The estimated prevalence of psychiatric disorders in nursing home residents ranges from 68 percent to 94 percent (1,2). The breadth of this range reflects the variation in definitions of psychiatric disorders as well as the number and types of nursing homes reviewed. Psychiatric disorders cause symptomatic distress to residents, complicate the course of medical illnesses, interfere with and confound nursing and medical care, increase residents' disability, raise health care costs, and diminish residents' quality of life (3). Pharmacologic agents and nonpharmacologic interventions are available to treat these disorders.
The Nursing Home Reform Act, part of the Omnibus Budget Reconciliation Act of 1987 (OBRA), restricted the use of antipsychotic drugs, antianxiety agents, sedative-hypnotics, and related drugs. The regulations require that clinicians try nonpharmacologic interventions first except when there are immediate concerns about residents' safety. However, cost and time limitations have led to a tendency to treat psychiatric and behavioral disorders with pharmacologic agents.
The most common cause of psychiatric and behavioral problems among nursing home residents is dementia (2). Three main psychosocial models describe the problem behaviors associated with dementia: the "unmet needs" model, a behavioral and learning model, and an environmental vulnerability model (4). Each model describes problem behaviors as complex phenomena that require individualized strategies based on a resident's unique characteristics. In a review of the literature on nonpharmacologic interventions, Cohen-Mansfield (4) identified 83 studies and categorized the interventions as social contact, behavior therapy, staff training, structured activities, environmental interventions, and a combination of therapies. Given the modest efficacy of pharmacologic interventions (5), treatment strategies should include nonpharmacologic interventions. However, without evaluation and mitigation of the challenges of delivering such interventions, nursing homes will continue to avoid this approach.
In this article we use an organizational framework drawn from the work of Porras (6) to discuss internal environmental challenges to the use of these interventions. We also discuss external challenges to the use of nonpharmacologic interventions posed by the regulatory, legal, and economic environments. Finally, we make general recommendations for overcoming these challenges.
Porras (6) describes the use of the stream organization model to guide planned organizational change. In this model, four interconnected dimensions describe the organization's work environment that shapes and guides the behavior of people on the job: organizing arrangements, including goals and strategies, formal structure, policies and procedures, and formal reward systems; social factors, including culture, interpersonal interactions, communication, influence and status, and individual attitudes and beliefs; technology, which encompasses technical expertise, job design, work flow, and information technology; and physical setting, including space configuration, physical ambience, and interior design. The design of one dimension affects the functioning of the others. In addition, a change in the factors of one dimension has an impact on the other dimensions.
The behavior of personnel at all levels is an important determinant in organizational performance and in the development of individual staff members (6). When organizational performance needs to improve, personnel must change their behavior. The following discussion explains how the organizational framework can address barriers to the use of nonpharmacologic interventions in a typical nursing home.
Unlike hospitals, nursing homes do not have standardized roles for and lines of communication between the residents' physicians, the medical director, and the nursing staff. No one, not even the medical director, monitors the practice of the residents' physicians. Often, numerous physicians see residents in a nursing home, and thus the nurses must communicate with a large number of physicians.
The medical director and the residents' physicians spend limited time at the nursing home (7). Consequently, there may be inadequate communication between them and the nursing home staff about the residents' needs, particularly those having to do with mental health and quality of life. Primary care physicians with large patient loads and limited psychiatric training provide care for most nursing home residents. They often base therapeutic decisions on staff reports or chart reviews and have little time to conduct interviews, establish relationships, or give adequate consideration to nonpharmacologic interventions. A survey of nursing homes in New York State found that the professional nursing staff made most of their initial reports about problems with residents by telephoning physicians (8). Thus the professional nursing staff must have skills in interpreting behavioral episodes to correctly describe them and help physicians distinguish behaviors that are best treated nonpharmacologically.
Usually unlicensed frontline staff members observe behaviors firsthand. Because their jobs are inadequately supported by training (9,10), it is difficult for these staff members to accurately interpret behaviors. Unlicensed staff members receive low wages and few rewards for providing quality care (9,10), and they rarely have a chance to make decisions (11).
