payment agencies provide funding for staffing at minimum levels. Quality of care has received increasing attention over the past 5 years, yet there has been relatively limited research examining the relationship of staffing to quality. There is some consensus about a limited number of indicators of quality care, but in general there remain issues of definition of quality. For instance, should the emphasis on quality measurements be on process or outcome? Who should define quality: the regulators, residents, family, or staff? Is there a consensus about measures of quality among these groups of stakeholders? Nursing Home Environment As the lines between the hospital, nursing home, and home care blur, the

OCR for page 426
--> definition of a nursing home has become less boundaried. The typical nursing home of the past provided care for primarily elderly individuals who needed assistance with some aspects of taking care of themselves or simply needed a sheltered place to live. The focus of care was generally custodial. If a resident had an acute problem they were promptly sent to the hospital. Nursing homes today provide a range of services to persons who are more disabled with an increased number of unstable chronic conditions (Morrisey et al., 1988; Shaughessy and Kramer, 1990). Care provided in a nursing facility may range from provision of ventilator assistance to rehabilitation for individuals with, for instance, hip fracture or stroke. It also ranges from care of a resident with an emerging acute care crisis, to caring for a resident whose death is expected and imminent. The general philosophy of care has shifted from one of custodial care to one of maximizing each resident's well-being. Specifically stated in the long-term-care regulation, "each resident must receive and the facility must provide the necessary care and services to attain or maintain the highest practicable physical, mental, and psychosocial well-being." Rehabilitation is now a basic conceptual framework for care. The types of services offered by nursing homes is changing as is the setting in which nursing home care is delivered. A nursing home may offer respite care or specialized care units such as an Alzheimer's unit. There are currently approximately 13,029 special care beds in nursing homes (Health Data Associates, 1994). A new category of care, subacute, is offered by many nursing homes. Subacute units provide care for residents with serious wound conditions, intravenous lines, rehabilitative needs including occupational therapy, physical therapy, and speech therapy, as well as medically complex patients. Traditional nursing home care is now offered in hospitals, life care communities, and residential care facilities. For some residents the nursing home will be where they live out the remainder of their lives, while others will only be in the home for a brief period of rehabilitation or respite care. With the requirement for advance directives, a major challenge to nursing homes is to provide more end-of-life care. Fewer residents are sent to hospitals to die (Sager et al., 1989). In addition, based on anecdotal information, more acute care problems are being managed within nursing homes to prevent the need for hospitalization. Characteristics of Nursing Homes A review of nursing home characteristics may be helpful in defining the context of nursing care. The characteristics that are both relevant and for which national data are available are included in Table 1. The 1994 data are obtained from the Nursing Home Yearbook (Health Data Associates, 1994). Data for the remaining years were directly computed from the On-line Survey Certification and Reporting (OSCAR) data set for those years. Occupancy has been about 85

OCR for page 426
--> TABLE 1 National Nursing Home Characteristics Characteristics 1992 (mean) 1993 (mean) 1994 (mean) Occupancy 84.81 85.07 84.79 Percent Medicaid 68.33 68.61 68.79 Case-mix 10.48 10.56 10.61 Hours of staff RN .31 .32 .34 LPN .60 .61 .62 NA 2.09 2.09 2.08 Total 3.00 3.01 3.04 Deficiencies related to nursing care Level A 0.13 0.15 0.13 Level B 8.11 8.01 7.41 Total 8.24 8.16 7.54 percent over the past 3 years. As indicated, Medicaid plays a large role in financing nursing home care with approximately 68 percent of residents over the past 3 years being covered by Medicaid. The case-mix index has increased slightly each year from 10.48, to 10.56, then 10.61. Registered nurse time has also increased slightly over the past 3 years, with 0.31, 0.32, 0.34 hour per resident per day respectively. Licensed practical nurse (LPN) time has also very slightly increased. Nurse assistant (NA) time has remained stable at about 2.09. The mean number of citations were calculated for nursing-care-related deficiencies. Expert nurse and ombudsman groups identified federal requirements that were most directly related to nursing care. Deficiencies in any of these requirements comprised the nursing care deficiencies (Johnson-Pawlson, 1993a). The mean number of level A deficiencies has been reasonably constant at 0.13. The mean number of level B deficiencies has decreased from 8.11 in 1992 to 7.41 in 1994. Nursing Staff Characteristics, Retention, Roles, and Educational Preparation Nursing Characteristics Nursing staff in a nursing home are primarily comprised of RNs, LPNs, and NAs. According to data from the 1992 nursing survey, 7 percent of RNs work in nursing homes (Moses, 1994). The proportion has been virtually unchanged since 1986 (National Center for Health Statistics, 1988). Of the RNs currently in nursing homes, 45.5 percent are diploma prepared, 31.7 percent are associate

