RNs, LPNs/LVNs, and NAs are employed in a wide variety of inpatient, home health, and ambulatory HCOs. Many of these organizations have undergone turbulent changes in response to the rapid evolution of the U.S. health care system over the last 20 years. The relationships between these organizations and their nurse employees have been turbulent as well. Many of these HCOs report large vacancies in nursing positions and serious difficulties in securing enough nursing staff to care for patients.
Wide Variety of Health Care Settings for Nursing Staff
While RNs are employed primarily in hospitals [see ], LPNs/ LVNs are about equally employed in hospitals and nursing homes [28 and 29 percent, respectively]. Another 14 percent of LPNs/LVNs work in physicians' offices and clinics [Bureau of Labor Statistics, undated]. Nursing homes employ the largest proportion of NAs [see ]. The populations served in these settings have some differences in their health care needs. These differences, changes in the U.S. health care system, and changes in the ways nursing care is delivered have shaped all nurses' work environments, but especially hospitals, nursing homes, home care and community-based organizations, and public health agencies.
TABLE 3-1
Primary Employment Settings of RNs Employed in Nursing, 2000.
Hospitals
Hospitals have historically been the largest employer of the nursing workforce and continue to be so today, although there has been a decline in the last two decades. The proportion of the RN workforce employed in hospitals peaked in 1984 at approximately 68 percent. By 2000 the proportion had declined to 59 percent as the result of a shift in care and nurse employment to noninstitutional settings [Spratley et al., 2000]. Most hospital nurses work on inpatient units; 53.7 percent of hospital RNs work in ICUs, step-down/transitional units, or general/specialty bed units [see ] [Spratley et al., 2000]. The deployment of nurses by hospitals has changed dramatically over the least two decades as hospitals themselves have changed.
TABLE 3-3
Types of Work Units in Which Hospital-Employed RNs Spend More Than Half of Their Direct Patient Care Time.
Fewer hospitals, fewer inpatient beds, and fewer [but more acutely ill] inpatients In the last two decades, hospitals have been under tremendous pressure to remain financially solvent in the face of a widely acknowledged oversupply of hospital beds, cost-containment measures resulting in changes in reimbursement from public and private payors, and demands for greater accountability for the quality of the care they provide. Between 1980 and 2000, the number of hospitals in the United States declined by 17 percent, the number of hospital beds by 28 percent, the number of hospital admissions by 10 percent, and the average length of patients' hospital stays from 7.6 to 5.8 days [American Hospital Association, 2002].7 Over about the same period, outpatient visits increased by more than 150 percent. By 1999, outpatient surgery constituted 50 percent of all hospital-based surgery—an increase from 16 percent in 1980 [American Hospital Association and The Lewin Group, 2001]. As a result of this downsizing and technological advances in care, patients admitted to the hospital today are more acutely ill than was the case in the previous decade [Medicare Payment Advisory Commission, 2001].
During this same period, the number of RNs working in hospitals increased substantially,8 although this increase was not uniform across all hospitals [Unruh, 2002], and not all of it should be assumed to represent an increase in RNs providing direct patient care. Data also indicate that many of the above downsizing initiatives were accompanied by reductions in unlicensed support staff, such as NAs [Aiken et al., 1996]. These changes were accompanied by changes in the ways nurses deliver care to patients and have been perceived as leading to an increased workload for nurses [as discussed further below].
Changes in approaches to care delivery As described in Chapter 1, many hospitals attempting to respond to the pressures of the last two decades have undertaken efforts to reengineer or redesign patient care processes to make them more efficient. Because nurses are the largest category of hospital workers, these reengineering efforts have often involved changing the ways in which nursing care is provided—typically through personnel reductions; cross-training of personnel to perform additional duties; changes in the mix of nursing staff [RNs, LPNs/LVNs, NAs]; reassignment of support services [e.g., laboratory, radiology] to nursing units; redistribution of patients across nursing units; redesign of patient care processes; use of clinical pathways; and other changes in organization structure, decision-making processes, and responsibilities of management and patient care staff [Aiken et al., 2000b; Norrish and Rundall, 2001; Ritter-Teitel, 2002; Walston et al., 2000; Walston and Kimberly, 1997].
