Nursing Homes: Special Care Units

A movement to establish dementia special care units (SCUs) began in the United States in the 1960s. The movement gained momentum, becoming widespread in 1980s and 1990s. This growth was fueled by the recognition that Alzheimer's disease and other progressive dementias represent a serious public health problem, and by the advocacy of the Alzheimer's Disease and Related Disorders Association to improve care for individuals suffering from dementia. SCUs were also established in many nursing homes because it was recognized that their presence improved the image of the nursing home and helped to attract privately paying patients. Another impetus for establishing SCUs was the Omnibus Budget Reconciliation Act of 1987 (OBRA 1987), which emphasized a decreased use of physical and chemical restraints. In addition, the growing number of old-old persons, and thus the growing number of persons with dementing illnesses, mandated the development of "quality care" for such persons. For these reasons, SCUs have become more prevalent—in 2001 they existed in almost one-quarter of the nursing homes in the United States, almost double the number of SCUs that existed in 1990. Special dementia care programs are also being developed in residential settings.

SCUs vary greatly in terms of the number and composition of staff, the nature of activity programs, and patient care, but several features are generally found in all such units: they house only residents with dementing illness, and they offer specialized staff training, specific activity programming, family involvement, and a segregated and modified physical environment. However, all of these features are not necessarily present in any single SCU. As a result, while some SCUs reflect an honest effort to enhance the quality of care given to this population, some reflect primarily the response to a marketing imperative, and the degree of care enhancement in such facilities, is, at best, questionable. For example, a survey of 436 nursing homes indicated that not all SCUs offered richer or more tailored services for dementia than did non-SCUs, and that most non-SCUs offered some dementia-specific features similar to those made available in SCUs. Another survey, which evaluated SCUs and non-SCUs according to the presence or absence of activity programming, mental health support, and nursing rehabilitation, found that 12 percent of SCUs did not have any of these special programs, and another 20 percent were deficient in two of the categories. Insufficient staffing and sparse activity programming was also found in a survey of nursing homes in the eastern United States: one-fourth of the facilities provided only minimal features, such as secure units.

Such differences fostered the creation by some states of standards for SCUs. These standards include elements such as: separate written policies and procedures, secured entries and exits, modification of the physical environment, including a separate dining room, access to a secure outdoor area, special staff education and training, specific staffing levels; and specific activities and social support. There is some evidence that these standards have had a positive impact on the type and quality of services provided by SCUs.

Prevalence estimates of dementia in nursing homes vary widely, mainly due to differences in study methodology. Studies that use direct assessment by trained clinicians yield higher estimates (65 percent to 75 percent) than surveys that use chart data or administrative data sets. A recent study of a new admission cohort (Moguziner, et al.) yielded an estimate of 48 percent. About 80 percent to 90 percent of nursing home residents are cognitively impaired. Recent findings suggest that all SCU residents suffer from a dementing illness. Among these residents, almost two-thirds are communication-impaired; 68 percent to 85 percent are in the late-stages of dementing illness; over half have severe-stage cognitive impairment; and, because of their level of dementing illness, almost half are not testable on neuropsychological tests. These prevalence estimates are significantly and substantially higher than are those found among non-SCU residents.

Perhaps dwarfing other considerations, the potential cost of providing care to persons with dementia in specialized settings is staggering. Nursing-home expenditures constitute a large and rapidly growing component of overall health care expenditures. In 1997, approximately $82 billion was spent on nursing-home care in the United States—about 8 percent of total health care expenditures. It is estimated that by 2030 the level of nursing-home health care spending will rise to $1,477.4 billion—9.25 percent of the estimated $15,969.6 billion total health care spending. While many disease processes lead to nursing-home placement, persons with Alzheimer's disease and related dementias are likely to contribute, at a minimum, 20 percent of total nursing home expenditures. However, given the increasing proportions of nursing-home residents with cognitive deficits, this almost certainly constitutes an underprojection.

A question remains as to whether the amount, and thus the cost, of personal care differs between special care and traditional care settings. A related question is the degree to which differential impacts are associated with such differences as have been found to exist. It is important to distinguish between costs (monetized service-time inputs) and charges made by nursing homes for services to residents. While the latter usually are regarded as estimates of the former, it is possible, perhaps even probable, that charges are sensitive to market pressures and, therefore, do not accurately reflect the actual costs associated with care for different individuals in different settings. The issue of possible differences in costs of staff inputs made in SCUs, as contrasted with counterpart costs in traditional care units, is particularly important for several reasons. First, staff inputs account for over 80 percent of the total costs of nursing-home care. Similarly, staff time expenditures constitute the largest component of cost of care that relates directly to the characteristics of individual residents. Second, in addition to the stipulation that SCUs house only residents with dementing illness, the majority of definitional criteria usually applied in making a distinction between special care and traditional care are staff related. Adhering to the view of SCUs as a separate form of nursing care, one would expect there to be higher staff ratios, more staff time spent in therapies and in staff support and supervision, and to have more SCU time spent in case management, including reporting and planning. Third, staff inputs are the most immediately malleable major component of nursing home care: staff ratios, assignments, and patterns can be changed at will. In contrast, environmental changes usually take far more time to plan and implement.

Most evaluations to date, however, have shown that there are few, if any, differences between SCUs and traditional nursing-home units in terms of any of the dimensions along which such differences might be expected. In fact, there is a growing tide of feeling which holds that special care is really quality care for the growing proportion (now the majority) of nursing home residents who suffer from dementing illnesses. On the other hand, there is some evidence that there are slight (and statistically significant) differences between SCUs and traditional units (favoring the former) in terms of aide time spend serving residents, and in the impacts that increases in aide time have in SCUs. For example, using advanced data-gathering techniques, it has been shown that, on average, SCU residents receive almost eight minutes more (of an average total of slightly more than fifty-one minutes per resident, per day) of aide time than do their non-SCU counterparts in facilities which maintain both kinds of units. This difference increases to almost twenty-one minutes per day when the SCU residents are contrasted with residents of traditional facilities that do not maintain SCUs. Perhaps more important, from a cost-of-service point of view, it was found that additional aide service inputs gave more "bang for the buck" in SCUs than in traditional units—that is, for whatever reason, substantial increases (more than forty minutes average increase per day, per resident) in aide time had far greater effect in reducing resident agitation in SCUs than it did in non-SCUs.

Generally, it appears that SCU residents incur greater costs of inputs than do residents of more traditional types of units in terms of aides and speech therapists. However, although statistically significant, the differences between SCUs and traditional units tend to be clinically and economically trivial (e.g., less than three dollars per resident, per day, between SCUs and traditional units). However, these differences do become substantial when applied to all residents on a unit. For example, for a typical thirty-six-bed SCU, the difference is the equivalent of about one extra aide, divided between the day and the evening shifts. Additionally, investigators have suggested that both SCUs and non-SCUs (in nursing homes which support both) provide more aide and RN time per day on the day shift than do traditional units. This may reflect the fact that the presence of an SCU in a facility can have a beneficial facility-wide effect in terms of staff training and lower staff turnover. It is discouraging to note, however, how relatively few minutes are spent in the provision of direct care in either type of facility.