OCR for page 426
--> degree prepared, 19.2 percent are bachelor's prepared, and 6 percent are master's prepared. The average turnover rate for nursing staff in 1994 was: RNs (56.3 percent); LPNs (52.5 percent); and NAs (100.4 percent) (American Health Care Association, 1995). The salary level for nursing staff in nursing homes is significantly lower than for hospital-based nurses. Based on the 1992 National Sample Survey of Registered Nurses, an RN in a staff position in a hospital earned $36,618, while staff nurses in a nursing home earned $31,298. The differential is greater at the supervisory level. The average salary of a supervisor in a hospital is $42,948, while in a nursing facility it is $32,569. The average nursing home LPN salary is comparable to the salary level of LPNs in hospital (American Health Care Association, 1994). Salary levels for NAs are lower in nursing facilities ($6.33 per hour) compared to the hospital setting ($7.31 per hour) (American Health Care Association, 1994). Retention As noted above, the average national turnover rate is very high for all levels of staff. A turnover rate of 100 percent for NAs and over 50 percent for RNs and LPNs indicates a major problem in continuity of care, which is an important factor in providing good care to residents with chronic conditions. Disruption of staff leads to residents having to constantly "train" staff, problems in carrying through care plans, inaccurate assessment because of failure to be familiar with the baseline status of a resident, and failure of a facility to fully develop a philosophy of caring. Inefficiency also exists because of the time needed for constant recruitment efforts. Staff are less productive as new staff spend time getting to know the residents and procedures of the facility, and stable staff spend time helping to orient new staff. It should be noted that even though NA staff turnover may be 100 percent, there is usually a reasonably stable group of NAs. The key issue is whether the stable core of NAs is large enough to create a critical mass to maintain stability of care. Role Licensed Nurses Licensed nurses, including RNs and LPNs, have two major functional roles in nursing homes: clinical and management. The clinical role requires a broad knowledge of nursing because of the wide variation in resident problems and needs. A licensed nurse is responsible for the overall care of the residents. This includes being able to accurately assess the status of residents, define physical, psychological, and social problems and strengths, and develop plans of care that extensively incorporate rehabilitation as well as health promotion concepts. Registered nurses work with residents whose main interests may not be based on their health problems but on the management of everyday living,

OCR for page 426
--> which may include concern about dying, relationships with family members, or what will happen to the house they have left. It is extremely important that the nurse accurately and comprehensively assess all problems. Life threatening problems such as fever, trauma from a fall, or change in mental status carry a particular burden for timeliness and accuracy. Accurate assessment and management can save the life of a resident and may assist in avoiding unnecessary transfer to the emergency room. It has been noted that early detection and monitoring of a resident with a fever can prevent dehydration and the need for hospitalization (Weinberg et al., 1994). Effective nursing interventions must be based on a broad range of competencies in the physical, psychological, and social realms. In order to provide consistent care to individuals, care that is planned by other health disciplines such as social work, physical therapy, or a consulting psychiatrist must be incorporated into the day-to-day care provided by nursing staff. In many cases nursing staff must substitute for health professionals that are not on site. An example is if a medical problem arises, most facilities have no daily on-site medical coverage (Karuza and Katz, 1994). One study reported that the mean response time between staff notification of an acute event and a physician response by telephone was 5.12 hours, which means the nurse is critical to the well-being of residents in these situations and requires a high level of nursing competency in clinical decision making (Brooks et al., 1994). The clinical care is very complicated, given that residents are likely to have multiple chronic illnesses requiring complete understanding of each illness and the interacting effects. The nursing care problems include not only physical problems, but mental and behavioral problems as well. There is a high prevalence of depression in nursing homes. In addition, a major challenge to nursing is the care of residents with dementia. It has been estimated that over half of all nursing home residents have memory impairment. This is frequently accompanied by agitated and aggressive behavior which has significant implications for the well-being of other residents and staff. Residents are frequently on complicated medication and treatment regimens requiring extensive pharmacologic knowledge. An RN or LPN may be the only licensed nurse on a particular unit or in the facility during evenings, nights, or weekends to ensure that the clinical needs of residents are met. Being the sole responsible clinician requires a nurse to have considerable clinical expertise. Whereas historically nurses in nursing homes were viewed more as house mothers than professionals, current resident care requires a highly qualified nursing staff, with licensed nurses needing a broad base of knowledge spanning basic nursing, rehabilitation, and psychiatric skills. The management role of licensed nurses is neither very well understood nor recognized. Most nurses, whether RNs or LPNs, have significant management responsibilities. A major management responsibility is staff supervision. A licensed nurse must oversee the direct resident care that is provided by NAs as

OCR for page 426
--> well as other licensed nurses. Given that NAs are minimally trained for resident care activities, oversight is crucial. Ensuring follow-through with therapeutic programs such as bladder training or behavioral management requires consistent staff performance and continual interaction between the supervising nurse and staff. In addition, establishing and reinforcing the expectation that all residents are treated with respect and caring is often a very challenging proposition. Management responsibilities in addition to staff supervision include resolution of conflicts, conducting patient care conferences, organizing systems of care such as bladder training programs, and conducting quality assurance activities. It has been recognized that organization of bladder training and restraint reduction take significant organizational skills on the part of licensed nurses (Schnelle et al., 1991; Werner et al., 1994). In addition, nurse managers must be able to work within interdisciplinary groups to conduct care planning. Nurses are usually responsible for interdisciplinary care planning and coordination of care. Nurse Assistants The nurse assistant has provided the backbone of care in nursing homes. This individual is responsible for most of the day-to-day assistance with activities of daily living (ADL), providing direct care and comfort, noting abnormalities such as pressure sores or change in mental status, and participating in rehabilitative care such as bladder training, use of assistive devices for eating, and strengthening exercises. The communication skills of an NA are very important in being able to effectively work with and involve residents in their care, convey a caring attitude, and provide information to residents and other care providers. Given the level of direct care responsibilities and the minimal preparation required for NAs, it is critical that licensed nursing staff be available and work closely through constructive supervision with nurse assistants. Education Understanding the educational process that prepares the nursing work force is critical to evaluating the capacity of the current work force to meet the needs of residents. Up until 1990, there was no uniform requirement for nurse assistant training. Some states required 100 or more hours of training, while others required none. With the passage of the Omnibus Budget Reconciliation Act of 1987, nurse assistants are required to have 75 hours of training and must complete a written and performance competency exam within 4 months of hire. While this is a landmark requirement, there has been no effective evaluation as to whether or to what extent this has improved the quality of care in nursing facilities. Licensed practical nurses provide a significant contribution to the care of residents and account for the majority of licensed nurses in nursing homes. Training of LPNs averages between 12 and 13 months, focusing predominantly on technical aspects of care such as medication distribution, wound care, and cath-