In addition, redesign and reengineering have changed the way nursing staff are organized to provide patient care. Restructuring initiatives often have been marked by a departure from primary nursing and a return to variants of team nursing [Norrish and Rundall, 2001]. As initially conceptualized, the latter approach involved a team of RNs, LPNs/LVNs, and NAs, with an RN serving as the team leader. The RN team leader determined assignments for team members consistent with their abilities and performed activities for which other team members were not qualified. At a daily team conference led by the team leader, patient care plans were reviewed. Ideally, the same team was assigned to care for the same group of patients each day. In practice, however, teams might include only a single RN. While team nursing was designed in part to make the most efficient use of RNs, it was criticized both for being overly task oriented and for resulting in fragmentation of care [Mark, 2002].9
Partly in response to this fragmentation of care, primary nursing became popular in the 1970s. This model of care delivery is characterized by the establishment of a direct relationship between an RN and a patient [Pontin, 1999]. The patient's primary nurse is responsible for all aspects of the patient's care, 24 hours a day, during the entire course of the hospitalization. This is achieved through a 24-hour plan of care created and implemented by the primary nurse, along with the use of associate nurses who care for the patient according to the plan in the absence of the primary nurse. Although primary nursing was not intended to require an all-RN staff, it was often interpreted in this way. The approach was viewed favorably by nursing staff because it emphasized the nurse–patient relationship and was perceived as most consistent with the practice of professional nursing [Norrish and Rundall, 2001].
Primary nursing still is often cited as the best way of organizing nursing care, although research on the effects of primary nursing has been hindered by the lack of a clear conceptual model [Pontin, 1999], and studies to date comparing team and primary nursing have had significant methodological weaknesses and yielded only equivocal results [Mark, 2002]. Moreover, some now assert that the question of which model is best is moot. Because levels of nursing expertise, support personnel, patient acuity and needs, and resources vary across nursing units, it is likely that the best nursing model in one unit is not the best for another. For example, a nursing unit with a high proportion of novice nurses is more likely to require a care delivery model that affords higher levels of clinical nursing supervision, such as a modified team approach, than a unit whose staff is stable and possesses higher levels of expertise. According to this view, care delivery models tailored to each nursing unit's structures, processes, and resources are most desirable [Deutschendorf, 2003].
Changes in workload Nurses in hospitals also report increasing workload as a result of the above changes [Aiken et al., 2001b; Hurley, 2000], and some have linked this increased workload to diminished patient safety [Kimball and O'Neil, 2002; Service Employees International Union, 2001; Sochalski, 2001]. Nurses' workload is discussed most often in terms of the number of patients assigned to each nurse [see also the discussion of staffing levels in Chapter 5]. In numerous surveys, nurses report inadequate numbers of nursing staff to accomplish their work [Kaiser Family Foundation and Harvard School of Public Health, 1999] and provide high-quality care [Aiken et al., 2001b]. Although evidence indicates that nurses' perceptions of staffing adequacy can be influenced by structural characteristics of hospitals and units, such as the number of beds in a nursing unit and higher levels of patient technology [Mark et al., 2002], the hospital industry itself reports difficulties in securing the number of RNs it needs to care for patients [AHA Commission on Workforce for Hospitals and Health Systems, 2002; American Organization of Nurse Executives, 2000; JCAHO, 2002]. Emergency room diversions, closures of nursing units, cancellation of elective surgeries, and other restrictions on service delivery have been documented as resulting from insufficient nurse staffing [First Consulting Group, 2001; Kimball and O'Neil, 2002; The HSM Group, 2002].
Staffing levels have been shown to vary considerably by hospital [Unruh, 2002]. This variation is illustrated by data for 1998–2000 from the California Nursing Outcomes Coalition, which maintains a statewide database of nurse staffing levels from California hospitals. Although these data constitute a convenience sample of 52 California hospitals voluntarily contributing staffing data, the data are useful because they were collected at the level of the nursing unit [as opposed to the aggregate hospital level], because common data definitions and reporting were used, and because ongoing verification was performed to ensure the data's accuracy. Data reported on 330 critical care, step-down, and medical–surgical units in these hospitals across nine calendar quarters revealed that RNs provided 92 percent of the care in ICUs, 87 percent of the care in step-down units, and 57 percent of the care in medical–surgical units. The RN–patient ratios across these facilities were as follows:
ICUs—a range of 0.5–5.3 patients for each RN [average 1.6]
Step-down units—a range of 1.5–11.6 patients for each RN [average 4.2]
Medical–surgical units—a range of 2.7–13.8 patients for each RN [average 5.9]
These findings did not vary over the nine quarters or by the size of the hospital [Donaldson et al., 2001].
Data from a fiscal year 2002 national convenience sample survey of hospitals on staffing, scheduling, and workforce management of nursing department employees further document this variation in staffing levels. The 135 hospitals responding showed variation in nurse staffing levels even with the shift and type of patient care unit being held constant. Although the average RN-to-patient ratio in medical–surgical units on the day shift was 1:6, the range was from 1:3 to 1:12. Twenty-three percent of hospitals reported that nurses in their medical–surgical units on the day shift were each responsible for caring for between 7 and 12 patients. On the night shift, 7 patients on average were assigned to each nurse, but 34 percent of hospitals reported between 8 and 12 patients assigned to each nurse. For critical care units, the average number of patients assigned to each nurse was 2 for both the day and the night shifts, but 7.4 percent of hospitals reported having nurses care for 3 or 4 ICU patients during the day shift, and 11 percent reported nurses caring for 3 or 4 ICU patients during the night shift [Cavouras and Suby, 2003].