Observational studies of nursing assistants indicate that they fulfill primarily a custodial function and have limited interactions with residents (12). Nursing assistants often have negative feelings when residents show aggressive behavior (13). They tend to avoid residents who are verbally or physically aggressive (14).
The cultural and ethnic background of the nursing home staff may influence their perceptions of problem behaviors and nonpharmacologic interventions. Everitt and associates (15) found a wide variance in the ability of individual caregivers to tolerate aggression. Beck (16) reported that nursing assistants in the Baltimore and Washington, D.C., areas recorded significantly fewer disruptive behaviors than nursing assistants in Arkansas. Staff members may hesitate to replace medications with nonpharmacologic interventions if they anticipate that this practice may lead to more disruptive behavior and add to their workload (17). They often see the sedating effects of psychotropic drugs as desirable and might prefer their use to resident activity.
Anecdotal evidence suggests that family members may hesitate to acknowledge a resident's behavioral problems. They may fear that the resident will have to leave the nursing home if the behavior persists (18), which would be particularly difficult for families who have few other options for care of the resident. In other instances, family members may want a resident to be medicated or restrained for his or her own protection.
A nursing home's lack of access to mental health specialists often prevents a thorough evaluation of behavioral symptoms (19). In a six-state survey of directors of nursing, 38 percent responded that they needed access to psychiatric consultants. Forty-eight percent, particularly those in rural areas and smaller facilities, felt that on-site consultation was inadequate. Few psychiatrists, particularly in rural areas, have training in geriatrics (20). Even if mental health consultation is available, no clear system exists for providing information from the consultant to the nursing staff. A lag occurs between the actual consultation and the time when the primary physician sees the consultant's report. Face-to-face professional communications are rare because of differences in schedules.
Nursing homes usually do not have a system for tracking residents' behavior and use of nonpharmacologic interventions. The nursing staff's inadequate assessment and insufficient communication about changes in residents' status compromise clinical decision making by physicians (21). Physicians depend on staff reports because their office charts contain only a minimal amount of information, and nursing homes rarely have computerized clinical information systems (22).
Nursing assistants receive inadequate training in nonpharmacologic approaches. When such training does occur, its effects quickly dissipate, because no system is in place to support the staff by providing feedback and reinforcement (23,24). Nursing assistant jobs are designed to get tasks done, so it is unlikely that nursing assistants would use time-intensive nonpharmacologic interventions. Because nursing assistants are usually excluded from resident care planning, professional staff may remain unaware of whether or when they use nonpharmacologic approaches.
The physical layout of many nursing homes is not conducive to observing patients' behavior and outcomes of nonpharmacologic interventions. Space configurations tend to aggravate problem behaviors, resulting in more use of sedation. Often, no one even considers modifying the multiple competing stimuli that may trigger behaviors before they resort to use of medications.
Nursing home residents may personalize their rooms, but common areas often have a cold institutional appearance and discourage interactions between people. In most nursing homes, residents share rooms, and this may lead to conflicts. Noises such as doorbells, alarms, public-address systems, telephones, and screaming are especially intrusive. Long halls and loud, crowded common areas may trigger wandering by residents (25).
Models for increasing the use of nonpharmacologic approaches need to acknowledge the regulatory, legal, and economic forces with which nursing homes must contend.
Although OBRA guidelines have been influential in decreasing the use of psychotropic medications (26,27), their influence on the use of nonpharmacologic interventions has not been demonstrated. The current regulatory climate emphasizes punishment rather than support. Regulations inadvertently promote pharmacologic interventions because of the lack of tolerance for safety violations, such as aggressive behavior, and a great concern to protect the rights of other residents from, for example, patients who wander. Surveyors may judge multiple incidents as a pattern that nursing home staff must curtail. Thus staff members may be reluctant to try nonpharmacologic interventions first and use medications instead.