OCR for page 426
--> eter care. A program may offer specific content in geriatrics. For instance, an LPN program in the District of Columbia offers approximately 108 hours of clinical experience and approximately 50 hours of classroom time in geriatrics. This same program also offers 1 week of leadership training. While the content of most LPN programs offer similar curricula, it is clear that LPN education prepares individuals for the technical role of nursing but is limited in its leadership and clinical decision making content, both of which are critical to effectively functioning in a leadership position in a facility. Registered nurses include three major groups: those prepared at the diploma level, the associate degree level, and the bachelor's degree level. The vast majority of nurses in nursing homes are diploma or associate degree prepared. Information from a survey done by the Community College-Nursing Home Partnership suggests that there is a limited inclusion of gerontology in the associate degree nurse (ADN) curriculum (Hanson, 1992). While 66 percent of programs provide a nursing home experience in the first year, only 20 percent include one in the second year. This suggests that the nursing home is the site for basic skills development, rather than the more advanced course work. Since that report, considerable work has been done by the partnership to expand the gerontology curriculum at the ADN level. Diploma programs are hospital based by definition. The number of these programs has decreased significantly over the past two decades with the rise of ADN and bachelor of science in nursing (BSN) programs. However, diploma nurses comprise a significant number of RNs in nursing homes. The hospital-based education of diploma graduates clearly limited their exposure to nursing homes. A recent report on BSN curricula suggests that 78 percent of respondents to a survey have participated in clinical experiences in long-term care and 83 percent have clinical experiences in geriatrics (Johnson, 1995). This is consistent with earlier information from the American Association of Colleges of Nursing that 75 percent of schools integrated gerontology content into clinical courses and 74.8 respondent schools incorporated long-term care into clinical course work. It is not at all clear from curricula information for BSN programs, however, that the focus of the content is to prepare nurses to take a leadership role in nursing home care. Based on anecdotal information and personal observation, most clinical experiences in geriatrics, if in nursing facilities, relate to basic nursing skills and are not offered as required senior-level nursing management courses. In addition, gerontology content could easily be taught in an hospital. BSN nurses have the most extensive education in leadership and clinical decision making skills of any of the RN levels. There are issues that affect all levels of licensed nurses. One is that there are no specific accreditation requirements related to experiences in nursing homes. Nursing homes have been more a setting of last resort than one that is valued as a legitimate practice site. Nurses from an intensive care background continue to

OCR for page 426
--> receive greater recognition than nurses in nursing homes. Another concern is the scarcity of faculty who are expert in nursing home care. There are increasing (although limited) numbers of nurses who have some expertise in geriatrics, yet have no nursing home experience. Most nurses who have extensive clinical experience in nursing homes do not have the academic credentials to be on a faculty of nursing, nor perhaps the desire. With a shortage of faculty prepared to be a role model and teacher, it is very difficult to build a cadre of graduating students with the knowledge and skills to effectively provide clinical and management expertise in nursing homes. Finally, few states have continuing education or re-examination requirements to maintain licensure. Health care information is changing at an ever more rapid pace, necessitating that nurses have current information. Without continuing education requirements for licensure, however, some nurses may be working with woefully outdated information. One might argue that the current regulations governing nursing homes that require continuing education for all staff address this issue. The quality and content of these programs, however, are extremely variable. Case-Mix and Staffing Needs Case-mix is an important factor in examining the relationship between resident needs and nursing staff required to meet those needs. Substantial and seminal work has been conducted as part of the Multistate Nursing Home Case-Mix and Quality Demonstration project sponsored by the Health Care Financing Administration (HCFA) beginning in 1989 (Fries et al., 1994). The goal of the project is development of a payment and quality monitoring system for nursing homes. The basic principle of the case-mix project is that resources should be allocated based on resident need and that sicker and more debilitated residents need more services, both in terms of amount of staff time as well as level of expertise. To date, work has included development of categories of resource need and measurement of staff time to meet those needs. The next step in the project is to link this with quality-of-care measures. As a result of research into case-mix, the third iteration of a resource utilization grouping (RUGS-III) has been developed. RUGS-III development, like other methods of relating resident needs to resource use, relies heavily on measures of activities of daily living which have been recognized as the most important predictor of the cost of nursing home care (Butler and Schlenker, 1989; Weissert and Musliner, 1992; Williams et al., 1994). RUGS-III is based on several levels of categorization of resident need. The first categorization identifies resident types, reflecting patients with certain conditions. The levels are: rehabilitation, extensive services, special care, clinically complex, impaired cognition, behavior problems, and (reduced) physical functions. These categories of residents represent a hierarchy of groups with different needs, the rehabilitative