In addition to staffing levels, work environment factors that have been identified as affecting nurse workload include RN expertise, patient acuity, patient turnover, physician availability, work intensity, unit physical layout, degree of teamwork, and available support staff [Pinkerton and Rivers, 2001; Salyer, 1995; Seago, 2002]. Many of these factors also have been affected by hospital reengineering and redesign initiatives. Workload factors for which there is a strong evidence base with regard to their effects on patient safety, as well as strategies for modifying the work environment to address these factors, are examined in the succeeding chapters of this report.
Nursing Homes
As patients move more quickly through acute inpatient settings or undergo complex procedures in outpatient settings, their needs for long-term care follow-up escalate. Further, as older adults increasingly constitute a larger proportion of the U.S. population, there is a concomitant increased demand for services for older patients who have higher dependency needs. As a result, nursing homes [sometimes called long-term care or nursing facilities] and the populations they serve have changed significantly in recent years.
Like hospitals, nursing homes are seeing an increase in the dependency and acuity levels of their residents [as described in Chapter 1] and an expansion of the nursing facility workforce. In contrast to hospitals, however, there has been an increase in the number of nursing homes and nursing home beds. Between 1987 and 1996, the number of nursing home beds in the United States increased by 19 percent, from 1.48 to 1.76 million, reflecting in part a 20 percent increase in the number of nursing homes nationwide [from 14,050 to 16,480] [CMS, 2000, 2002]. During this period, the percentage of nursing home patients whose care was paid for by Medicare increased from 3 to 9 percent, and the proportion of nursing homes certified to receive Medicare reimbursement increased from 28 to 73 percent, indicating that the number of nursing facilities planning to take residents with more acute illness or more complex needs rose substantially. Concurrently, the number of nursing home residents over age 85 increased from 49 to 56 percent for women and from 29 to 33 percent for men [Rhoades and Krauss, 2001].
Caring for individuals in nursing homes also involves some other special safety considerations. For many nursing home residents, the nursing facility is the home where they live as well as where they receive services. Patient safety in these facilities therefore requires consideration of patients' long-term living environment, as well as their clinical care needs. Further, many long-term care clients have some degree of cognitive impairment. Data from the 1996 Nursing Home Component of the Medical Expenditure Panel Survey [MEPS] revealed that nearly three-quarters [70.8 percent] of nursing home residents had some form of memory loss. About the same proportion had problems with orientation, such as knowing where they were or the identity of staff members. Many residents [80.6 percent] had difficulties in making daily decisions, and almost one-third [30.2 percent] exhibited at least one form of inappropriate or dangerous behavior—including wandering off or resisting care [12.5 percent]. Overall, nearly half of all nursing home residents had some type of dementia. These conditions make residents less able to participate in increasing their own safety, and in fact can cause behaviors that create their own threats to safety. Forgetfulness and disorientation in particular can be dangerous problems, requiring 24-hour supervision to provide for an individual's safety and well-being [Krauss and Altman, 1998].
This increase in resident dependency and acuity has important implications for staffing, oversight, work complexity, workload, and the overall nature of the work in nursing facilities. For example, the staff time required to meet basic needs of residents [such as feeding, toileting, and ambulation] increases with the overall dependency levels of residents [CMS, 2002]. Further, since Medicare residents often have complex health conditions or are recovering from serious health events, a more sophisticated knowledge base is required to care for these residents, as are higher levels of vigilance and monitoring and of professional staffing. Consistent with this observation, a higher ratio of RN staff to residents in nursing homes has been demonstrated to reduce adverse health care events for residents [CMS, 2000, 2002].
Along with the growth in the number of nursing homes, nursing home beds, and patient acuity has come a significant expansion of the nursing home workforce. There has been not only an overall increase in the total number of workers to care for the increased population of residents, but also an increase in the ratio of all categories of nursing staff to residents [CMS, 2000]. This increase is attributable in large part to the passage of the Nursing Home Reform Act in 1987, which mandated coverage by at least one RN for 8 hours a day, 7 days a week for all nursing homes accepting Medicare or Medicaid reimbursement. Additional legislation required facilities to have a licensed nurse [RN or LPN/LVN] on duty at all times. As discussed in Chapter 5, however, evidence indicates that levels of staffing above these minimums are necessary to ensure an adequate level of patient safety.