The use of psychotropic drugs is a key quality indicator and potential flag for surveyors because of the belief that many residents are overmedicated. For each of the antipsychotics, antianxiety agents, sedative-hypnotics, and related drugs, the OBRA guidelines provide a list of acceptable indications, prohibited agents, maximum dosages, requirements for monitoring treatment and adverse effects, and time frames for dose reduction and cessation (19). The guidelines instruct surveyors that before they find a nursing home out of compliance, they allow the facility to explain why a drug that deviates from the guidelines was used and why it served the best interest of the resident. Federal regulations do not strictly constrain prescribing options as long as physicians document their reasoning for determining that the benefits to the resident outweigh the risks of treatment (19). However, the regulations provide few instructions on the kinds of nonpharmacologic interventions that nursing staff should try and the methods they should use to implement and document such interventions.
Advocates for nursing home residents view lawsuits as a way to catch public attention when regulations are violated (28). Multimillion dollar settlements are becoming more common. Consequently, the cost of liability insurance for nursing homes has reached levels of $200 to $600 per bed (28). Nursing home operators who cannot pay the premiums go out of business. Thus, to those in the industry, the legal environment is often seen as punitive rather than as conducive to change. Although how much of this litigation is related to medication use or nonpharmacologic control is unknown, behavioral problems can be linked to concerns about abuse and neglect if they are handled inappropriately.
Would-be providers of specialty mental health services to nursing home residents face economic barriers, including fragmented payment sources, inconsistent application of coverage policy and medical necessity determinations, high copayment requirements, and a lack of reimbursement for midlevel providers who specialize in mental health (3,29). Mental health professionals generally receive reimbursement by fee-for-service billing under Medicare Part B or salary support from nursing homes or some combination of the two (30). Yet Medicare covers only 50 percent of approved psychiatric and psychological services, compared with 80 percent of approved medical services.
In addition, payments vary by region, because fiscal intermediaries have different interpretations of the regulations (31). In many states, Medicaid does not cover the 50 percent copayment, and families may refuse to pay the 50 percent out-of-pocket fee if they did not initiate the service (32). Residents or families also may refuse the use of antipsychotic drugs or nonpharmacologic interventions because they do not see the value of psychiatric consultation. Thus access to and promotion of mental health services in nursing homes have been limited.
State surveyors report that staffing shortages and inadequate staff expertise are major factors in poor-quality care at nursing homes (33). Multiple sources indirectly suggest that staffing of nursing assistants is inadequate (34,35,36). The American Health Care Association (37) reported turnover rates of 97 percent for nursing assistants, 52.5 percent for registered nurses, and 68.8 percent for total staff in 1996 because of factors such as poor wages and limited or no health benefits (10). The current shortage of nursing personnel makes replacement of nursing staff difficult. Thus although nursing home administrators support an increase in minimum staffing standards, they are also concerned about its feasibility. Some states penalize nursing homes financially if staffing drops below a minimum level. This practice has added to concerns about mandated minimum levels.
The appropriate use of nonpharmacologic approaches requires that the leadership of the organization emphasize integrated biopsychosocial care, interprofessional communication, and teamwork. Support for this shift in emphasis comes from recent research that examined variations in treatment culture and drug use in facilities in Wisconsin before and after OBRA (38). A reduction in use of antipsychotic drugs was more likely to occur in homes with a "resident-centered culture" characterized by stronger nurse beliefs favoring psychosocial interventions, more regular review of drug use and feedback from pharmacists to staff nurses, and involvement of mental health workers.
To make an appropriate treatment decision, the primary physician requires documentation of what preceded problem behaviors, the specific behaviors demonstrated, the frequency and duration of the behaviors, what others did in response to the behaviors, and how the resident reacted. Medication will likely have little effect on problem behaviors that were triggered by inappropriate interactions with staff members. Gathering data from as many sources as possible—for example, from different care providers, including family, over a 24-hour period—is helpful in this process. Placing a flag on the front of the charts of residents who are receiving nonpharmacologic interventions and attaching a form inside the chart to monitor the specific target behaviors may facilitate communication, informed decision making, and patient care and indicate the importance of the interventions.