OCR for page 426
--> group having the highest resource need. Each of these major categories are then further divided into subcategories which form 44 different RUGS-III groupings. Based on the case-mix project, it is evident that the need for nursing time and skills varies significantly depending on the resident classification. The case-mix project measured actual nursing times needed for various levels of nursing. Nursing times were classified as either resident-specific time or non-resident-specific time. Resident-specific time included time spent performing services for an individual resident and interactive time. Non-resident-specific time included included routine operations and procedures necessary for the day-to-day function of a nursing unit, including staff meetings, team conferences, routine charting and documentation, unit administration, supervision and the like. Based on the case-mix data, there is evidence of a wide variation of total nursing time needed for different categories of RUGS. For instance, the category requiring the least care is the first level of the reduced physical function group. Based on minutes per resident per day, the nursing times calculated for the group are: RN, 13 minutes; LPN, 24 minutes; and NA, 67 minutes. This is a total of 104 minutes. This is contrasted with a resident in the highest resource need group requiring the following times: RN, 82 minutes; LPN, 63 minutes; and NA, 186 minutes, to total 331 minutes per day of nursing care. As the resource need increases, so does the need for more skilled nursing time as well increased total staff time. There is evidence that case-mix has evolved toward a more debilitated resident population following implementation of diagnosis-related groups (DRG) (Morrisey et al., 1988; Shaughnessy and Kramer, 1990). Analysis of the OSCAR data set as presented above indicates a continuing increase in case-mix since 1992. The first surge in case-mix took place shortly after the implementation of DRGs. However, it is significant that the case-mix index is still increasing, which is consistent with what would be expected under continued pressures for hospitals to decrease length of stay. Quality of Care There has been considerable attention paid to the issues of how many staff and what types of staff are needed to meet the needs and expectations of nursing home residents as measured by outcomes. A review of current information on quality of care and nurse staffing suggests there is a relationship between staffing and resident outcomes. Mezey and Lynaugh (1989) reported a reduction in pressure sores, use of physical restraints, and falls when nursing school faculty participated in care at nursing homes. Munroe (1990) found a significant relationship between a higher ratio of RNs to LPNs and fewer deficiencies noted on surveys in California after controlling for several variables, including case-mix. A study of nursing homes in Maryland indicates that higher total staff levels are related to fewer nursing deficiencies (Johnson-Pawlson, 1993a). Another study

OCR for page 426
--> found a relationship between higher staff levels and lower RN turnover and resident functional improvement (Spector and Takada, 1991). Cherry (1991) found RN staffing to be associated with better resident outcomes as measured through a composite of indicators, including number of resident developing decubitus ulcers per immobile resident, number of residents catheterized per incontinent resident, number of urinary tract infections per incontinent residents, and rate of antibiotic use per resident. Finally, a study of pressure sores in three Department of Veterans Affairs nursing homes concluded that better outcomes in preventing pressure sores was likely due to more favorable staffing (Rodman et al., 1993; Brandeis et al., 1994). The measures relating to deficiency citations need to be carefully viewed. Federal nursing home requirements have been established using a very extensive and inclusive process to identify measurable criteria by which to determine whether or not a facility is providing adequate care, but there is evidence that the survey process is not applied uniformly. There are serious questions about whether the current survey process produces uniformly reliable and valid determinations of nursing home care (Abt Associates, 1993; Johnson-Pawlson, 1993b; Office of the Inspector General, 1993). It is likely, however, that surveyors are reasonably accurate at the extremes in identifying very good facilities and very bad ones. The determination of quality cannot be left solely to health professionals and regulators. The resident and family perception of quality of care has not received much attention. The expectations of family members or surrogates may differ from those of the resident. A study of the ability of a surrogate to represent the resident's view of quality of care indicates that surrogates cannot accurately express the resident's perspective in all areas of nursing home care (Lavizzo-Mourey et al., 1992). There may also be a difference in what a family member considers quality of care for a resident with severe dementia as compared to a family member of a resident who has severe emphysema. Information concerning what residents want is described in a study done by the National Citizen's Coalition for Nursing Home Reform. The findings of this study suggest that nursing home residents most value staff being nice to them, including smiling and treating them with dignity (National Citizens' Coalition for Nursing Home Reform, 1986). Focus groups conducted in Maine by DeSisto (1994) indicate that residents rarely identified clinical care, such as receiving medications, as being important. The areas they considered important included good food, privacy, being treated with dignity and respect, activities, and comfort. There is a study currently being conducted by the American Health Care Association on resident perceptions of quality. Data are not yet available from this study.

OCR for page 426
--> TABLE 5 T Test of Type of Ownership for High- and Low-Quality Facilities   Types of Ownership   For-profit Chain For-profit Nonchain Not-for-Profit Government   High Quality (N = 423) Low Quality (N = 95) High Quality (N = 433) Low Quality (N = 58) High Quality (N = 466) Low Quality (N = 54) High Quality (N = 107) Low Quality (N = 3) Staff RN .35a .28 .34a .26 .53a .34 .48 .34 LPN .62 .64 .58 .58 .68 .67 .73 1.02 NA 1.96 2.07 2.01 2.01 2.25 2.25 2.41 2.45 Total staff 2.94 2.99 2.93 2.85 3.47 3.27 3.62 3.80 Total beds 85.67a (42.16) 118.79 (47.07) 83.30a (63.18) 138.14 (80.86) 83.07a (62.39) 144.44 (66.62) 88.10 (95.98) 195.67 (65.62) Occupancy 87% (0.16) 89% (0.11) 88% (0.15) 87% (0.13) 87% (0.19) 89% (0.14) 83% (0.22) 82% (0.17) Percent Medicaid 63%a (0.26) 73% (0.13) 73% (0.23) 70% (0.21) 51%a (0.30) 63% (0.23) 61% (0.29) 76% (0.02) Case-mix 9.35a 11.15 9.41a 10.72 9.75a 11.11 9.99 11.66 a Significant at p < .0001.