NAs make up 60–70 percent of the total nursing staff in nursing facilities. They spend the most time with residents and provide 80–90 percent of direct patient care, working under the supervision of RNs and LPNs/LVNs [CMS, 2000; GAO, 2001a]. LPNs/LVNs constitute approximately 25–30 percent of all nursing staff and approximately two-thirds of licensed nursing staff. RNs represent the smallest proportion—10–15 percent—of nursing staff in nursing facilities [CMS, 2000]. In contrast to the hospital setting, physicians are less frequently on site in nursing homes.
There has also been an increase in the percentage of nursing homes that are large for-profit chains or networks of not-for-profit facilities, as opposed to being individually owned. The significance of this shift is unclear. However, a 1998 national study of 13,693 nursing homes10 comparing those owned by investors with nonprofit and public nursing facilities found higher rates of deficiencies in the quality of care provided and lower staffing levels among the former. Chain ownership also was found to be associated with higher rates of deficiencies in quality of care. In both instances, the analysis adjusted for case mix, location, percentage of patients covered by Medicaid, whether the facility was hospital based, and whether it served only Medicare residents [Harrington et al., 2001, 2002].
Home Care and Community-Based Organizations
Home care and community-based organizations encompass a wide variety of noninstitutional long-term care settings, ranging from an individual's own home to various types of congregate living arrangements. The boundaries between institutional and noninstitutional care settings are blurring, however. Many assisted-living “board and care” facilities are large buildings that resemble nursing facilities. Other residential care sites are small and homey. In contrast to nursing homes, which are licensed and regulated by the federal government as a condition of Medicare and Medicaid reimbursement, residential care facilities are generally licensed and regulated by states and local jurisdictions. Consequently, there is no single definition of “residential care” or tally of the number of such facilities nationwide. A 1999 national study counted 11,472 assisted-living facilities with approximately 650,500 beds. Other community-based long term-care settings include adult day care programs, in which disabled elderly individuals receive supervision, personal care, and social integration in a group setting, usually during the work week and normal work hours [Stone and Wiener, 2001].
Home health care was the fastest-growing employment setting for all nursing personnel throughout the 1980s and 1990s [Buerhaus and Staiger, 1999]. As of 2001, there were more than 20,000 home care agencies, approximately 7,000 of which were Medicare-certified. Free-standing, for-profit agencies represented 40 percent of that total and experienced the greatest growth. Hospital-based agencies made up another 30 percent of the total.
These free-standing and facility-based [usually hospital-based] Medicare-certified agencies, home care aide organizations, and hospices employ licensed nursing staff [as well as physical, occupational, and speech therapists] to provide such skilled services as illness management, medication management, infusion therapy, wound care, ostomy instruction, and end-of-life care to clients in their homes and other locations. Licensed home care nurses also supervise home care aides who provide such personal care services as assistance with bathing, eating, and ambulation, as well as monitoring of vital signs and patient status. NAs make up 54 percent of the nursing personnel working in home health care [GAO, 2001b].
The home care industry has experienced substantial turbulence. Since the 1960s, the National Association for Home Care [the home care industry association] has documented periods of rapid expansion and decline in the numbers of home care agencies [National Association for Home Care, 2001]. In particular, the Balanced Budget Act of 1997 changed the way the Medicare program pays Medicare-certified home health agencies from a cost-based method to a prospective payment system of fixed, predetermined rates. Subsequently, the number of Medicare-certified home health agencies decreased by 32 percent—from 10,556 in 1997 to 7,715 in 2000 [Office of Inspector General, 2001]. As with nursing home care, however, demands for home health services are expected to continue to grow because of reduced lengths of stay in acute care hospitals, advances in technology, and the aging of the U.S. population.
Public Health Agencies
Public health agencies comprise state, county, and local health departments that provide such health care services as immunizations, health education, case management for frail elders, and community assessment. All states have a public health structure and staff at the state level; some also have such a structure and staff in all counties or regions. Although many cities still have local health departments, the trend is toward decreasing duplication and cost by merging city and county units [Martin, 2002]. RNs employed in public health and community health settings increased by 155 percent between 1980 and 2000 [Spratley et al., 2000].
During the 1990s, various factors, such as substance abuse and its impact on high-risk pregnancies and newborns and the incidence of HIV/AIDS, stimulated growth in the public health sector and caused these agencies to reassess their mission and purpose. An earlier IOM study found that the public health system was in disarray and incapable of fulfilling the fundamental core functions of assessment, policy development, and assurance [IOM, 1988]. Following the events of September 11, 2001, and associated concerns about bioterrorism, the public health infrastructure began receiving additional attention.