Allowing nursing assistants to participate in residents' care planning would give them a better understanding of nonpharmacologic treatment approaches and a chance to provide valuable insight into the triggers of problem behaviors. Systematic supervisory performance monitoring and feedback (12,24), permanent work assignments (39,40), and a team approach (39,41) would also lead to greater use of nonpharmacologic interventions.
Because of the communal nature of nursing homes, the education of caregivers, visitors, and residents is a vital part of an effective plan to address behavioral disturbances. Everyone needs to know about the nonpharmacologic interventions both to understand the staff's actions and to support the plan of care in a way that respects the privacy and confidentiality of residents. Communication of the plan to handle the problem behaviors will reassure family members and other visitors and may prevent their inadvertently provoking disturbances. Educational programs for families need to address psychiatric illness, dementia, regulatory issues, participation in care planning, and communication with nursing staff. Family support groups need to promote family involvement, collaborative problem solving between the resident and staff, and consumer awareness of effective advocacy (42).
Environmental adaptations can compensate for a resident's particular cognitive deficit. For example, many residents respond to environmental cues, such as walking through doors that are left open, and physical cues, such as brushing their teeth if handed a toothbrush.
Reducing clutter and limiting the number of choices are other possible interventions. Staff members need to learn that residents may engage in problem behaviors to express discomfort from excessive heat or cold; unusual, loud, or repetitive noise; unusually bright or dim light; or altered routines. If staff members change these stimuli, residents may stop the problem behavior. The availability of physical amenities such as private lounges increases family involvement (43), and the provision of adequate areas for staff members to take breaks provides an opportunity for them to recover from the stresses of caregiving.
As the survey procedure changes from being process based to outcomes based, nursing staff should be able to incorporate creative solutions to behavior problems into care plans. Care plans need to allow for individual differences and weigh concerns about individual rights against concerns about the safety of others. The overall training for state surveyors needs to include guidelines for providing and evaluating the care of persons with dementia, as the state of Illinois has done (personal communication, Baker K, 2000). State surveyors need to be cognizant of the range of nonpharmacologic interventions and reinforce their appropriate use. Currently 26 states have adopted legislation to guide the development and practice of special care units (44,45), and many supported staff training in nonpharmacologic approaches to problem behaviors.
Nursing homes need a mechanism that helps administrators protect themselves against false accusations and minimize liability in unpreventable situations. Hospitals commonly use risk management, which has four purposes: to reduce preventable injuries and accidents, to minimize the financial severity of claims, to provide quality care, and to respond positively to unexpected events to prevent their recurrence (46). The adoption of risk management in nursing homes would promote the use of nonpharmacologic interventions.
The payment mechanisms and incentives for nonphysician mental health providers to work in nursing homes need improvement. Nonphysician mental health providers are more likely to use nonpharmacologic interventions. Regulations and payments for staffing need to support the use of more professional staff and more training for nursing assistants.
The most successful changes in nursing home care have involved reducing the frequency of certain poor care practices rather than adopting new practices (47). It is much more difficult to implement new care practices (22) than to prescribe medications. Initiatives to foster the use of nonpharmacologic interventions must be broad-based and multifactorial. There must be support, rather than contradictory messages, from the four organizational dimensions of the nursing home.
There also must be support from the regulatory, legal, and economic sectors of the health care system. Efforts to promote the use of nonpharmacologic approaches to problem behaviors must begin by encouraging outcomes that are meaningful to all—the nursing home operators, staff, families, residents, and policy makers.
Dr. Cody and Dr. Beck are affiliated with the department of psychiatry and behavioral sciences at the University of Arkansas for Medical Sciences in Little Rock and with the Center for Mental Health and Outcomes Research at the Veterans Affairs Medical Center in North Little Rock. Dr. Svarstad is with the School of Pharmacy at the University of Wisconsin in Madison. Send correspondence to Dr. Cody at the Veterans Affairs Medical Center, HSR&D, 2200 Fort Roots Drive, Building 58 (152/NLR), North Little Rock, Arkansas 72114 (e-mail, firstname.lastname@example.org). This article is part of a special section on mental health services in long-term-care facilities.