OCR for page 426
--> TABLE 6 T Test of Facility Characteristics by Urban or Rural Location for High and Low Quality   Rural   Urban     High Quality Mean (standard deviation) Low Quality Mean (standard deviation) High Quality Mean (standard deviation) Low Quality Mean (standard deviation) Staff RN 36a (0.28) .27 (0.22) .47a (0.37) .31 (0.18) LPN .59 (0.38) .63 (0.26) .68 (0.42) .64 (0.26) NA 2.04 (0.58) 1.99 (0.38) 2.17 (0.73) 2.15 (0.58) Total staff 3.00b (0.95) 2.89 (0.58) 3.32b (1.19) 3.10 (0.80) Percent Medicaid 60a (0.25) 75 (0.15) 63a (0.31) 68 (0.20) Case-mix 9.60a (1.65) 10.63 (1.38) 9.5a (2.47) 11.19 (1.29) Percent occupancy 88 (0.15) 90 (0.12) 86 (0.19) 87 (0.13) Total beds 70.00a (37.37) 103.98 (44.02) 97.59a (73.86) 143.22 (67.84) a Significant at p < .0001. b Significant at p < .001.

OCR for page 426
--> TABLE 7 Pearson Correlation Coefficients for Nursing Staff and Facility Characteristics   Hours per Resident per Day   RN LPN NA Total Staff Coefficients Probability Occupancy -0.32 -0.31 -0.30 -0.39 Medicaid -0.39 -0.18 -0.18 -0.30 Case-mix -0.11 -0.15 -0.26 -0.25 NOTE: All of the coefficients in this table are significant at p < .0001. quality urban facilities had 0.47 compared to 0.31 hours per resident per day of RN time. The percent of Medicaid-covered residents was significantly less in the high-quality rural facilities compared to low-quality facilities, 60 percent versus 75 percent respectively. This same relationship existed for Medicaid in urban facilities, with 63 percent of residents covered by Medicaid in high-quality facilities compared to 68 percent in low-quality facilities. Another consistent difference was that high-quality facilities have a statistically significant lower case-mix index, 9.6 in high-quality facilities versus 10.63 in low-quality facilities in rural areas, and 9.5 versus 11.16 respectively in urban facilities. Finally, high-quality facilities are smaller than low-quality facilities in both urban and rural areas. The average bed size in high-quality rural areas is 70.00 compared to 103.98. In urban areas, high-quality facilities have an average of 97.59 beds, compared to 143.22 beds. Bivariate analyses of nursing staff and facility characteristics, displayed in Table 7, indicate a statistically significant relationship between hours per resident per day for each level of staff and occupancy rate, percent of Medicaid residents, and case-mix. As one would expect, an increase in occupancy is related to a decrease in nursing time. Also, as the percent of Medicaid residents increases, there are fewer hours per resident per day of nursing time. Contrary to what would be expected as case-mix increases, the amount of staff time decreases. Correlation coefficients for facility characteristics are presented in Table 8. These data indicate a significant positive relationship between occupancy and percent Medicaid. There is a negative relationship between total beds and occupancy suggesting that larger facilities have lower occupancy. There is also a positive relationship between total beds and both case-mix and Medicaid, suggesting that larger facilities fill their additional beds with Medicaid residents and residents who are more debilitated.

OCR for page 426
--> TABLE 8 Pearson Correlation Coefficients for Facility Characteristics Coefficients Probability Occupancy Percent Medicaid Case-Mix Total Beds Occupancy — 0.26a 0.01 -0.39b Percent Medicaid — — 0.11a 0.08b Case-mix — — — 0.13a Total beds — — — — a Significant at p < .0001. b Significant at p < .01. Results of logistic regression are displayed in Table 9. Controlling for all other variables, RN time was significantly related to whether a facility provided high- or low-quality care. The odds ratio indicates that a 15-minute increase in RN time per resident per day increased by 1.73 the odds of a facility providing high-quality care as opposed to providing low-quality care. Size of facility provides a powerful explanatory variable for quality. Facilities with fewer than 60 beds are 9.74 times as likely to be a high-quality facility and facilities with between 61 and 120 beds are 2.28 times as likely to be high-quality facilities than the reference group. Each unit increase in case-mix (3–18 scale) decreased the likelihood of a facility being a high-quality facility by .66. The relationship of type of ownership to high- and low-quality facilities is significant. The reference group for the ownership variable was the for-profit chain group. Based on the regression results, a nonprofit facility is 1.69 times as likely to be a high-quality facility as the reference group; government-owned facilities were 7.02 times as likely to be a high-quality facility as the reference group; and for-profit, nonchain facilities were 1.62 times as likely to be a good facility as the for-profit chain facilities. The government-owned group is a very small group and limited conclusions can be drawn from these data. Finally, rural facilities are 1.58 times as likely to be in the high-quality category. Discussion It is important to note that the high-quality facilities selected in this study are not facilities that are necessarily providing care beyond federal and state requirements. They are facilities that, based on the criteria used for selection, are doing what they are supposed to do. The picture that emerges from this analysis is that RNs, case-mix, type of ownership, and urban or rural location are important factors in determining the probability that a facility will be a high- or low-quality facility. The influence of

OCR for page 426
--> TABLE 9 Logistic Regression Predicting High Quality Facilities; Predicators Re-Scaled for Clarity Variable Measurement Unit Parameter Estimate Standard Error p Odds Ratio Lower Limit Upper Limit Intercept   5.07 0.59 0.0001       Staff RN res/day 1/4 hour 0.55 0.12 0.0001 1.73 1.38 2.17 LPN res/day 1/4 hour 0.04 0.08 ns 1.04 0.89 1.22 NA res/day 1/4 hour -0.01 0.04 ns 0.99 0.91 1.07 Medicaid percent/10 -0.03 0.04 ns 0.97 0.90 1.05 Size < 60 beds 0–1 2.28 0.31 0.0001 9.74 5.25 18.05 61–120 beds 0–1 0.83 0.18 0.0001 2.28 1.60 3.27 ≥ 120 beds [reference group]             Case-mix 3–19 scale -0.42 0.05 0.0001 0.66 0.59 0.72 Ownership Nonprofit 0–1 0.52 0.21 0.01 1.69 1.12 2.54 Government 0–1 1.95 0.63 0.002 7.02 2.05 24.04 For-profit 0–1 0.48 0.20 0.02 1.62 1.09 2.41 For-profit chain [reference group]             Location Rural 0-1 0.46 0.19 0.02 1.58 1.09 2.30 Urban [reference group]            

OCR for page 426
--> Medicaid needs to be examined further. High-quality facilities had, in general, fewer Medicaid-covered residents except for the nonchain, for-profit facilities. Medicaid was not statistically significant in the logistic regression, however, probably due to colinearity with other variables such as size and RN, LPN, and NA staff, as evidenced in the bivariate analysis. These data indicate the importance of RN staff in a nursing home. High-quality facilities not only staff at higher levels than the low-quality facilities, they also staff at levels higher than the national 1994 mean (0.42 compared to 0.34 hours per resident per day) for RNs, while low-quality facilities staff at less than the national mean (0.29 hours per resident per day). The match between services needed by residents and the capabilities of RNs is the likely explanation for the RN effect. While nursing homes of the past may have cared for a relatively stable, self-sufficient population, that no longer accurately describes the current set of residents. Residents have multiple, complicated needs demanding considerable clinical judgment and capability. In addition, the management skills of RNs may also be reflected in the better outcomes. Registered nurses may provide the necessary supervision needed in working with a predominantly nonprofessional staff in a difficult environment. The finding that a low-quality facility would be nearly twice as likely to become a high-quality facility if it increased its RN staff by 15 minutes per resident per day (or 3.3 RN FTEs per day per 100 residents) has significant implications for staffing. A rough estimate of the cost of adding 3.3 RNs in the 210 low-quality facilities (assuming the average number of beds per facility is 100) would be approximately $22 million. This estimate is based on the national average salary of an RN in a nursing home being $31,298, and does not include fringe benefits or costs of recruitment. If the cost of staff required to bring the 14,000 facilities that fell into neither category into the high-quality category is considered, the total cost could well be in the hundreds of millions of dollars. The issues relating to the cost and quality trade-off are significant. State Medicaid budgets are already stretched beyond capacity, and additional RN time will cost additional dollars that currently are not available. The importance of staffing based on case-mix is critical. Clearly, as case-mix increases so does the need for additional staff, particularly RNs to meet those needs. The low-quality facilities were trying to care for residents who were significantly more debilitated than those in the high-quality group, using the same number of LPN and NA staff but fewer RNs. This situation then produced facilities that had the highest number of pressure sores per number of bedbound residents, the greatest medication error rates, and the highest use of restraints. The ongoing work of HCFA in the Case-Mix Demonstration Project is very important to determining the levels of staff needed to achieve desired outcomes. As nursing homes care for increasingly debilitated residents, there is a need for sufficient staff both in numbers and capabilities. The correlation between case-mix, percent Medicaid, occupancy, and total

OCR for page 426
--> beds supports the notion that larger facilities take more Medicaid residents and residents who are more debilitated in order to fill their beds. The impact on facilities of additional Medicaid residents who are sicker may be the inability to afford increased numbers of staff, particularly if payment is not adjusted for case-mix. Smaller facilities can be more selective in taking less debilitated residents, given they have fewer beds to fill. This study raises issues about the impact of ownership and quality. The for-profit facilities that are not chain owned fell into the high-quality category at as high a rate as nonprofit facilities while having substantially less staff and the highest proportion of Medicaid-covered residents. Exploring how these facilities manage to provide high-quality care under these circumstances could provide a better understanding of the influence that ownership exerts on care. These findings may reflect a direct commitment and oversight by the owners of nonchain, for-profit facilities that ultimately enhances care. It should also be noted that chain-owned, for-profit facilities fell into the poor category at a higher rate than expected. This could be due to chain-owned facilities not having a direct accountability or to a management structure in some chain-owned facilities that does not provide effective oversight of quality of care. It should be noted, however, that facilities represented in the low-quality group could be clustered among a few nursing home chains and do not represent all for-profit-chain facilities. Another important result in regard to type of ownership is that the not-for-profit group of facilities has the least percent of Medicaid residents, while the for-profit, nonchain facilities have the highest percent. One might assume that since the not-for-profit facilities receive some degree of public support though their tax status, not-for-profit facilities would have a greater commitment to providing a public good by taking more Medicaid-supported residents. This, however, is not the case. A possible, but unexplored, explanation may be that the not-for-profit facilities provide a greater amount of free care to residents and that residents simply don't apply for Medicaid assistance. The finding that rural facilities are more than twice as likely to be a high-quality facility may be explained by several different dynamics. A rural facility may be more sensitive to its community's needs than an urban facility. The community sensitivity may be due to staff knowing the residents they care for and being concerned about the reputation of the facility. Even in rural areas, however, facilities that fell into the low-quality category had a higher percent of Medicaid residents, larger facility size, higher case-mix, and fewer RNs. It is also important to further examine different impacts on rural and urban facilities, since urban facilities in general staff at higher levels than rural facilities. A possible explanation for the higher staff levels is that the turnover rate is greater in urban areas because of numerous opportunities for jobs in the minimum wage category. Large turnover may require additional staff to provide continuity of care. Nursing staff in rural areas may be ''locked in" to their jobs because of

OCR for page 426
--> limited options in the rural community. There could also be a payment bias for higher staffing in urban areas. Finally, overall case-mix was somewhat higher in urban areas, which may account for some portion of the difference in staffing. The consistent finding that high-quality facilities have fewer beds, after controlling for other variables, suggests that there may not be economies of scale for increasing the size of nursing facilities. The influence of large size may reflect the physical inaccessibility of staff to residents, inability of supervisory staff to effectively oversee care, or the difficulty of a facility in establishing a clear philosophy of caring. A larger facility may also have more difficulty recruiting and retaining competent staff. Finally, the quality indicators used to select facilities relate solely to physical care. It is not at all clear that the high-quality facilities are those that also provide humane, time intensive end-of-life care or manage residents with behavioral problems well. The authors believe that much could be learned by further examining the highand low-quality facilities extensively and systematically using qualitative as well as additional quantitative methods of investigation. Limitations The major potential limitation of this study is based on use of the OSCAR data for staffing information. Based on the authors' experience with the staffing data, the edits that were done to eliminate major errors produced staffing numbers that were within the range of staffing data reported in Medicaid and Medicare cost reports. In addition, the relationships that were found in this study were consistent with other studies. Another limitation is that quality was defined using physical parameters and not psychologic or social parameters, even though these may be very important to residents. Conclusions Based on information presented in both sections of this paper, the overview of nursing homes and the analysis of the high- and low-quality facilities, several important conclusions can be drawn. These include: While the total staff time per resident per day was virtually the same for each category of care, after holding case-mix and other variables constant, the amount of RN time was very important. By increasing RN time by 15 minutes per resident per day, a facility in the low-quality category would have increased the probability nearly twofold that it would fall into the high-quality category. Given the relationship between RNs and quality of care, RN participation in care is very important. The federal requirement of one RN for 8 hours a day, 7 days a week, is not sufficient to ensure quality of care to residents. Case-mix must be the basis for determining staffing needs. The low-quality

OCR for page 426
--> care facilities had a higher case-mix, the same level of total staffing, but fewer RNs. In order to provide adequate care, facilities must adjust staffing to accommodate residents who have a greater resource need. Increasing the number of RNs becomes very important as the case-mix index of a facility increases. For all types of ownership, high-quality facilities had higher RN staff levels than the low-quality facilities. The not-for-profit and government-owned facilities, however, had higher RN levels for high-quality facilities than either category of for-profit high-quality facilities. It will be useful to examine the for-profit, nonchain group of facilities in that they staffed at lower levels than nonprofit and government-owned facilities, yet had no higher than expected number of facilities in the poor care category. Any discussion of staffing must take into account financing of staffing needs. Medicaid is the major payer for nursing home care, but Medicaid budgets are likely to decrease rather than increase. Data from this study indicate that low-quality facilities have a higher proportion of Medicaid residents. Given that Medicaid rates are usually lower than private pay rates, this may contribute to financial constraints in hiring additional staff. Policymakers are faced with an extremely tough choice with the trade-off between quality and cost. Considering that the population is aging and there is no cure for Alzheimer's and other chronic diseases, the demand for nursing home care will likely expand and funding mechanisms will need to ensure adequate staffing to meet the care needs of residents. Additional research is needed to continue exploring the relationship between staffing inputs and resident outcomes. The case-mix project has gone a long way to examine resource needs and staffing needs. Yet the work of this project will not be complete until resident needs and staffing are linked to quality-of-care outcomes. In addition, a research agenda should be developed that focuses on qualitative aspects of nursing home care, including further study of resident and family needs and definitions of quality, and organizational characteristics affecting care. The categorization of high- and low-quality facilities was based on physical measures, which can at best provide only a partial examination of quality. Given the importance of RN staff, it is necessary that educational programs prepare RNs for a leadership role in nursing facilities. If LPNs are to substitute for RNs and remain an important level of nursing staff, their capabilities in managing their role in nursing homes will need to be enhanced so that they can have a greater effect on positive resident outcomes. References Abt Associates, Inc. Briefing Points on Preliminary Evaluation Requests. HCFA Leadership Conference, 1993. American Health Care Association. Facts and Trends: The Nursing Facility Sourcebook. Washington, D.C.: The Association, 1994.

OCR for page 426
--> American Health Care Association. Facts and Trends: The Nursing Facility Sourcebook. Washington, D.C.: The Association, 1995. Brandeis, G.H. Ooi, W.L., Hossain, M., et al. A Longitudinal Study of Risk Factors Associated with the Formation of Pressure Ulcers in Nursing Homes . Journal of the American Geriatrics Society 42:388–393, 1994. Brooks, S., Warshaw, G., Hasse, L., and Kues, J.R. The Physician Decision-making Process in Transferring Nursing Home Patients to the Hospital. Archives of Internal Medicine 154:902–908, 1994. Butler, P.A., and Schlenker, R.E. Case-mix Reimbursement for Nursing Homes: Objectives and Achievements. Milbank Quarterly 67:103–135, 1989. Cherry, R. Agents of Nursing Home Quality of Care: Ombudsmen and Staff Ratios Revisited. The Gerontologist 31:302–308, 1991. DeSisto, M. Residents Perspective of Nursing Home Care. Presentation at Maine Health Care Association meeting, 1994. Fries, B.E., Schneider, D.P., Foley, W.J., et al. Refining a Case-Mix Measure for Nursing Homes: Resource Utilization Groups (RUG-III). Medical Care 32:668–685, 1994. Hanson, H.A. Highlights of National Survey of Gerontological Nursing in the AND Curriculum. The Community College–Nursing Home Partnership Newslink, 1992. Health Data Associates. Nursing Home Yearbook. Tacoma, Wash.: Health Data Associates, 1994. Institute of Medicine. Improving the Quality of Care in Nursing Homes. Washington, D.C.: National Academy Press, 1986. Johnson, J.Y. Curricular Trends in Accredited Generic Baccalaureate Nursing Programs Across the United States. Journal of Nursing Education 34:53–60, 1995. Johnson-Pawlson, J. The Relationship Between Nursing Staff Variables and Quality of Care in Nursing Homes. UMI Dissertation Services. O.N. 9316112, 1993a. Johnson-Pawlson, J. Surveyor Performance Study. Washington, D.C.: American Health Care Association, 1993b. Karuza, and J., and Katz, P.R. Physician Staffing Patterns Correlates of Nursing Home Care: An Initial Inquiry and Consideration of Policy Implication. Journal of the American Geriatrics Society 42:787–793, 1994. Lavizzo-Mourey, R.J., Zinn, J., and Taylor, L. Ability of Surrogates to Represent Satisfaction of Nursing Home Residents with Quality of Care. Journal of the American Geriatrics Society 40:39–47, 1992. Mezey, M.D., and Lynaugh J.E. The Teaching Nursing Home Program. Nursing Clinics of North America 24:769–780, 1989. Morrisey, M.A., Sloan, F.A., and Valvona, J. Medicare Prospective Payment and Posthospital Transfers to Subacute Care. Medical Care 26:685–698, 1988. Moses, E. The Registered Nurse Population. Findings from the National Sample Survey of Registered Nurses, March 1992. Washington, D.C.: Division of Nursing, Bureau of Health Professions, Health Resources and Services Administration, U.S. Department of Health and Human Services, 1994. Munroe, D.J. The Influence of Registered Nurse Staffing on the Quality of Nursing Home Care. Research in Nursing and Health 13:263–270, 1990. National Center for Health Statistics. Characteristics of Registered Nurses in Nursing Homes: Preliminary Data from the 1985 Nursing Home Survey. Advance Data from Vital and Health Statistics, No. 152. Prepared by G. Strahan. Pub. no. (PHS) 88–1250. Hyattsville, Md.: National Center for Health Statistics, U.S Department of Health and Human Services, 1988. National Citizen's Coalition for Nursing Home Reform. Quality of Care from a Nursing Home Resident Perspective. Washington, D.C.: The Coalition, 1986. Office of Inspector General. States' Progress in Carrying Out Nursing Home Survey Reforms. Washington, D.C.: Department of Health and Human Services, 1993.

OCR for page 426
--> Rodman, D., Slater, E.J., Richardson, T.J., and Mattson, D.E. The Occurrence of Pressure Ulcers in Three Nursing Homes. Journal of General Internal Medicine 8:653–658, 1993. Rodman, D., Bross, D., and Mattson, D.E. Clinical Indicators Derived from the Patient Assessment Instrument in the Long-Stay Residents of 69 VA Nursing Homes. Journal of General Internal Medicine 9:261–267, 1994. Sager, M.A., Easterling, D.V., Kindig, D.A., and Anderson, O.W. Changes in the Location of Death After Passage of Medicare's Prospective Payment System. New England Journal of Medicine 320:433–439, 1989. Schnelle, J.F., Newman, D.R., Fogarty, T.E., et al. Assessment and Quality Control of Incontinence Care in Long-term Nursing Facilities. Journal of the American Geriatrics Society 39(2)165–171, 1991. Shaughnessy, P., and Kramer, A. The Increased Needs of Patients in Nursing Homes and Patients Receiving Home Health Care. New England Journal of Medicine 322:21–27, 1990. Spector, W.D., and Takada, H. Characteristics of Nursing Homes that Affect Resident Outcomes. Journal of Aging and Health 3:427–454, 1991. Weinberg, A.D., Pals, J.K., Levesque, P.G., et al. Dehydration and Death During Febrile Episodes in the Nursing Home. Journal of the American Geriatrics Society 42:968–971, 1994. Weissert, W. and Musliner, M. Access, Quality and Cost Consequences of Case-Mix Adjusted Reimbursement for Nursing Homes: A Critical Review of the Evidence. Washington, D.C.: Public Policy Institute, American Association of Retired Persons, 1992. Werner, P., Koroknay, V., Braun, J., and Cohen-Mansfield, J. Individualized Care Alternatives Used in the Process of Removing Physical Restraints in the Nursing Home. Journal of the American Geriatric Society 42:321–325, 1994. Williams, B.C., Fries, B., Foley, W.J., et al. Activities of Daily Living and Costs in Nursing Homes. Health Care Financing Review 15:117–135, 1994.